Fan Buffalo

o
fafi
03
O
CD
o
Q-
Copyright, 1914, by
The Buffalo Forge Company
Buffalo, N.Y.
llCt-CO^ tUFFAlO.
ENGINEERS HAND BOOK
O
OF TABLES, CHARTS AND DATA ^-^

ON THE APPLICATION OF CENTRI-
FUGAL FANS AND FAN SYSTEM o
APPARATUS, INCLUDING ENGINES
g
E
^
AND MOTORS, AIR WASHERS,
HOT BLAST HEATERS AND
SYSTEMS OF AIR DISTRIBUTION
PUBLISHED BY O
o
THE BUFFALO FORGE COMPANY -E
hp
BUFFALO, N. Y.
O
FIRST EDITION
O
-^
o

PRICE $3.00 _0
1o
a.
EDITED BY
WILLIS H. CARRIER, Chief Engineer

PREFACE
THE use of the fan in general engineering practice
increasing, making it imperative that both the engineer and
is rapidly-
the architect become familiar with the fundamental principles

governing the selection and application of fans for various
purposes. Some general information has been published on the
subject by the different fan builders, but what has been given
out has always been incomplete and frequently misleading, and
did not afford sufficient data to be intelligently employed by the
engineer. Heretofore no effort has been made to collect and
present under one cover the latest and most reliable engineering
data concerning fans and their application to various industrial
requirements.
This book is intended to be used as a guide in the selection

and application of fans, heaters and kindred apparatus, and an
effort has been made to so standardize the rules and data given
that they may be used with any standard make of equipment.
The greater part of the data presented is the result of tests and
research made by the engineering staff of the Buffalo Forge
Company in the testing laboratories of the company, many of
the investigations being made purposely to obtain data for this
book. The results of these investigations have in most cases
never been heretofore published except in the proceedings of
some of the engineering societies, where they were presented by
the engineers of this company, and others. The rules and ap-
plications as outlined are the same as are used in this company's
practice. In preparing this work the theory has been generally
omitted, except in such elementary form as was necessary to
an understanding of the facts given.
The information herein presented is in complete and reliable

form for standard applications, but there are many cases
requiring special consideration from the engineer familiar with
fan installations.
FAN ENGINEERING
OUTLINE OF CONTENTS
Part I PROPERTIES OF AIR.
Weight of air.
Specific heat of air.
Relation of velocity to pressure.
Effects of temperature, humidity
O
and barometric pressu^^
on the properties of air. ^
Relation of dry-bulb, wet-bulb and dew-point temperC
atures. O
Sensible, latent and total heat. C
Psychrometric charts and tables. 0
Part II APPLICATIONS. ^
Section I Heating.
Heat losses from buildings.
Cold storage insulation. OO
Heat loss from galvanized iron pipes. ^^
Heat loss due to infiltration.
Heat required for ventilation.
^
-^
Air quantity and temperature required for heating, q^
Section II Ventilation. f
Standard of ventilation. i^kfl
Special air requirements. C2
Application of the fan system of ventilating. LlJ
Section III Air Washing, Cooling, Humidifjring, Drying.'Z^
Air washing,
Humidity in heated buildings.
^
^"^
Cooling. O
Relation of room temperature to outside wet-bulb tem- Q^
perature. q^
Drying. O
Time required drying various materials.
for ^^
Practical examples in drying calculations. C^
Moisture removing capacity of air. O^
Section IV Mechanical Draft.
Forced and induced draft.
Draft requirement.

FAN ENGINEERING BUFFALO FORGE COMPANY
Amount of air required.
Measurement of air used.
Mechanical draft in connection with mechanical stokers.
Section V Exhaust Systems.

Size of pipes and velocity required.
Dust removal from grinding and buffing wheels.
Specifications from Labor Laws of New York State.
Dust collectors.
Friction loss in system.
Standard and slow speed exhausters.
Practical examples.
Section VI Miscellaneous Applications.

Part IIIAIR DUCTS.
Pressure losses in air ducts and elbows.
Pressure losses in nozzles and orifices.
The diverging nozzle in air ducts.
Theoretical outlet or blast area.
Proportioning the various losses.
Proportioning piping for exhaust systems.
Proportioning ducts for public buildings.
Proportioning piping for industrial buildings.
Equalizing friction for unequal length.
Piping layout.
Sizes and weight of piping.
Part IVAPPARATUS.
Section I Fans.
Fan design.
Pressure characteristics.
Special types and features.
Horsepower of a fan.
Relations of total, static, and velocity pressure.

The selection of a fan practical examples.
Section II Fan Testing.

The pitot tube.
The anemometer.
The orifice.
OUTLINE OF CONTENTS
The converging nozzle.
Coefficients of discharge for air measurements.
Fan Capacities.
Section III
Section IV Fan Dimensions.
Section V Heaters.
Buffalo standard pipe coil heater.
Heat transmission through metal surfaces.
Temperatures attained with indirect heaters.
Condensation in coils.
Velocity of air through heaters.
Application of heater tables and curves.
Frictional resistance of heaters.
Sizesand dimensions of Buffalo heaters.
Heater cases and by-pass proportions.
Determination of guarantees.
Section VI Air Conditioning Apparatus.

Air washers.
Types of humidifiers.
The dehumidifier.
The design of humidifiers.
Refrigeration required for dehumidifying.
Power required for operating humidifiers,
dimensions and capacity tables of Carrier air washers and
humidifiers.
Section VII Steam Engines.

M. E. P. of high speed engines.
Water rate of high speed engines.
Brake H. P. per R. P. M. of Buffalo engines.
Dimensions of Buffalo engines.
Section VIII Practical Applications and the Selection of

Apparatus for Heating and Ventilating.
Part V APPENDIX.
Specifications.
Miscellaneous engineering data.
Index.
,
PART I
PROPERTIES OF AIR
In this section will be found a discussion of the physical and> -^
chemical properties of air and their general relations with re-
spect to "fan engineering." A complete set of psychrometric "îT
charts and tables are included. q^
Air a mechanical mixture of various gases, ordinarily con- CZ
is
sidered as consisting of oxygen and nitrogen, but also containing <^

a portion of moisture and carbonic acid, and a very small part"
of other constituents. The proportion of these components will -
vary under different conditions, but ordinarily pure dry air is ;
^ j
composed as follows, in per cent.:
By Volume By Weight oio
Oxygen 20.9 23.1 . -.
Nitrogen
The moisture will
79.1
vary with the humidity of the air, from O*-.
76.9
^
to 4 per cent., and the carbonic acid will vary with the purity q^
of the air from perhaps 0.03 to 0.30 per cent., or as usually ex- O
C
pressed, from 3 to 30 parts in 10,000.
Weight of Air '^
The weight of the air varies with its temperature and baro- C
metric pressure and also with the amount of moisture it contains. LjlJ
The weight of one cubic foot of pure dry air expressed in pounds ,..
may be determined by the formula
^^^982p
459.2 +t
where p = absolute pressure in pounds per square inch.
^
^
^
a>
t=- temperature of the air in degrees F.
A convenient formula for expressing the weight of dry air CD
at any conditions of temperature and pressure as used by ^
Frank H. Kneeland* is
0253b
^
+t
459.2
(2)
^
c
*^
where b = corrected barometer reading in inches oi mercury
t = temperature, deg. F.
1.3253 = weight in lbs. of 459.2 cu. ft. of air at 0 F. and
1" barometric pressure.
* "Some Experiences with the Pitot Tube on High and Low Velocities"
Am. Soc. Mech. Engrs., Dec. 1911.
1

A formula expressing the weight of humid air is given in the
Smithsonian Meteorological Tables as
0.080723 b- 0.378 e
1 +0.0020389 (t- 32) ^ 29.921
^"^^
where t = temperature, deg. Fahr.

b= height barometer in inches of mercury
of
e = pressure due
to vapor in the air in inches of mercury.
According to the latest data the above values should be
slightly changed, and we will then have the following formulae
as convenient forms for calculating the weight per cubic foot of
either dry or moist air.
r?
For A
dry air w
^^ =
0.0028862 b
^^^
1 + 0.0021758
.^
For moist
.
air W = 0.0028862
, b- 0.001088 e
^'^
iô.0021758t
This last gives the weight of a cubic foot of the mixture of
air and vapor, either for saturated or partly saturated air.
The weight of the dry air contained in one cubic foot of
saturated air may be determined from the formula
^y _ 0.0028862 b- 0.002886 e
1 +0.0021758 t
^^
The weight vapor or density in pounds per cubic foot of
of
saturated vapor at temperature t is given by the following:
_ S (144X0.4908 e) ^^
'
53.35 (459.2 + t)
where S is the specific weight of water vapor and may be found as
S =0.6221 +0.001815l/e"+0.000005lT/e3 (8)
The relationship temperature and specific

between the
weight of vapor is shown by the diagram on page 14 taken from
W. H. Carrier's paper on "Rational Psychrometric Formulae."*
An approximate value for the weight of water vapor con-
tained in one pound of dry air saturated with moisture may be
determined from
G^^^
b-e (9)
It may be noted from the curve on page 14 that the value of

0.624 for S in the above is only correct at about 70.
"Rational Psychrometric Formulae " Am. Soc. Mech. Engra., Dec, 1911.
12
PROPERTIES OF AIR
PROPERTIES OF DRY AIR

Barometric Pressure 29.921 Inches
0.64 60
0.6440
0.642C c^ 53 35(t'4596)Ds
I44p
fi.f>400
. Ds Density in lb per cu. ft
t Temperature. Fahr
O.iSSO
p Pressure in lb persq.m
0.6360
Equa tion of curve
5 --0 6Z2l'>0O0l6l5'ie*0Q00OOSlS?
^0.6300
0.6280
0.6260
0.6Z40
20 40 60 60 100 120 140 l60 180 200 220 240 260 280
Temperature
Specific Weight of Water Vapor
The table on page 13 gives the properties of dry air for

various temperatures, and the table on page 15 the properties
of saturated air. These are both based on the standard baro-
metric pressure of 29.921 inches.
The table on page 17 giving the weights of saturated and
partly saturated air for various barometric and hygrometric
conditions will be found especially convenient in making calcu-
lations based on other than standard conditions. The weight
in pounds per cubic foot of saturated air is given for even baro-
metric pressures and temperatures. The decrement per degree
rise in temperature and the increment per 0.1" increase in baro-
meter are also given, thus readily giving the weight of saturated
air at any other temperature and pressure. The last column in
the table gives the approximate average increment per degree
wet-bulb depression which is to be added to the weight of satu-
rated air to obtain the corresponding weight of partly saturated
air.
Example As an example of the use

. on page 17 we
of the table
will assume a case where it is desired to weight in pounds
find the
per cubic foot of air at a temperature of 83 dry- and 68
14
PROPERTIES OF AIR
PROPERTIES OF SATURATED AIR

Weights of Air, Vapor of Water,and Saturated Mixture of Air and Vapor at
Different Temperatures, Under Standard Atmospheric Pressure
of 29.921 Inches of Mercury
wet-bulb (or a depression of 15) when the barometric pressure is
29.40 inches. From the table on page 17 we find that the
weight of saturated air at 80 and 29.00 inch barometer is
0.07034 lb. per cu. ft. Also the decrement to be subtracted
is 0.00015 lb. per degree of temperature above 80. That is,
the weight at 83 and 29.00 inches would be 0.07034- (3 X 0.00015)
= 0.06989 lb. per cu. ft. The increment to be added per 0.1"
increase in barometer above 29.00 inches is 0.00025, so that the
weight of the saturated mixture at 83 and 29.40 inches will be
0.06989 + (4X0.00025) =0.07089 lb. per cu. ft. From the last
column in the table we find the increase in weight for each degree
wet-bulb depression for a temperature of 83 to be 0.000034-1-
0.3 (0.000039 - 0.000034) = 0.0000355.
Then the weight of moist air at 83, 15 wet-bulb depression,
and 29.40 inch barometer will be
0.07034 -0.00045 -hO.OOl-H 0.00053 =0.07142 lb. per cu. ft.
Specific Heat of Air
The heat of air is the ratio of the heat required to
specific
raise the temperature of a given weight of air through one degree
as compared to the heat required to raise the temperature of the
same weight of water from 62 to 63 degrees Fahr., i. e., it is the
B. t. u. required to raise one pound of air one degree Fahr.
The specific heat of air may be expressed as either of two
factors, specific heat at constant pressure or at constant volume.
It is the specific heat at constant pressure that is ordinarily
referred to. The factors commonly used heretofore have been

those determined by Renault specific heat at constant pres-
sure =0.2375, and at constant volume = 0.1689. But recent
investigation tends to show that the value 0.2375 is too low, and
that it should be Cp = 0.24112 4-0.000009 t or for ordinary pur-
poses approximately 0.2415.*
For the specific heats of various substances see the table on
page 78.
Relation of Velocity to Pressure
The laws governing the flow of air are perhaps less under-
stood than almost any other branch of engineering data. The
flow of air under high pressures must necessarily be investigated
thermodynamically and the formulae are more or less compli-
* "Rational Psychrometric Formulae" by Willis H. Carrier. Am. Soc.
Mech. Engrs., December. 1911, also W. F. G. Swan, Phil. Trans. Royal Soc,
Series A. Vol. 210, pp. 199-238.
16
I
PROPERTIES OF AIR
uoisssjdsa qing
)i{8p/W ui 3se3J3ui
8JJ3AV "xoiddv
05 00 00 C<3CC(N
4Ba UI
3S!^//rOJ3d ooo ooo ooo OOO OOO ooo (N(NW
OOO
ooo ooo ooo
ooo ooo ooo ooo
ooo ooo coo ooo
ooo ooo OOO
qjng Xjq OSOOOO t^t^CO COCDCO tÔOOi OfN-* 2 00.
'3UI '83Q Jdd ooo ooo ooo

OOO ooo ooo ooo COCOrf
OOO <N<Neo ooo
OOO
ooo ooo ooo ooo ooo ooo
OOO ooo
coo OOO ooo
>!V
OINCO 0500
OTftOO .-HCOiO OTflO
pajBjniBS 'Jd .-iair>. .-lOJt^ O05 O5COC0 OCOCM
OOtN-t^ CD CD
no J3d -JAV ooo OOO ooo CDooo OOO OOO
jBg UI 05 00 00 t^t^CO ^ T}< CO CO00<N r-Ô
<NCN(N (N<N(N
9si^//roJ3d ooo ooo
ooo coocoo ooo ooo
coo cooooo coo
OOO ooo
coo ooo ooo
ooo ooo coo ooo
ooo
qing Xjq 000000 t^<OCO OC00 CDt>-M 050<N
(N(NCO
DUf *33(] J3d
coo ooo coo
OOO
ooo coo
ooo ooo coo coo
ooo coo
ooc ooo ceo
ooo ooo
ooo
ooo
osioea COCCTf< ooo CD OOt^N
CO ooo 10 05C0 r>-C(N
p3)cjn)BS JJ eO'-o oocoio ccô OOtîO OOTfO
00 00 00 CD CD CO
no J3d lAv coo coo coo ooo
jBg UI
qing ^jq
3UI '330 J3d
p3BinBS -Id
t^cooo
txDn
005t^
ccma.
OS-* 00
OOSt^
^ô ^CO^
OiiOM
(NC0 05
<t(N05
cO'i'C^ .--05l^ CO<N00
O0J>t> t^COCO CO CO CO COOiO
no J3d -JAV ooo OOO ooo ooo ooc
JBg UI 05 00 00 r^t^co CDOiO TfTj^CO MCC(N
ooo ooo ooo ooc <N(N(N
ooo OOO oco
ooc oocooc cor
3S!^//rO-3d
ooo
ooo coo coo ooo
ooo ooc coo ooc oo<
COCOCO CCOÔ iooo CCCCt>- 00 os-^ ceo 00 ^CDO
qing Xjq
DUI aSQ J3d ooo ooo ooo
coo coo coo OOO'
oco ooo
oco ooo
oco OCO'
JAV s.J3aa ooo ooo ooc oco OOO
OOOCO 00 ooo ec-ô rfOSO
!V OOCîO OS'* 05 TTOJ-^ OOIMCO 00 05 00 lOOW OTjtiO
TfOCO'
pajBjniBS -jj t^cOrf IN^05 00 --om occO'* <NCt^
COCOCO COCOCO lOiCiO
no J3d -JAV OOO ooo oco coo oco ooo OOOI
t^t^co
JBg UI
3S!^//rO-ad
AV s.-'ûi
COCOCO UJiCO lOiOiO
qing /(jq
ooo coo
ooo oco
'3UI'S9Q J3d ooo
IA\ s.jDaa ooo
!V
pajBjn^BS Jd
no J3d lAV
jqsa S33J83Q sjni
-Bjadiosx qfng Ajq r
17
cated. For ordinary fan work, however, where air is at low
pressure but slight error isthe same formulae are
introduced if
applied to the flow of air as are commonly used for th^flow of

water. The basic formula for such calculations is
V = ^2gh (10)
where Vb = velocity in ft. per second, or
V = 60^^|2^ (11)
where V = velocity in ft. per min.

g= acceleration due to gravity in feet per second
h=head in ft. causing flow
But we also have
h= h'^ (12)
where h'= head expressed in in. of water

d = density of water
W= weight of air in lbs. per cu. ft.
Then at 70 F. and 29.92'' barometer and with dry air
12W ^^''^
12X0.07495
and we have
V = 60^2gh'-^=4005^^ (13)
Thus we see that the velocity at standard conditions stated for

a pressure of one ipch of water will be 4005 ft. per min., and for
one ounce per square inch will be
4005 a/i.734 = 5273 ft. per min. (14)
The weight of dry or saturated air at other temperatures

may be found from the tables on pages 13 and 15, or for any
special condition of temperature, barometer, or humidity from
the table on page 17, the use of which has already been explained,
(see page 14).
The most convenient formulae for determining the velocity
or pressure of air under different conditions of temperature,
barometer and humidity, when computing test results are the
following:
V = 1096.5^-^ (15)
18
PROPERTIES OF AIR
where V= velocity in ft. per min.

p = pressure in in. of water.
W = weight of air in lbs. per cu. ft.
The quantity of air discharged through an orifice or nozzle

due to a difference in pressure may be determined from
Q = 1096.5 CA-J^
where C = coefficient of discharge.
A = area of orifice in sq. ft.
p = pressure head in in, of water causing flow of air

through orifice.
For values of coefficients of discharge see "Coefficients of

Discharge for Air Measurements" in Part IV, Section II.
In case the pressure is expressed in ounces per square inch
these formulae become:
V = 1444.5.^ (17)
Vl444.5/
W (18)
and Q = 1444.5 CA-v/J^

\ W
The value to be used for W to be determined for each specific
case, as already explained.
Example . As an example of the application of the above we

will assume a case of a fan test made under the same atmospheric
conditions as those assumed for the example on page 14. That
is, the air to be at 83 F. and 15 depression, with the barometer
at 29.40 inches. What will be the velocity of this air at a

pressure of 1.5 inches of water as measured by a pitot tube? As
determined on page 16 the weight of air under the above con-
ditions will be 0.07142 lb. per cu. ft. Then from formula (15)
we find the velocity to be
The above formulae are sufficiently accurate for low pres-

sures such as are ordinarily used in fan work, but for high pres-
19
sures such as are met compressed air work, the error becomes
in
excessive and it be found necessary to use the following
will
thermodynamic formulae. For the flow through an orifice
from a higher to a lower pressure, where the absolute initial
pressure is less than twice the absolute pressure of the discharge
region,
V2 = 6552^Ti[l- (^)'''] (19)
where V2 = velocity in ft. per min. at discharge.

Pi = absolute initial press, in lb. per sq. in.
P2 = absolute final press, in lb. per sq. in.
Ti = absolute temp, degrees F. of entering air.
The discharge through an orifice into a region where the
pressure is greater than half the initial pressure, expressed in
cubic feet of free air per minute, may then be determined by the
formula
Q=63.oocA^V(irrM?fr ^^"^
where Q = cu. ft. free air per min.

C = coefficient of discharge.
A = orifice area in sq. ft.
As already shown for dry air at 70 F. and 29.92 inch baro-
metric pressure, the velocity due to a pressure of one inch of
water is 4005 feet per minute and for a pressure of one ounce
per square inch is 5273 feet per minute. Since the velocity
varies as the square root of the pressure, we have
^=^/ôrV
Vo \ po
= VA/Pl
\
(21)
Po
Taking and Vo the velocity corresponding
po as unit pressure,
thereto,assuming dry air at 70 F. and 29.92 inch barometer,
the above relation reduces to
V = 4005l/'p~ (22)
When the pressure is taken in inches
or = 52731/"?" (23)
when the pressure is expressed in ounces.
The table on page 21 gives thevelocity of dry air at standard
conditions for various pressures expressed both in inches and
ounces. The two tables on pages 22 and 23 give the correspond-
ing velocities of dry air under standard barometric pressure of
20
PROPERTIES OF AIR
CORRESPONDINQ PRESSURES AND VELOCITIES OF DRV AIR AT 70

AND 29.92 INCHES BAROMETER
II
in
m
t/3
u
ai
0.0!
UJ
r
M
PROPERTIES OF AIR
OOO ooo ooo oo
OC^-* iOiMW OOOO JOO
ccoec ^CoM 20)b-
0500
MOOC^
.-^(N
t-c^
(NCO
OOOiOi
OCO<N
t^ 04 CO IM 05 * MM
05 cc t^ . o * r^ 00X00 00050 OOi
00 (N
CO-* 00 <*?0<N Mt-O Tt^M-* OOOM J^MO ^^--ÔJ

^'^^ ^
^'^
QO
SSo> ooTt<o3 SJi^S M
ooo 'OO
-^<>'z! ^t:5-
O5 00-*
,-i(NM
OS-*00 (N-rOS MCC05
JcSS ^^t-
,_^
i>t^t^
OSTt-q?
tôboo oo
t~- "-I
rfiMM OOIMI^ Tt^t^CÎ t^00 ICM

OINM (N<NO Tj<M<N
(NCO 00 Xf-'^ OCOO T*<0<N (N^QO 'to
COM(N (n;oo rô<N -ôoM r^^"
u ooeO<N 0<M0 5o^-*
U5C05D <C?Ot~ t^tÔO OOOO OO
Oi
ooow MOCO 00 (MOO

<*0<N
lO O
lOOO
.-I <N (M 00
COCO'-i
tôoo MiOCD
lOOM l>OM ^(CO t>.t>.00 00O
CO CO CO
2s
c^q
>< lo
ICO
ooMO
C1MO
iOooiM
MTf^T}'
oioio
CJOCM
^<M2S
t-^M !S;*J2
lOt-CD
oocot^
MOM
OMt^ OM
OOOO
CO 00
00 00 00
IM MM-* -^-^lO OiO0
<is
^Ô M.^^ OOtÔ ^MOO Tt*iC'-' COO

M00
OOOO-H
rîo-*
COCO(N lOCOO (MOCOI COO<N M'^M
bCrîO
O'ÔO
OOMiO
OM
0<M
COMO Mt TtitÔ MiOOO OC2>0 QCO
li ,-<(N<N MM TjH-îO lOiOiO CCCOCD COt^t^ t-0000
o 00 t^ OcOh. OOOO w^
U. iôooo looio (Ncoej w:::;^! 1-1
(MCOO (MM
coSoo M<Noo ^rô t-wj;- .-iM0
ow* h-tîO
COO-* OOt-iC 00^
COMOO
)-hNN
MtÔ
mm-*
TftÔ
Th-'tiO
(N^Jb-
iOiO0 coco CO CCI>t^ tôo xo
r^^ lorô OCC-* TftîO iOOO QCt>-

covi-^
SmS ooo t^t^co 555215
J'QCJ^ CCXO ^.-O ifiO<N -tiO
tô
COMOO MCOO MCDO (M-*t-- O -"t coOM ^-03;
Sc^?l MM^ Tf-*-* lOiOiO lOCOcO
(NXt^ OCOt <NcOt1< CDOiO M<N0 ^'<^22 S'^

tîCO (N(NO t>-C^CO 0<NM ^Tf-4 t^M lOCO
WCOO MCOO t-i'^CO X .COM COOM COO
XX
'2???5 MM-* Tt
rfi Tj" iO00 lOCOCO eocot^ t^xx
CDOQ
Tj<XC^ OOiC OM o>oo <NOO Oj2
s^
MCOO MtÔ "^r^Q -^t
X M OX
O
CO lOClO
OX M lO O <Mf CO t^ '<< * '^. "^ ^. "^^
t~-_
Ci'mM M^^ Tf lO
'înê <ôi "^"^"J

i(K(r<)
^"t*^ rJ ririri w^
23

29.92 inches for different pressures and temperatures. One
table gives the velocity for even parts of an inch and the other
for even parts of an ounce, with the corresponding pressure in the
other unit.
Effect of Temperature and Barometric Pressure on Velocity

considered at the same pressure the effect of changing the
If
temperature of the air will change the corresponding velocity in
direct proportion as the square root of the absolute tempera-
tures. That is
V = Vo\^0l
^\460 (24)
+ to
The tables on pages 22 and 23 give the corresponding velocities
for dry various pressures and temperatures, but the veloc-
air at
ity for any other temperature may be determined from the
above formula.
In connection with fan work we have the same relation
that is speed, capacity and horse-
at constant pressure, the
power square root of the ratio of the
of the fan varies as the
absolute temperatures. At constant velocity the weight and
pressure of the air handled will vary inversely as the ratio of the
absolute temperatures.
The velocity of air at constant pressure not only varies with
any change in temperature, but also with every change in barom-
eter. The velocity of the air varies inversely as the square root
of the ratio of the barometric pressures.
Then we will have
V = Vo>J-^ (25)
or where we wish to correct for both temperature and barometer
V'VoVUxV? (26)
In the above formulae V represents the velocity of the air at

temperature t degrees Fahr. and barometer b, while Vo is the
corresponding velocity at temperature to and barometer bo-
Relation of Altitude to the Properties of Air

The diagram on page 25 shows graphically the effect of dif-
ferent altitudes on the properties of air, and the two lines of
relative air velocity and of air pressure and density are espe-
cially convenient in fan calculation.
24
PROPERTIES OF AIR
AdnD^Btî JO 93H0NI Nl 3dnSS3dcl Dldl3jM0dV8

'
5 '
5 '
o S ' 5
aN3Dd3d
25
As an illustration of the use of this diagram assume a case
where a fan is to handle 150,000 A. P. M. at 0.5 inch static pressure
at an altitude of 5000 feet. We must determine what sea level
conditions correspond to these conditions at the given altitude
and so be able to select a fan of the required capacity. From
the chart we find the relative pressure at this altitude is 0.825,
so that the sea level pressure corresponding to 0.5 inch at 5000
feet altitude will be 0.5-7-0.825=0.6 inch. The horsepower
required to operate this fan will be 82.5 per cent, of the rated
horsepower as given in the fan tables for the corresponding
pressure of 0.6 inch static.
Any given amount of air ascommonly specified will be in-
creased in volume by this same ratio when we consider an alti-
tude of say 5000 feet. Thus if we ordinarily require a definite
quantity of air for a certain purpose this volume should be
divided by 0.825 to determine the capacity required if the ap-
paratus is to be installed at an altitude of 5000 feet, and a fan
selected to handle this greater volume.
Effect of Temperature on the Volume of Air

Air at constant pressure changes its volume almost exactly
in proportion to its absolute temperature (460 -}- temperature
deg. Fahr.) The table on page 13 gives the relative volume
of a given quantity of air at various temperatures as compared
to the volume at 70. For instance, the volume at 160 will
be 1.17 times the volume at 70. Expressed as a formula
for usewith other temperatures than those given in the
table we will have, where Q is the volume at temperature t
and Qo the volume at to
The effect of temperature on the various properties of air is

shown graphically by the diagram on page 27. As air at 70 is
commonly used as a standard, these curves give the various
relationships relative to air at 70. Inasmuch as the fan tables
given herein are based on air at 70 F., the upper curve of this
diagram is especially applicable to fan calculations. Thus we
see that if the air to be handled by a fan is at 140, the velocity
and fan speed would have to be increased to 106.5 per cent, of
that given in the tables. Or if the velocity remains the same,
the pressure and weight of air handled at 140 will be only 88
per cent, of the rated capacity.
26
PROPERTIES OF AIR
Effect of Humidity on Velocity
consider the
It be noted that the tables herein given,
may
the standard temperature and
various properties of dry air at
pressure. But as a matter of fact, atmospheric air
barometric
is not dry, so that a
correction is necessary in order to reduce
the actual observed velocity to the standard condition of dry
This may be accomplished by
means of the following
air
cubic feet per pound of air being determined from
relation, the
the psychrometric charts on pages 36 and 37.
Actual vel. Cu. ft. per lb. air as observed

/
(28)
Vel. Dry Air \ Cu. ft. per lb. dry air
HUMIDITY
Humidity the moisture or water vapor mixed with the air
is
in the atmosphere, and the weight of water vapor a given space

will hold is dependent entirely on the temperature. The amount
of vapor in any given space is independent of the presence of
air, the only effect the air has being due to its temperature.
Absolute Humidity
Absolute humidity is the weight of a given volume of water
vapor at a given temperature and percentage of saturation and
is usually expressed as grains per cubic foot. The tables on
pages 38 to 45 give the weight of water vapor per cubic foot at
different temperatures and percentages of saturation.
Relative Humidity
Relative humidity is the ratio of the weight of water vapor
in a given space as compared to the weight which the same space
is capable of containing when fully saturated at the same tem-
perature. It is the ratio of the absolute humidity for the given

condition to the absolute humidity at saturation. The quantity
of moisture mixed with the air under different conditions of
temperature and saturation is usually determined by means of
some form of instrument in which a dry-bulb and a wet-bulb
thermometer are used.
Dew-Point
The dew-point is the temperature at which saturation is
obtained for a given weight of water vapor, or the point where

any reduction in temperature would cause condensation of
some of the water vapor. Any given amount of moisture must
have some temperature at which saturation will occur and any
further lowering of the temperature will cause condensation.
This then will be its dew-point.
Dry-Bulb and Wet-Bulb Thermometer

Usually the temperature of the air is determined by means
of an ordinary or dry-bulb thermometer. The wet-bulb ther-
mometer has the bulb covered with a piece of clean soft cloth
and should be wet or dipped in water before taking a reading.
Care should always be taken to keep the cloth free from dirt
and to use pure clean water. This thermometer will give a
depressed or lower reading than that of the dry-bulb thermom-
eter in proportion to the evaporation from the wet surface of the
28
HUMIDITY
cloth, and is a measure of the amount of moisture
this depression
in the air. This depressed reading corresponds to the tempera-
ture at which the air would normally saturate without any
change in its heat contents. That is, the total heat in the air
remains constant at a constant wet-bulb temperature. In order
to obtain a true reading it is necessary that the thermometer be
placed in a strong current of air.
The Sling Psychrometer

This instrument consists of a wet- and a dry-bulb thermometer
mounted on a strip of metal and provided with a handle which
permits of the thermometer being rapidly whirled through the air.
When being used the instrument should be whirled continuously
until no further drop in the wet-bulb reading is noted. The differ-
ence between the readings of the two thermometers is the wet-bulb

depression, and by referring to the tables on pages 38 to 45 or to
the charts on pages 35 to 37 the corresponding psychrometric
conditions may be determined. There are other forms of instru-
ments, generally of some stationary type, used for taking humidity
readings, but the instrument described is reasonably accurate
and is the one used by the United States Weather Bureau.
The Aspiration Psychrometer
The aspiration psychrometer shown herewith is more accurate
than the sling psychrometer, as it will be noted from the cut
that the bulbs of the two thermometers are enclosed in highly
polished tubes so the temperature is not affected by radiation
from surrounding objects.
A circulation of air is induced through the tubes by a small
suction fan located in the top of the instrument. This fan is
29
fflsJ ^^^
Aspiration Psychrometer
driven at a constant speed by means of the clockwork, which

will drive the fan several minutes with one winding, giving a
uniform strong current to produce a rapid evaporation from the
wet-bulb. The small rubber bulb is used to moisten the wick or
muslin covering the wet-bulb.
Relation of Humidity to Heating
To understand more thoroughly the relation of humidity to
heating, it is necessary to know that the temperature felt by the
body, or the sensible temperature, as it is called, corresponds
to the temperature of the wet-bulb thermometer; hence, the drier
the air the greater is the difference between the actual and sensible
temperatures.
Dry air heated much above the normal will still
be chilly, slight drafts are very noticeable and colds are easily
contracted.
The excessive evaporation from the skin lowers the tempera-
ture of thebody very rapidly, and as a result higher tempera-
tures are required than would be necessary for comfort if the
proper amount of humidity were present. On the other hand,
if the percentage of humidity is excessive, evaporation from the
body is below normal, with the result that the body heat is not
radiated as speedily as is necessary for comfort. In general,
the higher the humidity maintained the lower the temperature
required for the same degree of personal comfort.
Relation of Dry-Bulb, Wet-Bulb and Dew-Point Temperatures
The relation between the temperature as shown by the dry-
bulb and wet-bulb thermometer, and the relation to the dew-
point should be thoroughly understood by those expecting to
become at all familiar with the subject of humidity.
Dew-point, as previously stated, is the temperature at which
saturation is obtained for a given amount of water vapor. In
other words, the air is at the dew-point when it contains all the
30
HUMIDITY
moisture hold at a given temperature, and when it is im-
it will
possible to get the air to absorbmore water vapor without rais-

ing the temperature. When air has been reduced to the dew-
point, both wet- and dry -bulb thermometer register exactly the
same; for instance, air at 50 temperature, and 100 per cent,
saturation will contain 4.076 grains of moisture per cubic foot,
under which condition the dry-bulb and wet-bulb thermometer will
both register 50. If this air is heated, both thermometers will
rise, but the wet-bulb temperature will rise more slowly and the
relative humidity will be rapidly reduced: the dew-point re-
mains constant at 50, since any given number of grains of mois-
ture per cubic foot has a fixed and definite dew-point or tempera-
ture of saturation.
If a cubic foot of air at a temperature of 87, containing 4:076
grains per cubic foot with the wet-bulb temperature at 65, is
passed through a fine spray of recirculated water, it will absorb
moisture; the dry-bulb temperature will immediately begin to
fall,but the wet-bulb temperature will remain absolutely con-
stant at 65 until the dry-bulb temperature has dropped to the
wet-bulb temperature, namely, 65. As the absorption takes
place the dew-point will be gradually rising from 50 to 65,
when saturation is obtained. At ordinary temperatures the
absorption of one grain of moisture per cu. ft. lowers the dry-
bulb temperature approximately 83^.
Sensible, Latent and Total Heat

The is composed of the sensible heat or heat
total heat of air
due to the temperature of the air as indicated by the thermom-
eter, and the latent heat or heat of vaporization of the moisture
or vapor in the air. The total heat is a constant quantity for
any certain wet-bulb temperature irrespective of any change in
the dry-bulb temperature. This fact has been termed by
W. H. Carrier* "One of the four fundamental psychrometric
principles," and expressed as
"The true wet-bulb temperature of the air depends

entirely on the total of the sensible and the latent heat
in the air and is independent of their relative proportions.
In other words, the wet-bulb temperature of the air is
constant, providing the total heat of the air is constant."
"Rational Psychrometric Formulae" Am. Soc. Mech. Engrs., Dec, 1911.
31
Thus, if moisture is introduced into a certain quan-
sufficient
tity of air, the dry-bulb temperature of the air will be lowered
until it is the same as the wet-bulb temperature. This is simply
an exchange from sensible heat into latent heat required to
vaporize the moisture, keeping the total heat the same. If a
further lowering of the temperature takes place, the wet-bulb
temperature will lower and the corresponding total heat will be
less. If the air should be heated without the addition of more
moisture, the dew-point temperature of the air would remain
constant but the wet-bulb, as well as the dry-bulb temperature
would increase, and the total heat of the air would increase a
corresponding amount. The two psychrometric charts on
pages 36 and 37 will be found especially convenient for deter-
mining the total heat of the air for any wet-bulb temperature.
Psychrometric Charts and Tables
Psychrometric charts giving the properties of air as calcu-

lated by W. H. Carrier and published in his paper entitled
"Rational Psychrometric Formulae," which was presented
before the A. S. M. E. at the 1911 annual meeting, are shown on
pages 36 and 37. These two charts are to be used when calcu-
lations are being made in terms of pounds of air, while the chart
on page 35 should be used for cubic feet of air. For most pur-
poses of calculation it will be found preferable to use the pound
as a unit.
The various curves shown on these charts will be found espe-

ciallyvaluable in making air calculations. The grains of mois-
ture per pound of dry air are read by passing directly from the
dew-point, or intersection of the wet- and dry-bulb temperatures,
to the scale on the left edge of the chart. The B. t. u. required to
raise one pound of dry air one degree when saturated with mois-
ture, as also the vapor pressure, may be determined by passing
vertically from the dew-point to the proper curve, and then to
the corresponding scale on the left edge of the chart. The total
heat, in B. t. u., above zero degrees contained in one pound of
dry with moisture may be found by passing ver-
air saturated
ticallyfrom the wet-bulb temperature to the total heat curve
and then to the left edge of the chart. The volume of air in
cubic feet per pound may be found by passing vertically from
the dry-bulb temperature to either of the two volume curves
32
HUMIDITY
and then to the left edge of the chart. One curve gives the
volume of dry and the other of saturated air.
Example . of the use of this chart we will
As an example
assume dry-bulb temperature and 60 per cent, relative
air at 75
humidity. From the chart we find that the wet-bulb tempera-
ture will be 65.25, the dew-point 60, the grains of moisture
per pound of dry air 77; the heat required to raise one pound of
dry air saturated at 60 through one degree is 0.24664 B. t. u.;
and the vapor pressure of air saturated at 60 is 0.523 inches
of mercury. Passing vertically from the wet-bulb temperature
of 65.25 to the total heat curve and thence to the scale on the
left, we find the total heat above zero in one pound of dry air
when saturated at 65.25 to be 29.75 B. t. u. This, then, is

also the measure of the heat in a pound of air at 75 and 60
per cent, relative humidity, since the wet-bulb temperature is
the same.
The may be found by passing
cubic feet per pound of air
vertically from the drj^-bulb temperature to either of the two
volume curves, depending on whether the volume of drj' or of
saturated air is desired. To determine the volume of one pound
of partly saturated air as here assumed, we will have from the
chart,
Cu. ft. per lb. at 75 sat. = 13.88
Cu. ft. per lb. at 75 dry = 13.48
.40= Moisture
.60= Rel. humidity
.24
13.48
Cu. ft. per lb. at 75 and 60% = 13.72
The psychrometric chart on page 35 and tables on pages 38 to
45 are taken from the catalog of the Carrier Air Conditioning
Company of America. They show the grains of moisture per cubic
foot of saturated air at various temperatures, as well as the relative
humidity, the dew-point temperature and the grains of moisture
per cubic foot of air for different temperatures as determined
by the wet and dry-bulb of the sling psychrometer.
As an example of the use of the chart on page 35 we will
assume a case where the drj^-bulb temperature is 80 and the
wet-bulb thermometer reads 70. or a 10 depression. From the
intersection of the corresponding lines through these two tem-
33
COMPANY
FAN ENGINEERING-BUFFALO FORGE
to be 62 per cent. Pass-
peratures we find the relative humidity
to the left from this point of intersection to the
ing horizontally
saturation curve) we find
wet-bulb temperature line (called the
temperature to be 64.5. If the temperature of
the dew-point
the dry and wet-bulb readings
the air should be reduced, both
until they both read 64.5, when the air wil be
will be lowered
foot
saturated. The grains of moisture contained in each cubic
horizontal
be
of this air will found by continuing to the left on the
edge of the chart,
line through the 64.5 dew-point to the left
reading of 6.65 grains. If the temperature
where we have a
of the air be further reduced,
part of the moisture content will
the dew-point or saturation temperature will be
be condensed,
cubic foot will be corres-
lowered, and the grains of moisture per
pondingly less.
than can be deter-
In case more accurate readings are desired
as the one on page 35, the
mined from a chart on as small a scale
psychrometric tables may be used.
w,.th
shown on pages 36 ^nf. ^7
NOTE.-Large psychrometric charts asfurmshed on "-fj^^^^bythe Garner
sub-diiisions for accurate readings, will be New York l.ity.
Air Conditioning Company of America. 39 Cortlandt St.. ^
34
HUMIDITY
Noiivunivs IN33 uad

? r- ^
l (0 lO
36
FA N v.NaTNEERING-BUFF ALO
FORGE COMPANY
36
HUMIDITY
37
COMPANY
FAN E N G TKFF.RING-BIJFFALO FORGE
MOISTURE
RELATIVE
^ HUMIDITY, DEW-POfNTS AND GRAIIHS OF
PER CUBIC FOOT
^ r>
HUMIDITY
RELATIVE HUMIDITY, DEW-POINTS AND GRAINS OF MOISTURE.
PER CUBIC FOOT Barometric Pressure, 30 Inches
FAN
HUMIDITY
RELATIVE HUMIDITY, DEW-POINTS AND GRAINS OF MOISTURE
PER CUBIC FOOT
Barometric Pressure 30 Inches
PER CUBIC FOOT
HUMIDITY
PER CUBIC FOOT
^.N ENGINEERING BUFFALO FORGE COMPANY
RELATIVE HUMIDITY, DEW=POINTS AND GRAINS OF MOISTURE
PER CUBIC FOOT
HUMIDITY
RELATIVE HUMIDITY, DEW=POINTS AND GRAINS OF MOISTURE
PER CUBIC FOOT
PART II
APPLICATION
The principles ofFan Engineering have found application
for a great variety ofpurposes and an ever increasing use in the
manufacturing industries. These applications will be briefly-
treated under their proper heading, such as Heating, Ventilation,
Drying, Cooling, Mechanical Draft, Planing Mill and other ex-
haust systems and various other miscellaneous uses. Some of
these applications, such as heating and ventilation, are so com-
mon as to be more or less familiar to all engineers, while others
are of a more special nature and the requirements as well as the
apparatus used are not so generally understood.
SECTION I
HEATING
Although the heating of buildings is accomplished in many
ways, the fundamental requirements and the results desired
are the same in all systems; that is, to provide sufficient heat
to take care of the radiation and infiltration losses, and if re-
quired, to warm the air needed for ventilation.
Heat Losses from Buildings
The heating capacity depends on the amount of the heat
losses, so evidently the first step in laying out any heating sys-
tem is to determine the extent of these losses. The main source
of loss will be due to radiation, and as accurate data in the form
of factors or coefficients for the various building materials used
are available, the total heat loss may be determined when the
extent of the surface is known.
Each of the various materials used in building construction
has a certain capacity for transmitting heat, or we may say
exerts a certain resistance to the transmission of heat, and the
transmission of heat may be shown to be the reciprocal of the
resistance. This transmission is due to two components, radia-
tion and convection from the surface, and conduction through
the material. The radiation and convection factor is independ-
ent of the thickness, but varies with the height of the wall, with
the difference in temperature between the two sides of the
46
HEAT LOSSES
material, and with variation in the air movement or velocity
over the surface. While for extreme accuracy all of these
variables should be considered, for ordinary calculations we
may use a coefficient which will meet the average conditions.
We by NK
will represent this factor where N varies with the
;
temperature difference according to the accompanying table,

and K is a constant for any given material,
NK= Surface transmission for each material.
Tj^r^ = Surface resistance to the transmission of heat by

radiation and convection.
The conductivity of any material will vary with the thick-
ness, so that we will have
A = Conductivity of material itself from surface to surface
per unit thickness.
= Resistance of the material per unit of thickness to the
transmission of heat.
W = Thickness of the material.
- = Conductivity of the material.
W
7- = Resistance of the material to conduction.
A
L = B. t. u. transmitted per sq. ft. per hour per deg. difference.
As the total resistance is composed of the two factors ,-77^
W .
^^
and-^, we have the transmission in B. t. u. per square foot per
degree difference in temperature between the two sides of the

material,
_J_4.^" (29)
NK * A'

and the total transmission per square foot per hour with tem-
peratures ti and t2 on the two sides of the material will be
NK "^
A
In case we are to consider a double wall or a wall made up
of more than one material, we will have greater resistance due
to the extra surfaces adding their resistances and also to the
added resistances to conduction, thus giving a lower rate of heat
transmission. This will then give us
47
L= (31)
(ni]lKi
+ N2K2
'
-)-(lM:--)
In case the materials considered are very thin, but slight
error will be introduced if we neglect the conduction factor and
consider only the resistance of the surface.
This theory of heat transmission was first deduced by Peclet
and has been used by the majority of investigators for deter-
mining factors of heat transmission. The following values for N,
K and A are adapted from the original tables of factors for use
in these formulae. The values given for A are for a unit thick-
ness of one inch.
VALUES OF N FOR VARIOUS TEMPERATURE DIFFERENCES

Difference Between
HEAT LOSSES
factors as given apply to wall construction, where there is more
or less movement of the air over the surfaces due to the current
of air passing up the walls. With a roof or a floor this move-
ment of the air will be greatly lessened and the transmission
consequently decreased. The necessary modifications of the
factors for other than wall construction are in a great measure
a matter of judgment.
The application of the foregoing formula may best be shown
by a practical example, in which it is required to determine the
coefficient of transmission per square foot per degree difference
in temperature for a wall composed of two-inch pine boards
covered with building paper and corrugated iron, with a differ-
ence of 50 in temperature between the two sides. From the
accompanying table we find the value of N to be 0.956.
RNe BOARD BUILDING PAPER
ORRUGATED IRON.
N)K|i.Zl8
NjKz- I.I 5 N,Ka 1.43a
A 0.97
HEAT TRANSMISSION THROUGH WALLS.

We will have six surface resistances to consider, NK for the
two surfaces board being 1.218 and for the two surfaces
of the
of the corrugated iron 1,433. On account of the fact that the
paper and the iron are quite thin the resistance to conduction
would be very small and may be omitted for these two materials
without serious error.
We will then have the surface resistance ^^ = t o.o

INllVl l.Jlo
for the two board surfaces, and assuming the same surface loss
40
for paper as is given for sheet iron, we will have the resistance
for the two sides of the paper ,.

^ ^ N2K2
= . . ^
1.15
As already stated the resistance of the two surfaces of the
iron will be ^^
N3K3
= , .-^ The resistance to conduction
1.433
of the two-inch pine board will be -r-

W = 2
7^-7^;=- Then we will have
A. (j.yy
the total transmission in B. t. u. per sq. ft. per degree difference

in temperature between the two sides of the wall as,
L= 2~ = 0-224 B. t. u. per hour.

"^ 1.15"^ 1.433 "^
1.218 0.97
If we leave out the paper and consider a wall composed of

two-inch boards and corrugated iron, the transmission in B. t. u.
per sq. ft. of wall per degree difference in temperature between
the two sides of the wall would be
L= \1 jr = 0.279 B. t. u. per hour.

1 ^
"*" "^
1.218 1.433 0.97
The and 52 are compiled from the best au-

factors on pages 51
thorities andbe found to agree with modern engineering prac-
will
tice. Much of our information on this subject dates back to the
time of Peclet and other early investigators, but a great deal of
work has been done more recently in an effort to obtain authen-
tic data. The principal reason for discrepancies in this part of
the work is due to the great difference in building construction.
Thus, although as ordinarily considered, the radiation loss is the
principal factor to be considered and reliable data is available,
nevertheless, due to poor construction, the convection losses
may become so great that the apparatus will be unable to furnish
the heat required.
For this reason these factors are subject to modifications to
care for special cases such as exposure to winds, unequal distribu-
tion of heat, infiltration of cold air, etc. The German government
standards require these factors to be increased as follows:
Ten per cent, where the exposure is a northerly one and the
winds are to be counted on as important factors.
50
HEAT LOSSES
Ten per cent, when the building is heated during the day-
time only, and the location of the building is not an exposed one.
Thirty per cent, when the building is heated during the day-
time only, and the location of the building is exposed.
Fifty per cent, when the building is heated during the winter
months intermittently, with long intervals of non-heating.
It may
be noted that some engineers are inclined in a few
instances to use slightly higher factors than here given. For
instance, N. S. Thompson in his "Mechanical Equipment of
Federal Buildings" gives constants for concrete that are 50 per
cent, greater than the constants for brick herein quoted instead
of 20 per cent, greater as given below, while the constants for
brick agree very closely.
HEAT LOSS FROM BUILDINGS

B. t. u. Transmitted per Hour per Square Foot of Surface per
Degree Difference in Temperature
CONSTANTS FOR BRICK WALLS
Thickness
Inches
FOR VARIOUS WALL CONSTRUCTIONS
Thickness
of Board
Inches
HEAT LOSSES
Heat Loss Through Cold Storage Insulation
Extensive experiments have been made by various inves-
tigators who were interested in the subject of heat transfer
through the various materials commonly used for insulation '^
in cold storage work. Any apparent discrepancy between the^^*^
figures quoted by different authors for the same material is "-^
probably due to the different conditions under which the tests
were conducted.
^
The following table of coefficients has been compiled by E
F. E. Mathews, principally from data furnished by the Arm-,S*
strong Cork Co. This table was published in Power, August ^
8, 1911. These values are for use under the best conditions, ^^
and builders are advised to increase them by 25 to 50 per cent.,^^
depending on the physical condition of the insulation. O^
COLD STORAGE INSULATION fcj()
Transmission in B. t. u. per
in
Square Feet per Hour per Degree Difference
Temperature for "
d
r"
Insulating Slabs Q^
"Nonpareil" cork board (pure cork, no foreign binder)
1" 0.271 C^
I" "Rock" cork (water-proofed rock-wool composition board) 0.308 J^
1" "Lith" plain (mineral wool flax-fibre composition board) 0.329* i
1" "Lith" water-proofed (same as above, water-proofed)
1" "impregnated cork board" (gran, cork and asphaltic binder)
0.350 ^H
I" indurated fibre board (indurated wood-pulp board)
0.371 ifr^
0.417 C
Built-up Insulation (wood and air space)
1" American spruce 0.700 i^^
%" C3
( J-i"
dressed and matched spruce
sp.X
sp.
( J^" sp. paper Ji" sp.) (K"
sp.) (1" air space) ( H" sp. paper Ji" sp.)

paper H"
sp. paper J/g"
0.198
0.177
^ ^
^
6 thicknesses, J^" sp., 3 papers, 2 air spaces arranged as above 0.144 ^t^
8 thicknesses, J4" sp., 4 papers, 3 air spaces arranged as above 0.113 _
Built-up Insulation (wood, paper and fill)

Q>
H" sp. paper %" sp.) ( %" sp. paper Vg" sp.) 0.198 t-
(
( H"
(H"
sp.
sp.
paper %" sp.) (4" min. wool) ( %" sp. paper "J^" sp.)
paper J4" sp.) (8" mill shavings, damp) (H" sp. paper
0.092
sp.).. ..0.088W _. ^
W
( H" sp. paper J4" sp.) (1" mill shavings, dry) ( Vs" sp. paper Vs" sp.) 0.066
( 14" sp.
CA" sp.
paper ]4" sp.) (8" granulated cork) ( %" sp. paper %" sp.)
paper J4" sp.) (1" Nonpareil cork) Hi" sp. paper %" sp.)
0.079
0.129
^
^
( J^'sp. paper) (1" Nonpareil cork) (paper H" sp.) 0.136 V^
( H" sp. paper) (2" Nonpareil cork) (paper Vh" sp.)
( H" sp. paper) (3" Nonpareil cork) (paper S/i;" sp.)
"-jlx?
"5?j;
O -
( H" sp. paper) (4" Nonpareil cork) (paper sp.) M 9-^^

0.202
( Vs" sp.) (1" pitch) ( H" sp.) .
(K" sp.) (2" pitch) iVs" sp.) 0.177
Built-up Insulation (wood, paper, air space and fill)

( H" sp. paper '^4" sp.) (1" air space) ( li" sp.) (6" min. wool) ( H" sp. paper
14" sp.) 0.062
(14" sp. paper Ji" sp.) (1" air space) ( H" sp.) (6" gran, cork) ( H" sp. paper
14" sp.) 0.061
(14" sp. paper H" sp.) (1" air space) ( 14" sp.) (2" Nonpareil cork) {14" sp.
paper 14" sp.) 0.067
63
( H" sp. paper %" sp.) (1" air space) (2" Nonpareil cork) (paper H" sp.) .0.088 .
( Vi" sp. paper V^" sp.) (1" air space) (3* Nonpareil cork) (paper V^" sp.) .0.071 .
ili" sp. paper H" sp.) (1" air space) (4" Nonpareil cork) (paper J^" sp.). .0.050
(Ji" sp. paper Vs" sp.) (1" air space) (5" Nonpareil cork) (paper J^" sp.). .0.038
Brick Wall and Sheet Cork

(13" brick wall) (2" Nonpareil cork) 0.115
(13" brick wall) (4" Nonpareil cork) 0,061
NOTE Sp. designates American Spruce,
The data given in the table following is an extract from a

report on Heat Transmission of Building Materials, submitted
to the American Society of Heating and Ventilating Engineers,
1913, by Prof, L. A, Harding,
These results were obtained in a specially constructed test-
ing plant, using temperature differences that ordinarily occur,
either in heating or refrigerating practice. The testing boxes
were made up of the various materials stated, having approxi-
mately 100 sq, ft, of surface. Heat was introduced inside the
boxes by means of electrical resistance coils, the air surround-
ing the box being artificially cooled, A strong circulation of
air was maintained both inside and outside of the box by fans.
TABLE OF UNIT HEAT TRANSMISSION

B. t. u. Transmission per Sq. Ft. per Hour per Degree Difference in Temperature
of Air in Contact with the Two Sides
One 4-in. Hollow Tile 0.625
1 in. Concrete (1-3-5 Mix.) 4.29
Vi in. Lumber (T and G) 0.83
One Air Space (from 1 in. to 6 in.) 1.66
1 in. Mineral Wool (Dry) 0.666
1 in. Pitch 0.79
1 in. Shavings (Dry) 0.666
1 in. Granulated Cork 0.479
1 in. Corkboard (all Cork) 0.26
1 in. Hair Felt 0.31
1 in. Indurated Fibre Board 0.416
1 in. Compressed Mineral Wool Board 0.33
Heat Loss from Galvanized Iron Pipes

As already explained (page 47) the resistance of any surface
to the transmission of heat the sum of the various resistances
is
to be met in each case, and the transmission or heat loss is the

reciprocal of this total resistance. In the case of a galvanized
iron duct the conductivity of the material may be neglected,
and we have to consider only the resistance to the transfer of
heat from the air to the metal on one side and from the metal
to the air on the other side. The air in the pipe would have a
direction of flow parallel to the surface so that the heat
transfer would be as shown in formula (97), page 404, for
54
HEAT LOSSES
longitudinal flow. The conditions on the outside of the pipe or
duct would be approximately those of any direct radiation
placed in the room.
According to the best authorities the coefficient of trans-
mission K for direct radiation from the outside will vary from
1.6 to 1.8 B. t. u. per sq. ft. per hour per degree difference in
temperature between the material and the external air. A
factor of K = 1.7 B. t. u. is an average value commonly used.
Assuming a velocity of 1500 ft. per minute for the air through
the pipe, we find from the diagram giving "Rate of Heat Trans-
mission for Longitudinal Flow of Air," on page 406, that for
the internal surface of the pipe K = 7.5 B. t. u. The total
resistance of the duct will then be
K = 1.38
For average conditions as outlined above we may assume
for the heat transfer through galvanized iron ducts per sq. ft.
per hour per degree difference in temperature between inside
and outside
K = 1.4B. t. u.
H=K (ta-tr)=1.4 (ta-tr)
where H = heat lost per square foot per hour.
ta = temperature of the air in the pipe.
tr = temperature of the room.
Heat Loss Due to Infiltration

The heat due to leakage and infiltration is often a diffi-
loss
cult quantity to determine, inasmuch as it may depend on so
many different factors, such as size, height and construction
of the building, distribution of doors, windows and ventilators,
and the object for which the building is used. For this reason
no fixed rule can be given, and the allowance to be made for
this loss is necessarily a result of experience and good judgment.
It is customary to allow from two changes per hour to one change
in two hours as a measure of this loss, depending, as already
stated, on the circumstances of the case considered. For the
average application one air change per hour is usually a satis-
factory allowance.
Thus if the building is very large the ratio of cubic contents
to wall surface would be greater than with a smaller building,
hence the air change due to leakage would be less frequent,
55
The existence of large doors frequently opened or of ventilators
makes a more rapid air change due to leakage, as does also
loosely fitting windows or other faulty construction. Many of
the materials used for roofs and siding, such as tile or galvanized
iron, are notoriously bad as regards leakage.
The heat required in B. t. u. per hour to warm this air from
the outdoor to the room temperature w411 be
TT _ cubic contents X changes per hour X temp, rise
'
55:2
Heat Required for Ventilation
The heat required for ventilating is easily computed when

the air supplied per hour is actually known. If we consider the
specific heat of air at constant pressure to be 0.2415 and the
weight of one cu. ft, of air at 70 is 0.0749 lbs., one British thermal
unit of heat will raise the temperature of one cu. ft. of air
= 55 90 F
0.2415X0.0749
From this we may deduce the formula.
H'=?î|zA) (32)
where H' = B. t. u. per hour required for ventilation.

Q = cu. ft. of air per min. required for ventilation.
tr = room temperature.
ti = entering, or outdoor, temperature.
Examples above formula will be
illustrating the use of the
found in Section VIII, Part IV. For a consideration of the
heat given off by the occupants of a room or by various sources
of heat, such as lights and machinery, see "Special Ventilation
Requirements" on pages 61 and 62.
Air Quantity and Final Temperature Required for Heating
In heating a building by means of the fan system, there are
three factors to be considered. These are, the heat loss due to
transmission and infiltration; the quantity of air required as a
heat carrier; the final temperature and the temperature rise
of this air in passing through the heater. Ordinarily the heat
loss and one of the other factors are given, with the third to be
determined. It may be required to use all return or room air,
all outdoor air, or a mixture of the two. The relations between
the above factors may be expressed by the following equations:
56
HEAT LOSSEl
Q 55.2 H
and , 55.2 H ,
, ....
*^ = -60-Q+*^ (^^^
where H = heat loss in B. t, u. per hour due to transmission

and infiltration.
Q = cu. ft. of air per minute.
t2 = temperature of air leaving heater.
tr = temperature of room.
When using all or part outdoor air there will be sufficient
heat required at the heater not only to take care of the heat loss,
but also to raise the temperature of the air from entering tem-
perature ti, to the room temperature, tr. The total tempera-
ture rise will then be t2 ti, and
jj,^ 60Q(t.-tO ^35^

OO.ii
where H' = total heat in B. t. u. per hour required at the

heater,
ti = temperature of air entering heater either out-
door or a mixture of room and outdoor air.

The amount of steam required will be
T
TI/
r = : 7 I lbs. per hour.
heat
latent steam of
TT/
or approximately lbs. steam per hour =
67
SECTION II
VENTILATION
Any room or building used for the habitation or
congrega-
tion of human beings should be provided with a plentiful
supply
of fresh air. Strictly speaking, good ventilation is
merely a
relative term, and the standards as ordinarily
accepted are a
compromise that will answer the purpose of keeping
the air in
a building in a fairly fresh condition. The requirements of
ventilation are, first, maintaining certain standards of purity
of the air within the room or building;
second, the removal and
prevention of odors; third, the removal of the
bodily heat of
the occupants together with the heat from
such other sources
as illumination and power; and fourth, the
prevention of exces-
sive rise in humidity which usually
accompanies the rise in
temperature from bodily heat.
Many of the existing standards of ventilation have been
founded on the belief that carbon dioxide was the
dangerous
element in expired air. The requirements of ventilation
as to
air purity are more or less arbitrary,
and no rational standard
has ever been fixed. Later investigations would indicate that
carbon dioxide is harmless, and interesting only as indicating
how much respiration the air has undergone. In this way it
serves as an index of the contamination of the air with
organic
impurities from the lungs and bodies of the occupants.
These
organic poisons are little understood, although they
undoubtedly
constitute the real danger in impure air. The standard of purity
which has usually been considered satisfactory is from six to
eight parts of carbon dioxide in 10000 parts of air, but it is
certain
that ten times this amount would not be injurious if provision
were made for the removal of organic impurities. In all prob-
ability the best index of
good ventilation in so far as purity is
concerned is freedom from objectionable odors.
It estimated that the average adult, at rest or doing light
is
work, willbreathe approximately 0.25 cu. ft. of air and exhale

0.01 cu. ft. of CO2 per minute (0.6 cu. ft. per hour), and that
58
VENTI LATION
only about five per cent, or less of the oxygen is
taken out of a
The air of poorly ventilated rooms will show a
breath of air.
slight diminution in the oxygen, accompanied

by a correspond-
ing increase in carbon dioxide, organic pollution,
and moisture.
The poisons in the air due to the presence of too many
persons
relative to the supply induce a lowering of the vital processes

and a loss of muscular strength.
Ordinary outdoor air will contain on an average about
four
ventilation is ordinarily
parts of CO2 in 10000 and fairly good
not more
considered to exist in a room where the air contains
than from six to eight parts of CO2 in 10000. That is, if a great
amount of CO2 exists in the air, it is considered as having been
The fol-
breathed too often and unfit for further respiration.
per hour by the
lowing table gives the amount of air required
hour, if it is desired
average person, exhaling 0.6 cu. ft. of CO2 per
to maintain the corresponding number
of parts of CO2 the air m
per 10000.
with outdoor air containing four parts of CO2
Parts of CO, in 10000

FAN ENGINEERING-BUFFALO FORGE COMP
ANY
except where the building is artificially
cooled. In crowded
audience halls, however, and even in
school-rooms it is the de-
termining factor. Each adult occupant
gives ofT an average of
400 B. t. u. per hour, of which approximately 150
B. t u may
be assumed to be latent heat of evaporation,
while not more
than 250 B. t. u. will be sensible heat given
off by the breath
and by convection to the surrounding air.
On this basis if
each occupant is supplied with 30 cu. ft.
of air per minute or
1800 cu. ft. per hour there will be a rise of approximately
eight
degrees above the temperature at which
the air is introduced
mto the room, so that in order to maintain we will say
70 degrees
in the room, the air would have to
be reduced to 62 degrees
There is evidently a limit to the difference of
temperature
allow-
able between incoming air and room
temperature,which depends
largely on the size and arrangement of inlet
openings as effecting
the production of cold drafts. The practical limit
is found in
standard methods of ventilation to be between 5 and
10 degrees.
Therefore, the limiting quantity of air required for
the removal
of bodily heat must be taken approximately
between 25 and 50
cu. ft. per minute per adult occupant.
This, as may be noted,
also gives a very satisfactory standard of
purity.
While 1800 cu. ft. of air per hour or 30 cu. ft.
per minute
(expressed as 30 A. P. M.), when used as a standard
for over-
head ventilation, is in the average case amply sufficient
to take
care of the heat and moisture from the body,
when the air is
supplied through many small openings distributed
about the
room a smaller quantity of air may often be supplied.
Several
different systems of this character have been
used, such as in-
troducing the air under the seats in a theatre, or
through a small
opening at each desk in a school. By this means
a more uni-
form distribution of the air is obtained than is
possible with the
over-head system, with greater assurance that
each occupant
of the room will receive the desired amount of fresh air.
Carbon Dioxide Determination

Various methods for making analyses of the air to
ascertain
the CO2 content have been used, but it is
generally considered
that the more simple methods are little better
than approxima-
tions. The more dependable methods require carefully prepared
apparatus, and an operator who has had considerable
experience
in this or similar chemical work.
One of the methods quite
60
VENTILATION
generally used for this determination is called the Pettenkofer
method.
The Pettenkofer method of analyses for carbon dioxide is
absorb the CO2
based on the fact that barium hydroxide will
from the atmosphere. A measured bottle or flask is used to
of bellows or air pump
collect the sample of air, some form
the room air into the bottle. This operation
being used to force
is continued for several
seconds, or until the air in the bottle
has been changed quite a number of times
(six to ten).
After the sample is collected, a definite quantity of standard

barium hydroxide is inserted into the bottle to absorb the CO2
few drops
from the sample of air contained. At the same time a
of penolphthalein is added to
the barium hydroxide in the
bottle, giving the mixture a reddish
color. The sample should
being frequently
then be allowed to stand for at least an hour,
shaken, although the final operation may be left till the next
day if desired. The excess of barium hydroxide is then

titrated
acid, by dropping the acid into the reddened

with standard oxalic
solution untU the color disappears. This oxalic acid should
be dropped into the sample bottle from a graduated
burette,
titrate the barium
so that the exact amount of acid required to
hydroxide not absorbed by the carbon dioxide may
be meas-
Previous determinations having been made to find
the
ured.
oxalic acid required to titrate a quantity of barium
amount of
in the
hydroxide equal to that put into the bottle, the difference
the sample a measure of the
acid used before and after taking is
barium hydroxide uniting with the CO2. One c. c. of the oxalic
acid is equivalent to ^ô c. c. of carbon dioxide.
Special Ventilation Requirements

There are other factors that may have to be considered in
making a determination of the air quantity to be supplied for

ventilating purposes. of these special cases are where
Some
provision is to
to be made
care for the heat given off by furnaces
lighting apparatus;
or machinery; the effect of gas jets or other
heat radiated from the bodies of the occupants
or to remove the
of the building.
The allowance to be made for many sources of heat, such

as
a matter of experience and good judgment,

furnaces, is often
power the
but in case of machinery using a known amount of
61
FAN ENGINEERING-BUFFALO FORGE CO MPANY
heat expended may be determined on the
basis of 2545 B. t u
per hour or 42.416 B. t. u. per minute per
horsepower.
The following data concerning the heat given off
to the air
by various electric lights is based on the standard of
2545 B t u
per hour per H. P.; and 1 H. P. =746 watts,
giving 3.41 B. t. u
per watt as the heat radiated. This gives the B. t. u. per hour
for the following lamps:
25 watt lamps q-
50 watt lamps -
2X
400 watt lamps ; ' ; ,^^X
600 watt enclosed arc ;.;.;; .' .' .' .' ." ." ."
;

; ' ' ; ; ; ; l^l^

Prof. Kinealy quotes the following values for
the heat radia-
ted in B. t. u. per candle power per hour.
Gas, ordinary split burner onn
Gas, Argand burner ^XX
Gas, Auer burner '^x"
Gas, Welsbach burner, 16 c. p. . . A
Petroleum .'......
1^0
Electric, incandescent ?V
Electric, arc
43
The following data
gives the values commonly quoted as the
bodily heat given off in B. t. u. per hour per person.
Child 6 years old o.^
Adult at rest '.'.'.'.'.'.'.'.'.'.'.'.'.'.'..'.
Man 3I0
30 years old in an atmosphere with a temperature of 31 F fiOO
Adult in old age ' '
ogX
The amount of heat in B.
usually assumed as given off
t. u.
per person per hour in an atmosphere at 70 F. is
400 for adults
and 200 for children. These are the figures generally
used when
the heating effect of the occupants of assembly
halls, audito-
riums, or factories is taken into account.
The proper allowancefor the above sources of heat is of
especial importance in the design of apparatus used for cooling
a building in the summer-time. The heating effect
of the direct
sun on walls and glass surface has also to be
considered, the
ordinary factors in B. t. u. per hour per square foot of surface
being:
Sun effect 13-inch brick wall, 6.0 B. t. u. per sq. ft. per hr.
Sun effectglass, 150.0 B. t. u. per sq. ft. per hr.
The Fan System for Heating and Ventilating

While for heating purposes the fan heating system
may or
may not be used, depending on circumstances and the require-
ments to be met, yet for purposes of ventilation the fan
system
is practically without competition. The fan system may be
62
VENTILATION
Three=Quarter Housing Fan, Left=Hand Top Horizontal Discliarge, Blowing Air

Through and Underneath Heater into Brick Plenum Chamber
used to supply both heat and fresh air for ventilation or it may
be used in conjunction with some form of direct radiation which
is to care for the heat losses. When used for ventilating pur-
poses the fan will be required to supply whatever amount of air
is specified to meet the ventilation requirements.
Arrangement of Apparatus
Fan system apparatus, consisting of a fan andsome form of
indirect heating coil, may be arranged either to exhaust the air
through or to blow through the heater, commonly called the
exhaust -through or draw-through and the blow-through appara-
tus. Each arrangement possesses its own peculiar advantages
but the selection depends largely upon the individual require-
ments of the installation. An exhaust fan is slightly loss effi-
cient than a blower, but in a draw-through system the air blows
directly from the fan into the piping system with but little, if
any, change in velocity. On the other hand when using a blow-
through apparatus the velocity of the air leaving the fan must
be reduced through the heater and then raised again through
the piping system, both changes entailing a loss in pressure.
6.3
The exhaust-through apparatus is usually employed in fac-
tory buildings on account of its compactness as well as the

advantage gained by connecting directly to the piping system.
In this case the temperature of the air delivered will be the same
to all parts of the building. The blow-through apparatus is
used in public buildings or wherever different temperatures and
independent temperature regulation are required for different
rooms of the building. The use of the by-pass around the heat-
ing coils permits the mixture of hot and cold air in any desired
proportions, by the use of a mixing damper at the point where
the two ducts from the heater and from the by-pass join to form
one duct leading to the room. In the case of public buildings
the fan frequently blows the warm air into a space called a
plenum chamber, from which the air ducts radiate to the various
rooms of the building. For this reason this system is sometimes
called the plenum system of heating and ventilating.
Upward and Downward Systems of Ventilation
For audience halls the problems of air distribution and avoid-

ance of drafts are greatly increased owing to the usual large
dimensions of such buildings, and the density to which they are
peopled. Two plans of ventilation are in vogue, usually dis-
tinguished as the upward and the downward systems of ventila-
tion. In the upward method the air is admitted through
perforations in the floor underneath the seats and is allowed to
escape through ventilators in the roof. In the downward system
the air is admitted through registers in the walls at a height of
several feet above the floor, and removed through vent registers
in the walls at the floor line in the same manner as in school
buildings. In large auditoriums the upward method is doubt-
less preferable when the makes it permissible.
architectural design
A perfect distribution of the air can be secured, and the air flow
is upward in accordance with natural currents induced by the
heat of the body and the breath, the products of respiration are
immediately carried away, and the incoming air is uncontam-
inated. This method of ventilation is exceedingly efficient, as
a high standard of purity can be maintained at the breathing
linewith a comparatively small air supply.
Upward ventilation to be successful requires a very careful
arrangement of the supply openings on account of the greater
64
VENTILATION
liability of drafts. The velocities are necessarily low, and the
registers are so small that a very large number is needed to
convey the necessary air. The plenum chamber for the supply
is sometimes out of the question, and on this account the down-
ward system, which is in almost universal use in schools, is

extended to churches, theatres and halls with high ceilings.
With a proper arrangement of fresh air and vent registers and
ample air supply excellent results are obtained. To insure such
results exhaust systems are frequently relied upon, the vent
registers being connected with suction fans which maintain a
steady draft.
Schoolhouse Ventilation
Modern school buildings offer most exacting requirements in
heating and ventilation. On account of the large number of
pupils seated in one room, a very rapid air change is required,
and must be accomplished without drafts. The tempera-
this
ture must be uniform everywhere, and ventilation must be ade-
quate. Even elaborate systems can not secure entirely perfect
distribution of air, and the only practical and successful method
of insuring ample ventilation in all parts of the room is to supply
air considerably in excess of the theoretical requirements. The
necessity of this added capacity, or factor of safety as it may be
termed, is often overlooked in writing specifications for school
buildings. Thirty cubic feet of air per pupil which is usually
specified will allowfrom six to seven parts CO2 in 10000. In-
dividually this is ample, but collectively insufficient, since to
insure that this per cent, of CO2 is nowhere exceeded, it would
probably be necessary to supply an average of nearly 40 cu. ft.
per pupil.
Ventilation of Industrial Buildings

Where an industrial building is heated by means of a fan
system apparatus and no special air requirement exists, except
in certain cases, it is not customary to provide for ventilation
aside from taking the air for heating purposes from the outdoors.
Certain industrial processes require ventilation either for cooling
or for the removal of obnoxious gases and fiimes or of steam.
For such cases advisable to use an exhaust system where
it is
practicable. should be exhausted if possible at the

The air
point where the heat or objectionable gases escape.
65
It is customary to suspend a hood directly above the source
of heat or gases, this hood being connected by a duct or pipe to
the inlet of an exhaust fan or to a vent stack. These hoods are
usually so proportioned as to obtain a velocity of from 75 to 250
feet per minute over their area, according to the location of the
hood. Full directions for the design of such a system, with
data on the size of hoods and piping to be used, will be found
under "Exhaust Systems" on page 93, and on "Proportioning
Piping for Exhaust Systems" on page 129. Data on the design
of piping systems for various purposes will also be found under
"Air Ducts," Part III.
A few of the more common air changes provided for ordinary
conditions are:

Laundry 1 to 3 minute air change depending on the size
of the room and the concentration of the heat.

Hotel kitchen 4 minute air supply and 3 minute exhaust.
This tends to place the room under a slight vacuum, so that any
leakage at the doors is into rather than outward from the kitchen.
Engine and boiler room 3 minute supply and 4 minute
exhaust.
Foundry 15 minute change when
air air is taken from
outside.
Roundhouse 10 to 12 minute supply in order to keep the
air free from steam and smoke.
Cooling occupied rooms in summer without refrigeration
usually calls for from 4 to 6 minute air supply.
66
SECTION III
Air Washing, Cooling, Humidifying,* Drying
AIR WASHING
On account of dust and soot introduced by a ventilating
system, some form of air washer or air filter
is essential where
cleanliness is of importance. The spray type has superseded
cloth screens and other methods of wet cleaning on account
of
Its greater efficiency and is now standard practice
for ventilat-
ing equipment.
The advantages to be derived both in industrial establish-
ments and in public buildings by maintaining any desired
degree
of moisture in the air, as well as freeing it
from impurities, have
become very widely recognized. This process is
generally
termed air conditioning. This conditioning can be
most suc-
cessfully accom.plished by passing the air
through a spray type
of air washer or humidifier where additional
moisture is desired,
or by using a spray type dehumidifier when the
moisture con-
tent of the air is already too great and requires reduction.
Humidity in Heated Buildings

In schools and other public buildings the humidity of the
air is of more consequence than is usually
supposed. The
amount of moisture which air can hold at saturation per unit
volume increases rapidly with the temperature as shown by the
psychrometric chart on page 35. Air normally has a humidity
varying from 40 to 50 per cent, of saturation, while if much
above or below these limits it becomes uncomfortable if not
actually injurious to the health. Hence air at 70 should contain
from 3.5 to 5.5 grains per cubic foot, while at 0 it contains only
about 0.3 grains and at 32 about 1.25 grains, so that in the usual
systems of heating, with 32 outside, the humidity of the air when
heated to 70 would be only 15.5 per cent. The effect of this ex-
*NOTE. For a general discussion of the subject of Humidity see page 28,
Part I.For details of the performance and dimensions of Carrier Air Con-
ditioning Apparatus see Section VI, Part IV.
67
treme dryness is undoubtedly very harmful to the mucous mem-
brane in the nose, throat and the lungs, and may be considered
a contributing source of many throat and pulmonary diseases.
The proper humidity to maintain in public buildings is from
35 to 50 per cent. The humidity to be recommended in good
practice is 40 per cent., with a room temperature of 68 F. This
corresponds to about three grains of moisture per cubic foot of
air and a dew-point of 42. Even this will cause slight conden-
sation on the windows in extremely cold weather and a lower
humidity should be maintained in very cold weather if con-
densation on the windows is objectionable.
Humidity in Manufacturing Establishments
In the case of industrial installations the amount of moisture
required in the air will vary widely according to the nature of
the process, some requiring high and others low relative humid-
ity. In textile mills the necessity of humidifying and cooling
the air has long been understood. Just as in many instances
the fan system has superseded other methods of drying materials,
it is to be expected that air conditioning apparatus with auto-
matic control of humidity will find new applications in which

economy of operation will justify the expense.
COOLING
One of the special developments in connection with the fan
system of ventilating the cooling of a building so that the
is
indoor temperature in summer will be lower than that outdoors.

A limited amount of cooling may be obtained by passing the air
through an air washer in which cold water is used, and for many
purposes this will be found sufficient. In case a considerable
temperature difference is desired, or when a great amount of
heat is to be taken care of, as from machinery and other appara-
tus in a factory, a special form of washer known as a dehumidi-
fier is used. This is generally operated in two stages or sets of
sprays, one using cold well water and the second using refrig-
erated water. With such an apparatus any desired dew-point or
per cent, of relative humidity may be maintained in the room to
be conditioned, and the air may be delivered as low as 39 or 40.
It is not generally considered desirable to have too great a
difference between the room and outdoor temperatures, on
account of the chilling effect to persons coming mfrom the out-
68
COOLING
door air. A difference of from 10 to 15 will generally be found
the most desirable. The incoming
must be cooled to a
air
temperature enough lower than the room
to take care of any
heat generated in the room, as well as
the heat transmitted
through the walls from the warmer outdoor
air. In case the
relative humidity of this air is then too
great, the moisture con-
tent may be reduced by lowering the
temperature still further
so condensmg a portion of the moisture from the
air The
amount of moisture contained at different saturation
tempera-
tures may be found from the
psychrometric tables on pages 38
to 45.
Relation of Cooling Effect to Percentage of

Relative Humidity
In the moist air system of humidifying it
is evidently essen-
tial, as shown in the table on
page 71, that the difference between
the dew-pomt temperature of the incoming
air and the room
temperature shall not exceed a predetermined value,
depending
upon the percentage of humidity to be maintained.
The mini*^
mum temperature at which air can be introduced is evidently
the dew-point or saturation temperature at the
apparatus. This
permissible temperature rise limits the possible
cooling effect to
be obtained from each cubic foot of air as shown
in the table of
cooling capacities. This relationship is of primary importance
m the design of the humidifying system and the
disregard of it
has been the chief cause of failure or of unsatisfactory
operation.
COOLING CAPACITY OF CARRIER HUMIDIFYING SYSTEM
Per Cent. Humidity

I
FAN ENG NEERING BUFFALO FORGE COMPANY

dependent on the other. In every industrial air conditioning
plant there are four sources of heat which must be taken into
account in the design of the system:
(a) Radiation From Outside Owing To The Maintenance
Of A Lower Temperature Inside. At ordinary humidities
this is negligible, but at high humidities and in dehumidi-
fying plants it is an important factor, owing to the in-
creased temperature difference. This may be calculated
from the usual constants of radiation.
(b) Heating Effect Of Direct Sunlight. This is especially
noticeable from window shades and exposed windows and
skylights where the entire heat energy of the sunlight is
admitted to the room, and from the roof which constitutes
the greater amount of sunlight exposure and which in
ordinary construction transmits heat much more readily
than the walls. Precautions should be taken where high
humidities are desired to shade exposed windows and to
insulate the roof thoroughly. Ventilators in the roof are
of great advantage in removing the hot layer of air next
it and those of ample capacity should always be
provided.
(c) Radiation Of Heat From The Bodies Of The Operatives.

This amounts to about 400 to 500 B. t. u. per hour per
operative, about one-half of which is sensible heat, .the
other half being transformed into latent heat through
evaporation.
(d) Heat Developed By Power Consumed In Driving
Machinery And In Manufacturing Processes In General.
According to the laws of conservation of energy, all power
used in manufacturing is ultimately converted entirely
into its heat equivalent. Each horsepower of energy,
therefore, creates 42 1^ B. t. u. of heat per minute, which
must be cared for by ventilation. In high-powered mills
this is the chief source of heating and is frequently suffi-
cient to overheat the building even in zero weather,
thus
requiring cooling by ventilation the year round.
It must be remembered that in cooling moist air the

latent
heat removed in condensing the moisture is usually of more

importance than the reduction in the sensible heat of the air
itself. The total heat removed may be determined from the
It should
total heat curve of the diagrams on pages 36 and 37. -
also be noted that the total heat of the air is dependent

on the
wet-bulb temperature only, and the wet-bulb temperature
should always be used in such calculations.
70
'
HUMIDIFYING
duiax uioo^ pue )U!0(} i
oiou: Scoooo
-M3Q U3dM)3q 33U3J3JJin X oi C5 OO'^'-^
.vj :îTi:>) CO MM CO
u.
o duiax }U!Od-M3Q
u duiax uioo^ puB juioj ooo toooo

M3Q Ud3M)dq 33U3J3jjj(J
>u
^^ duiax )ui0cf-M3a 000 lOOiop
duiax moo^ puB juiod i Oiot2 J5{2S5t-

M3Q U33A\)3q 33U3J3JJIQ , CjMm' MM^.*
O 0 1>- t^
duiax Juiod=M3a t^
C^J CO
uj o
Q S CO ^^ "M ^' '
uo
o
Otf duiax uioo^ puB luioj
=M3(I U33M}3q 33U3J3JJIQ
uz
c^< diuax }U!0d=-M3Q
w>
tu f-
dUI3X uioo^ pUB ?UIOd
^f*<^(\ Ud3M)3q 33U3J3jJ!(I
Sg COt>.'l^ t^ GC GO 00
u. duiax |uiod=^3(i coot>. CO 05 -^ M

odco'i^ <N-43(
QO
m t dlU3X UIOO^ pUB JUIOJ <N'.qoq(N
=A\3(I U33MJ3q 33U3J3JjI(I ^j 0 to lO o
duiax lU'Od-Aiaa
N. CO O 00 * M 00
Qe uj oioo *'c>-<to6
"5 lO x: CC -i t^ tN.
^^
S tu dUI3X uiooî pUB JUIOJ M, 0 t>. OC CI CO t*
-AVaQ U33/H)3q 33U3J3JJIQ cvjoitN w'coco'co
z2 00 o M w r^ CO
dm3X |UI0d=.M3Q t>.

c*i t>.' cî N.' -h" o
^ o duiax uioo^ puB juioj

<" A\3(J U33M)3q 33U3J3JJ!(] 666
U Z
IE dUI3X }U!0d-M3Q
r>;
^'
?0 CO
6 *
C^
6 -4
^ 00 o
(Xi co'
dUI3X UIOO^ pUB JUIOJ

o r AV3Q U33M)3q 33U3J3jJig
000000 00000)0
'dUISX )UI0d-M3Q rococo c^MOir^
<ot>.rs oao
71
FAN ENGINEERING B UFFALO FORGE COMPANY
Relation of Room Temperature to Outside Wet-Bulb Tempera-
ture
During cool weather the dew-point or saturation tempera-

ture at the apparatus is secured and controlled artificially at
whatever point required. During warm weather however, it
is impossible during the greater part of the time to obtain as
low a dew-point as desired without refrigeration, which in the

majority of cases of humidifying is impracticable. The lowest
saturation temperature that can be obtained with an efficient
spray system is the same as the outside wet-bulb temperature,
as has been shown; therefore the dew-point in the room will
always be the same as the outside wet-bulb temperature. Since
the difference between the dew-point and the room temperature
is dependent upon the percentage of relative humidity main-
tained, the minimum room temperature and the percentage of

humidity required in the enclosure will be as shown in table on
page 71. It will be noted that the lower the humidity carried,
the lower the dew-point must be for any given room tempera-
ture.
Dew-Point Method of Humidity Control

Any one of the three spray types of air conditioners previously
described are admirably adapted for humidity control by what

is known as the dew-point method. This system is applicable
only where the absolute moisture content of the air in the room
is unaffected to any great extent by extraneous sources of
mois-
ture supply or by moisture absorption. It depends upon sup-
plying the enclosure with conditioned air having a definite dew-
point and maintaining a predetermined relationship between
the dew-point temperature and the room temperature. The
dew-point of the air supply is determined by saturating the air
and removing free moisture at the apparatus at a definite
all
temperature. This dew-point will evidently remain constant

regardless of subsequent variations in air temperature. It
may be shown that the percentage of relative humidity in an

enclosure is dependent upon the difference between the dew-
point temperature and the room temperature, and that it is
substantially constant for any variation in room temperature
so long as the difference between the dew-point and room tem-
perature is maintained constant. (See table page 71.)
72
HUMIDIFYING
HEAT REQUIRED TO CONDITION 1000 CU. FT. OF AIR (MEASURED AT 70

DEQ. FAHR.) FROM VARIOUS ENTERING WET-BULB TEMPER-
ATURES TO VARIOUS DEW-POINT TEMPERATURES
If
It is evident that this system is particularly adapted to
thermostatic control of (a) the dew-point (saturation tempera-
ture at the apparatus) and the room temperature independently;
(b) of the dew-point with reference to a variable room tempera-
ture; or (c) of the room temperature with reference to a variable
dew-point temperature. System (a) is generally applied to air
washers and humidifiers under winter conditions, where the
outside temperature is considerably lower than the room tem-
perature and to dehumidifiers where it is possible to maintain a
definite dew-point temperature throughout the entire year.
However, during summer conditions the saturation point

and unavoidably be higher
at the apparatus will frequently
than the required minimum dew-point. Under such variable
temperature conditions it is necessary to control temperature
with reference to the dew-point according to system (c) and a
humidifier is employed to give the air complete saturation under
these conditions. A dififerential thermostat effects this control.
Automatic Humidity Control
In many industrial installations where humidifying or de-
humidifying systems are used, some means of positively and
accurately maintaining the proper temperatures and humidities
is essential. While much can be accomplished by hand regu-
lation, this would require the constant attention of a highly
skilled operator which in most instances is impracticable. In
many processes of manufacturing, as, for example, the weaving
of silk and in the conditioning of tobacco for the manufacture of
cigars, a uniformity of humidity conditions is quite as essential
as the quantity of moisture, as any variation in humidity, either
above or below a standard, reduces the output and causes lack
of uniformity in the product. In many cases a sensitive auto-
matic humidity control is as important as some means of humidi-
fying. There are three distinct methods by which such auto-
matic control can be secured:
(a) By two separate thermostats, one of which is placed at
the humidifier to control the temperature of the dew-point
by an automatically operating valve or damper, governing
a means of varying the temperature of the spray water, of
the entering air, or of both in conjunction. The other
thermostat, placed in a room where the humidity is con-
trolled, maintains a constant room temperature, either by
controlling the temperature of the air entering the room,
74
DRYING
or by controlling some source of heat within the room.
With these two temperatures maintained constant, the
percentage of humidity in the room will remain constant,
and will depend upon the difference between the dew-
point temperature maintained at the humidifier and the
temperature maintained in the room, as shown by the
table on page 71.
(b) By a differential thermostat. This type of dew-point
control is required wherever it is impracticable to maintain
either a constant dew-point or a constant room tempera-
ture. In this method there are two elements, one of
which is exposed to the dew-point temperature, while
the other is exposed to the room temperature. They are
so connected that they act conjointly upon a single ther-
mostatic valve connected with operating motors arranged
to control the dew-point temperature in relation to the
variable room temperature, or to control the room tem-
perature with respect to the variable dew-point tempera-
ture.
(c) By means of some form of differential hygrostat. This
controls the wet-bulb temperature with respect to the
dry-bulb temperature, so as to maintain a constant rela-
tive humidity without regard to the dew-point or varia-
tion in room temperature.
DRYING
The drying of materials of various kinds may be accomplished
either by means of direct radiation from some source of heat, or
by means of air currents, depending on the character of the in-
stallation or the requirements to be met. Drying by means of
air currents may be done either by means of natural circulation,
or by the use of some form of fan
either of the disk wheel or
steel plate type. This air is usually warmed, either by some
form of heating coil or by means of waste heat, the temperature
ranging say from 70 to 200 depending entirely on the nature of
the substance to be dried. In some cases this temperature is
varied at different periods of the operation. The time required
may be anywhere from a few minutes to several days. In many
cases a combination of the above systems are used
that is,
both direct radiation and air circulation.
Dryers are ordinarily built in either the room type or the
continuous (or progressive) type. In a room dryer the mate-
rial to be dried is placed in the room and left for a certain period
until drying is accomplished. In the progressive type wet

material enters at one end, and is taken from the other in a
76
FAN Ex\GINEERING BUFFALO FORGE COMPANY
dried condition. The^ entering end of the djyer is termed the
green or wet end and the leaving end the dry end. In a con-
tinuous dryer it is customary to introduce warm air at the dry
end and exhaust it at the wet end, this air being either discharged
to the atmosphere or returned to the fan to be recirculated.
The apparatus is usually so arranged that any desired propor-
tion of this return air may be recirculated, depending on the
varying atmospheric conditions. The drying apparatus itself
may take any one of several forms, depending on the material
handled, which may be spread on trays, placed in a revolving
cylinder, on a traveling conveyor, or in a room or kiln. A dry-
ing room or compartment is frequently referred to as a drying
tunnel. Except in the case of cylinder driers or continuous
conveyors it is customary, whenever practical, to load the mate-
rial on trucks or small cars in order to facilitate the filling and
emptying of the drying tunnel or kiln. The dryer should be so
designed that a clear area for the passage of air is provided
equal to 3^ to % the cross sectional area of the dryer.
The amount of moisture carried in the air is of as great im-

portance as proper temperature in many classes of drying
is
work which require either a high or a low moisture content, or

often a varying amount of moisture at different periods of the
drying for different substances. Any desired amount of mois-
ture may be obtained by passing the air through a humidifier
or a dehumidifier, depending on the conditions desired, or the
humidity of the air may frequently be controlled by recirculat-
ing varying amounts of the moist air leaving the dryer.
The temperature drop through the dryer or tunnel must be

sufficient to care for the following heat losses; (a) Radiation
from the walls, (b) Heat required to temperature of the
raise the
material being dried, including the contained moisture, as well
as the trucks or other apparatus from the room to the dryer
temperature, (c) Heat required to evaporate the moisture re-
moved by the air, which is the principal requirement. Suffi-
cient quantity of air must be supplied to act as a heat carrier
without having the temperature leaving the dryer drop too low.
The air quantity must also be sufficient to remove the desired
weight of moisture without bringing the air to saturation at the
green end of the dryer. The relative humiditj^ of the air leav-
ing the dryer is ordinarily kept below 75 per cent.
76
DRYING
The quantity of air tobe supplied by ,the fan, or frequency
of the air change in thedrying chamber, depends upon the rate
at which moisture is given up by material. This will vary with
every class of installation, in some cases as high as ^
minute
air change being used. The theoretical amount of moisture
which the air will remove is directly proportional to the differ-
ence between the wet and dry-bulb temperature of the entering
air, while the actual amount absorbed by a given quantity of
air is measured by the drop in dry-bulb temperature between
the air entering and leaving the dryer, less a slight correction
for radiation. For the same reason the higher the temperature
of the entering air (the initial moisture content remaining the
same) the greater will be the amount of moisture removed per
given quantity of air and the greater will be the economy of the
dryer.
The temperature of the air will drop approximately S}^

for each grain of moisture absorbed per cubic foot of air meas-
ured at 70, or 0.71 of a degree for each grain of moisture absorbed
per pound of air. Approximate calculations may be based on
air volume, but for exact determinations the weight of air
handled should be used, on account of it being a fixed quantity
at all temperatures. Knowing the rate of drying desired and
the amount of moisture to be removed, it is a simple matter to
determine the quantity of air required.
The following table of drying conditions is given by H.
C. Russell* and will show some of the variations required in
this work:
CONDITIONS FOR DRYING DIFFERENT MATERIALS
Material
FA^ ENGINEERING BUFFALO F.ORGE COMPANY
The table on page 79 giving the "^foisture Removing
Capacity of Air in Fan System Dryers" will be found especially
convenient for use in drying calculations, as it will serve as an
indication of the results to be obtained under any given condi-
tions. Two sets of values are given, for air entering the heater
either at 50 per cent, or 100 per cent, saturated. It will be
noted that the results given are based on the assumption that
the air leaves the dryer saturated, but as already explained, in
practice the air absorbs only about 60 to 70 per cent, of this
theoretical amount. Under these conditions about two-thirds
as much moisture will be removed per cubic foot or pound of air
as is given in the table, and the air quantity would be increased
50 per cent.
The specific heat of various substances will be found below.
DRYING
MOISTURE REMOVIMQ CAPACITY OF AIR IN FAN SYSTEM DRYERS
u
moisture content of this air will still be the same as first con-
sidered, that is, 50 grains per pound of air.
The amount of moisture which this air will remove may be

determined from the high psychrometric chart on page 37. The
first step is to find the intersection of the horizontal line through
50 grains per pound with the vertical line through 150 dry -bulb
temperature, and by following the diagonal through this point
to the saturation curve, we have a wet-bulb temperature of 81,
which is the temperature the air would assume if brought to
saturation. Following the horizontal from this point to the
left edge of the chart, it will be seen that at saturation this air
would be capable of containing 165 grains per pound, or an
increase of 115 grains. In practice it is impossible to bring the
air to saturation, the limit generally being from 65 to 75 per cent,
of the theoretical increase. Assuming that the air under con-
sideration will absorb 70 per cent, of the 115 grains indicated
above, gives us 80 grains of moisture absorbed per pound of air.
Inasmuch as a draw- through outfit will generally be used
for this class of installation, the fan will handle the air at a
temperature enough above 150 to care for the radiation loss
from the connections. The quantity of air handled will then
be based on 50 grains per pound and a temperature of at least
150, or, as found from the psychrometric chart, approximately
13.7 cu. ft. per pound. Then 80-^13.7 gives 5.85 grains ab-
sorbed per cubic foot of air. On the basis that each grain ab-
sorbed will reduce the temperature of the air 8.5, we will have
a drop in temperature through the dryer of 8.5X5.85=50.
There will be an additional drop in temperature due to the
radiation loss from the walls of the dryer, as well as due to the
heat required to raise the contents to the temperature of the
dryer.
80
SECTION IV
MECHANICAL DRAFT
There are two methods in common use for removing smoke
and gases from a boiler, by means of a chimney or natural draft,
and by means of a steam jet or a fan, commonly called mechan-
ical draft.
Mechanical draft possesses many advantages over natural

draft, in that it is independent of atmospheric conditions, and
absolute command of the draft enables the boilers to be operated
at capacities greatly in excess of that possible when depending
on natural draft. Indeed it is directly responsible for the high
rates of combustion and the increased efficiency obtained with
modern boiler plants. Heat in the escaping gases may be largely
used in economizers, since it is not required to create draft, re-
sulting in a considerable saving. It is also possible to burn a
cheaper grade of fuel when using the mechanical draft systems.
Mechanical draft produced by a fan is commonly classified
under two heads, forced draft and induced draft. Each of these
systems has its advantages and each has special features which
recommend it for different cases.
Forced Draft
In the case of forced draft, air is forced by the fan through
the maintaining a pressure in the ash pit and furnace greater
fire,
than that of the atmosphere. Forced draft is applied in two

ways: The plenum system used in steamships, where forced
draft creates a pressure in the entire stoke-hold, and the direct
system where the fan discharges directly into the ash pit
beneath the grates. Forced draft is always used with underfeed
stokers on account of the restricted area of tuyere openings.
With forced draft the blower should be run at a pressure
sufficient to overcome the resistance of the grate; the pressure
SX
FAN ENGINEERING BUFFALO ^FORGE COMPANY
losses in the tubes and breeching all being taken care of by the
stack, with an indraft above the fire of from 0.05 to 0.10 of an
inch. Hence forced draft requires a higher stack than would
an induced draft system, since its action practically ceases at
the surface of the fire. If a greater pressure should be carried,
the result would be a pressure in the fire-box greater than that
existing in the room, so there would be an out-rush of flame and
smoke when the fire-door was opened.
Induced Draft
In the induced draft system the fan is placed at the base of
the stack and handles the smoke and gases. By this means a
partial vacuum is maintained within the furnace, closely imitat-
ing the action of a chimney. Induced draft should not be ex-
pected to create an excessive vacuum through
the fuel bed itself.
In case an excessive furnace draft is maintained the loss due to
air leakage through the boiler setting is greatly increased. A
combination system of both forced and induced draft is fre-
quently used to good advantage where a considerable overload
capacity is required of the boilers. Thus it is seen that an in-
duced draft system is intended to either supply the draft ordi-
narily obtained by means of a stack, or to so increase the capacity
of a stack that variable or overload capacities may be carried.
The use of mechanical draft makes a much more flexible
system than does any system of natural draft, since the pressure
or intensity of the draft is under perfect control of the fireman.
Extreme fluctuations of load may be cared for, and peak loads
that would be impossible when depending on natural draft may
be readily carried.
Draft Requirement
The draft or pressure required for a boiler is due to the com-
bined effect of two causes, the resistance of the fuel bed and the
resistance of the boiler itself. In case the breeching and uptake
are considered, they will cause an additional pressure loss. The
amount or intensity of the draft required varies with the rate of
combustion, thickness of the fuel bed, and character of the fuel
used.
82
MECHANICAL DRAFT
Boiler and economizer losses for which draft is required
follow the usual law for frictional resistance. That other
is,
conditions remaining the same, the draft loss varies as the

square of the velocity and therefore approximately as the square
of the per cent, of rating developed. For example, an increase
of 50 per cent, in capacity requires a draft which is 2.25 times
the draft at normal rating, and for forcing to 100 per cent, over-
load would require four times the draft used at normal rating.
The following table gives some of the commonly accepted
values for draft required in the furnace to overcome the resist-
ance of the fuel bed under the different conditions stated:
FURNACE DRAFT IN INCHES OF WATER

FAN ENGINEERING BUFtFALO FORGE COMPANY
âXVAA AO SaHDNl Nl Xji^ftiO dO ÔHOd
84
MECHANICAL DRAFT
30 per cent, greater than for a steel flue. The loss of draft in an
easy right angle bend or elbow will be approximately equal to
0.05" of water. The drop in pressure through an economizer
will ordinarilyrun from ^
to 3^ inch, at rated capacity.
Draft intensity of a chimney is proportional to its height.
For a chimney 100 feet high, with flue gases at 350 above
the outside temperature, the draft intensity will average ap-
proximately 0.5'' of water.
Amount of Air Required
In estimating the amount of air to be supplied, as also the
size of fan required, for a forced draft system it is necessary to
assume both the rate of combustion and of evaporation for the
plant under consideration. These will depend on the size of
the plant and the class of equipment installed. The weight of
air actually required for the combustion of one pound of coal is
approximately 12 pounds, but owing to the fact that it is im-
possible to perfectly intermingle the air and gases rising from the
grate, more air must be provided than is theoretically required.
This will vary from 18 to 30 pounds depending on the installa-
tion, in the average case between 20 and 25 pounds of air per
pound of combustible being allowed.
It is customary practice in selecting apparatus for mechan-
ical draft purposes to allow for 100 per cent, excess air for hand
fired boilers, or 16.70 cu. ft. of air per minute at 70 per boiler
H. P. for a forced draft fan, and 32.40 cu. ft. per minute at 550
for an induced draft fan. An allowance of 50 per cent, excess
air is made where the boiler is equipped with a stoker, or 11.70
cu. ft. per minute at 70 per boiler H. P. for a forced, and 22.80
cu. ft. per minute at 550 for an induced draft fan.
Assuming 20 pounds of air at 70 per pound of coal, and 4.5
pounds coal containing 11450 B. t. u. per pound with a boiler
efficiency of 65 per cent, as equivalent to one boiler horsepower,
gives 90 pounds of air per hour, or 20 cu. ft. per minute at 70
per boiler horsepower. Then the total horsepower of the boiler
multiplied by 20 gives the required capacity of the fan with the
air at 70 F. This will be the proper size for a forced draft
system, but with induced draft where the fan handles the gases
and excess air at flue gas temperature, due allowance must be
made for the increased volume to be handled, and a larger fan
chosen. A fan for this purpose should be of special design,
with bearings protected from the heat of the flue gases handled.
85
There is a definite relation between the analysis ef the flue
gases passing from the boiler and the quantity of air required
or supplied, and, as explained on page 89, the results of the
analysis may
be used to determine the amount of air or gases
being handled by a forced or induced draft system. The
method of taking flue gas analysis by means of the Orsat appara-
tus is quite generally understood, and will be found described
in the various standard works on boiler performance.*
A pound of carbon requires for complete combustion 2.67
pounds of oxygen, or a volume of 32.64 cu. ft. at 70. Consid-
ered at 70, the gaseous product, CO2, would occupy the same
volume as did the oxygen. The volume of the carbon dioxide
(CO2), as also its proportion to the nitrogen, would be the same
after combustion as had been the proportions of oxygen and
nitrogen originally in the air used. Then the complete com-
"
bustion of carbon, with no excess of air, would give a volumetric-

flue gas analysis of
Carbon dioxide C02 = 20.91%
Nitrogen N =79.09%
If the supply of air is in excess of that required to supply the
oxygen needed, the combined volumes of the carbon dioxide
and oxygen are still the same as that of the oxygen before com-
bustion. The action of hydrogen in the coal is to increase the
apparent percentage of nitrogen in the flue gases. Thus the
sum of the CO2 and O2 from flue gas analysis will not be found
equal to the theoretical 20.91 per cent., but will approach this
amount as the amount of excess air is increased. This is shown
by the values given in the table on page 87.
Quantity of air required may be determined approximately
by means of the following formula when the ultimate analysis
of the fuel is known.
Pounds of air required per pound of fuel
= 36.56 (^ + H (36)
-4)
where C, H and O are per cents, by weight of carbon, hydrogen
and oxygen in the fuel, divided by 100.
When the proportionate part, by weight, of the carbon in the
fuel (C) is known, and also the carbon monoxide (00), carbon
Experimental Engineering by R. C. Carpenter.
Steam Boiler Economy by Wm. Kent.
MECHANICAL DRAFT
ft
(JUC
si
FAN ENGIxN BERING BUFFALO FORGE COMPANY
dioxide (CO2) and nitrogen (N) in per cent, by volume are de-
termined from the flue gases, the total amount of air supplied
may be found from the following formula,
Pounds of air supplied per pound of fuel
= 33Kco;TCo)xC + (1-A) (37)
where A represents the proportionate part, by weight, of ash

in the fuel.
Ratio of air supplied per pound of fuel to the amount theoreti-

cally required is
^^^^
N- 3.782
The heat loss in the flue gases is
H=0.24W(T-t) (39)
where H = B. t. u. lost per lb. of fuel.
W = Wt. of flue gas in lb. per lb. fuel.

T= Temperature of flue gas deg. F.
t = Temperature of air deg. F.
0.24= specific heat of flue gas.
The table on page 87 has been calculated on the basis of

burning a good grade of bituminous coal having a combustible
containing 87 per cent, carbon and 5.5 per cent, hydrogen. The
amount of CO in the flue gas has been considered a negligible
quantity, and only the CO2 and O2 used in the calculations.
This table gives the air required for different dilution coefficients
and the air quantities to be allowed for either forced or induced
draft work w^hen operating with different amount of excess air.
It also gives the per cent, of excess air corresponding to different
per cents, of excess oxygen as determined from the flue gas
analysis. This may be readily calculated from the formulae on
page 89. As explained on page 89 under air measurement, the
values in this table give a measure of the air being handled when
the flue gas analysis and also either the weight of combustible
burned or of water evaporated are known.
On page 87 the column of air quantity at 70 per H. P. would
apply to forced draft work; the column headed 300 would apply
to the average induced draft conditions where an economizer was
in use, while the column headed 550 would be used for induced
draft fans handling gases directly from the boiler. These values
88
MECHANICAL DRAFT
then indicate the increase in volume of a given weight of air at a
constant pressure for the different temperature conditions.
The tables on pages 321 to 330 give the capacity under differ-
ent conditions, both for standard and for special high efficiency
fans when used for induced draft service. The standard Plan-
oidal, Niagara Conoidal and Turbo-Conoidal fan capacity tables
in Part IV, Section III, give the required information for forced
draft work.
Measurement of Air Used

Amount by either forced or induced draft
of air being used
systems under different conditions may be determined by either
of the three following methods:
1 Pitot tube readings in the breeching or forced draft con-
nection.
2 By weighing the coal and ash and taking the flue gas
analysis.
3 Approximately by weighing the water and taking the
flue gas analysis, in case of a boiler with a good setting.
The theory and methodof using the pitot tube will be found
fully given on page 190 under "Fan Testing" and need not be
repeated here. If intelligently used this instrument has proven
itself to be a simple and accurate method of measuring air or
gases. It may be used alone or as a check on either of the other
two methods mentioned, and in the case of a forced draft system
any difference between determinations made by the pitot tube
and from the flue gas analysis would be an indication of air
leakage through the boiler setting.
Amount of air used in per cent, of that theoretically required
may be determined from the proportions of excess oxygen and
carbon dioxide as indicated by the flue gas analysis. The rela-
tion between these three factors may be expressed by the follow-
ing formulae, where 02 = excess oxygen and C02 = carbon
dioxide from the flue gas analysis, and K = per cent, of excess
air being used.
96.6j02_
(40)
20.9 -02 ^ ^
1760
K= L/U2
g^^^^ (4j)
Theoretical amount of O2 or CO2 in the flue gas for any per

cent, of excess air may be found from the following:
0.=
'"'^. (42)
0.966+ ^
100
17.60
CO2
(43)
0.966 + 3^
Since it requires approximately 12 pounds of air to burn one
pound K as derived from the formulae
of combustible, in case
shows 100 per cent, excess air, we would know at once that 24
pounds of air were being supplied for each pound of combustible
burned, or that 25 pounds of gases were being handled if an in-
duced draft system was installed. The table on page 87 gives
the pounds of air required per pound of combustible burned for
different amounts of excess oxygen in the flue gas.
Thetable just referred to also shows the cubic feet of air
and chimney gases per boiler H. P. for the different flue gas
of
analyses. Thus we see that if the H. P. developed is known
from the weight of water evaporated, the flue gas analysis gives
an indication of the amount of air used. Any determinations
of this character are subject to corrections on account of vary-
ing amounts of leakage through the boiler setting.
Mechanical Draft in Connection with Mechanical Stokers
Mechanical draft may be used in connection with boilers
fitted with automatic stokers as readily as with hand fired in-
stallations. With some forms of stokers either forced or induced
draft may be used while others are only adapted to forced draft,
or in some special instances to some form of steam jet blower.
The manufacturers of the Parsons Mechanical Stoker or-
dinarily install a steam jet blower in connection with their
apparatus, although if conditions will permit, forced draft by
means of a fan may be used. Their practice is to allow for maxi-
mum conditions a pressure in the ash pit of 2^
inches for anthra-
cite, 1}4: inches for bituminous coking coals, and 1 inch for non-
coking coals. The damper regulators are ordinarily so adjusted
as to give a maximum indraft if 0.05 of an inch over the
fire, or a
condition of minimum inflow at thedoors or corresponding

fire
parts. In special double deck boilers this may be increased to
0.15 of an inch.
90
MECHANICAL DRAFT
Forced draft is used in connection with the Jones underfeed
stoker system, sufficient pressure being maintained to force the
air required for combustion into the air chamber and practically
to the top of the fuel bed. The stack is then depended upon to
produce the necessary furnace draft and to overcome the re-
sistance through the boiler. To meet the maximum require-
ments it is customary to provide for supplying 200 cu. ft. of air
per pound of coal at a pressure of not less than two ounces in the
air chamber. For approximating the probable coal consump-
tion the company's engineers ordinarily assume five pounds of
coal per horsepower which will provide a sufficiently large factor
of safety and allow for some reserve power. This allows 16.67
cu. ft. of air per minute per boiler horsepower.
Forced draft is used in connection with the Taylor stoker,
giving a very wide range of over capacity to the boiler. The
pressure required in the tuyere chamber may vary from one to
six inches of water according to the conditions and capacity
developed. Recent tests on a well known installation of Taylor
stokers, with the boilers using an average of approximately 35
per cent, excess air, required the following pressures in inches of
water in the tuyere chamber for the corresponding per cent, of
rated boiler capacity.
TAYLOR STOKERS
Per Cent, of Rated Boiler Capacity

from each other, or from a stationary grate if it is kept clean
and free from clinker, and an interesting study of results
obtained from such an installation will be found on page 84.
This data was furnished by Joseph Harrington, Chief
Engineer of the Green Engineering Company, based on actual
tests under the conditions stated. While no one set of curves
will give the draft losses for all the various kinds of coal or all
the different types of boilers, yet a diagram such as the one
referred to is of especial value inasmuch as it gives a basis for a
comparative study of other conditions.
Niagara Conoidal Type "N" Fan Direct Connected to

Buffalo Double Vertical, DoubleÂcting Engine
92
SECTION V
EXHAUST SYSTEMS
An extensive field has been developed for the use of exhaust
fans of more or less special design, for the removal of refuse and
industrial waste in shops and factories, as well as removing the
heated or foul air, or gases, resulting from various industrial
processes. Such an exhaust system consists of the proper hood
or receptacle at the receiving end, the necessary piping to con-
nect to the exhaust fan, and if refuse is handled, some form of
dust or refuse collector. In laying out an exhaust or convey-
ing system the usual method of procedure is to determine: (1)
the number and size of branch pipes necessary to properly do
the work; (2) the design and arrangement of piping to give the
best results with the least power consumption; (3) the size and
most economical type of exhaust fan, and (4) the disposition of
refuse.
Size of Pipe
Proper is for the most part a matter
size of piping required
of experience, although practice has established standards for
the more common applications. The tables on page 94 give the
usual sizes of galvanized iron piping to attach to the hoods of
the machines indicated. For branch pipes over 25 feet long,
increase the size 10 per cent, for each additional 20 feet.
Hood Construction
It almost impossible to give standard practice in hood
is
construction, since there is such a variety of makes and sizes of
machines as to obviate the possibility of having any standard

design. Furthermore, a hood must be constructed to suit the
character of the work to be done. In designing hoods, a prin-
ciple to keep in mind is to so shape them that the refuse from
knives or wheels, due to their centrifugal action, is thrown
directly to a point where it will be caught by the highest velocity
of air. Hoods should always be made to fit as tight and close
as possible, since the suction effect is lost, resulting in poor opera-
tion, if this feature is disregarded.
03
PIPE SIZES FOR WOODWORKING MACHINES

EXHAUST SYSTEMS
For the removal of smoke or fumes it is good practice to make
the mouth of the hood extend out over the kettle or furnace at
least six inches in every direction, if the hood is not elevated
over two feet. For every additional two feet elevation, the
size of the hood should be increased six inches each way. The
area of the branch pipe should then be made one-sixteenth of the
hood mouth. For instance, a furnace 2x4 feet in size, having
the bottom of the hood four feet above it, would have a hood
4 ft. x 6 ft. and the area of the pipe should be one-sixteenth of
this, or 1.5 sq. ft. This branch should therefore be 17 inches in
diameter. The velocity at the mouth for average conditions
should be 75 to 250 feet per minute.
In some manufacturing processes it will be found necessary
to provide some means for the removal of poisonous and noxious
gases that will be more certain in its action than is the common
form of open hood or canopy. This can be accomplished by the
use of a double hood with about an inch or less of clearance be-
tween the edges of the outside and inside hoods. There should
also be an opening in the top of the inner hood, located under the
exhaust pipes. These openings should be of such a size that a
velocity of about 1000 feet per minute will be created through
the slot, around the edge and through the central opening.
Size of Main Pipe

It the common practice in blow-pipe construction to add
is
the area of the branch pipes and make the area of the main pipe
equal to their sum. This process should be continued back to
the fan, choosing a fan with an inlet equal to or greater in area
than the main pipe.
Velocity Required
The subject of the proper velocity of the air throughout the
system is an important one, and while sufficient velocity should
be provided to insure the removal of the material being handled,
any excess means an unnecessary increase in power consump-
tion. It must be borne in mind that the power required in-
creases as the cube of the speed or velocity, hence to double the
velocity will require eight times the horsepower.
In planing-mill work, it is customary to allow a velocity of
2400 for light shavings, 3000 for dry saw dust, and from 3600
to 4000 for knots, blocks, etc. This corresponds to operating
the exhaust fan at a speed to give approximately 2)^ to 5 ounces
FAN ENGINEERING BUFFALO FO RGE COMPANY
of pressure,depending upon the length of the piping and the
velocity required. The velocity in the piping should either be
uniform throughout the entire system, or else higher in the
branches than in the main pipe.
Dust Removal from Grinding and Buffing Wheels

Specifications for the design, construction and operation of
exhaust systems for grinding, polishing and buffing wheels to
comply with the Labor Law of New York State, have been pre-
pared by William Newell, of the Department of Labor, as fol-
lows: Accompanying the specifications are certain recommen-
dations by The Buffalo Forge Company on the design of these
systems.
1. Minimum size of branch pipes allowed for different sized
emery or other grinding wheels are given in the accompanying
table. In case a wheel is thicker than given in the tabulation,
or if a disc instead of a regular wheel is used, it must have a
branch pipe no smaller than is called for by its grinding surface.
GRINDING WHEELS
Diameter of Wheels
EXHAUST SYSTEMS
""^ ^'^""'^ P^P"' ^"^^d f^ different
sized ^î'^'^^'^'^vt'"
.iJri buffing, pohshmg, or rag wheels, as they
are variously^
called, are given in the table.
Buffing wheels six inches or less in
diameter used for jewelrv
work may have a three-inch branch
pipe
The thickness given for buffing wheels applies to the thickness
of the wheel at the center.
In case the wheel is thicker than
given in the tabulation, it must have
a branch pipe no smaller
than IS called for by its grinding surface.
3. Branch pipes must not be less than the
sizes specified
above throughout their entire length.
4. AH branch pipes must enter the main suction
duct at an
angle not exceeding 45, and must
incline in the direction of the
air flow at junction with main.
5. Branch pipes must not project into the

main duct.
6. All laps in piping
must be made in the direction of the
air now.
All bends, turns, or elbows,
7.
whether in the main or
branch pipes, must be made with a radius in
the throat at least
equal to ly^ times the diameter of the
pipe on which thev are
connected.
8. The inlet of the fan or exhauster shall be at least
20 per
cent, greater in area than the sum
of the areas of all the branch
pipes and such increase shall be carried
proportionately through-
out the entire length of the main suction
duct, i. e., the area of
the main at any point shall be at least
20 per cent, greater than
the combined areas of the branch pipes
entering itbetween such
point and the end or dead end of the system.
tail
For the convenience of those wishing to
use it, the table
on page 98 is given, showing what the size of
the main suction
duct should be at any point for any number
of uniform-size
branch pipes when the main duct is made 20
per cent, greater
than the combined areas of the branches
entering it, the mini-
mum required by these specifications.
9. The area of the discharge pipe from the fan shall be as
large
or larger than the area of the fan inlet
throughout its entire length.
10. The main trunk lines, both suction and discharge, shall
be provided with suitable clean-out doors
not over ten feet apart,
and the end of the main suction duct shall be blanked off with
a removable cap placed on the end.
97
PROPORTIONS OF MAIN DUCT TO ACCOMMODATE BRANCHES

EXHAUST SYSTEMS
11. Sufficient static suction head shall be
maintained in
each branch pipe within one foot of the hood to
produce a differ-
ence in level of two inches of water between the
two sides of a
U-shaped tube. Test is to be made by placing one
end of a X*
rubber tube over the small hole made in the pipe,
the other enc^S
of the tube being connected to one side
of a U-shaped watei^
gauge. Test is to be made with all branch pipes
open and un-"^
obstructed.
^
In addition to the foregoing specifications, which
pulsory, a number of
are com-
"Recommendations" are given below,

O,
which, observed, will make for
if still more efficient operation'^
and longer life of the system. ^m
Recommendations
1. Emery wheel and buffing wheel exhaust systems should^
be kept separate owing to danger of sparks from the
former Kfl
setting fire to the lint dust from the latter, if both
into the same suction main.
are drawn
" r-
C
2. In the case of undershot wheels, i. e., the top of the wheel
runs toward the operator, which is almost always the
^
of rotation of both emery and buffing wheels, the
direction
main suction "J
^
duct should be back of and below the wheels and as close
to 2P
them as is practicable; should be fastened to the ceiling or
or it
the floor below, preferably the former. If behind the
wheels,
it should be not less than six inches above
the floor at every point
c^
to avoid possible charring of the floor in case of fire in
the main
duct and also to permit sweeping under it. For similar reasons
O
C->
it should be at least six inches below
any ceiling it may run -rn
under.
q^
3.
made
Both the main suction and discharge pipes should be
and with as few bends as possible, to avoid loss
as short
o
*0
by friction. If one or the other must be of considerable length,
it is best to place the fan not far
beyond where the nearest
^
O
branch enters the large end of the main, as a long discharge Q
main is a lesser evil than a long suction main.
4. Avoid any pockets or low places in ducts where dust
might accumulate.
5. The main suction duct should be enlarged between every
branch pipe entering it, whenever space permits, and in no
case
should the main duct receive more than two branches in
a section
99
of uniform area. All enlargements in the size of the main should
be made on a taper and not by an abrupt change.
6. If there isa likelihood of a few additional wheels being
installed in the future, it is advisable to leave a space for them
between the fan and the first branch and to put in an extra
size fan. Or, a space may be left beyond the fan so that the
fan may be moved along and the main extended when it is
actually decided to install additional wheels, provided the fan
is of sufficient size to still comply with these specifications after
the additional branches are added.

7. Branch pipes should enter the main on the top or sides
never at the bottom. Two branches should never enter a
main directly opposite one another.
8. Each branch pipe should be equipped with a shut-ofT

damper or blast-gate as it is which may be closed,
also called,
if desirable, when the wheel is not in use. Not more than 25
per cent, of such blast-gates should be closed at one time; other-
wise, the air velocity in the main duct may drop too low and let
the dust accumulate on the bottom.
9. Itis very important that the lower part of the hood shall
come enough forward beneath the front of the wheel so that

far
the dust will enter the hood and not fall outside of it altogether,
even if the accomplishment of this result necessitates leavin-;
considerable space between the wheel and the lower part of the
hood in order that the hood shall not interfere with the work.
10. Branch pipes should lead out of the hood as nearly as
possible at the point where the dust will naturally be thrown
into them by the wheels. This is very important.
11. An objectionable practice sometimes found where
small work is polished is the use of a screen across the mouth of
the branch pipe where it enters the hood. Such screens are an
obstruction to the passage of material, and the ravelings from
buffing wheels are held against the screen by the suction, with
the result that in a short time the draft is almost entirely cut off.
12. The use of a trap at the junction of the hood and
branch pipe is good practice provided it is cleaned out regularly
and not allowed to fill up with dust. This will catch the heavier
particles and so take some wear off the fan. It will also serve
to catch any nuts, pieces of tripoli, etc., dropped by accident,
100
EXHAUST SYSTEMS
and in the case of work on small articles, will enable them to be
recovered when dropped in the hood.
13. All bends, turns, or elbows, whether in the main or
branch pipes, should be made with a radius in the throat of
twice the diameter of the pipe on which they are connected,
wherever space permits.
14. Elbows should be made of metal one or two gauges
heavier than the pipe on which they are connected as the wear
on them is much greater.
15. The withdrawal from a room by an exhaust sys-
of air
tem naturally tends vacuum and for this reason
to create a slight
inlets for air at least equal to the sum of the areas of the branch
pipes should be left open.
16. Recommendations for the size of the cyclone separator
or dust collector, as it isoften called, are hard to give, as the
separator must be proportioned to suit operating conditions,
light dusts requiring a larger separator than heavier dusts. A
separator should be selected with an inlet area at least as large
as the area of the discharge pipe from the fan.
For light buffing dusts, lint, etc., the air outlet from the top
of the separator should be so large that the velocity of discharge
will not exceed 300 to 480 feet per minute; then select a separator
of which the other dimensions are proportionate. The air
outlet should be provided with a proper canopy or elbow to ex-
clude the weather, but should be otherwise unobstructed.
Dust Collectors
The air with its contents of shavings and dust should be
delivered by the fan into a separator or collector. Whatever
heavy matter the air carries here settles to the bottom and is
discharged into the proper receptacle, leaving the air to escape
to the atmosphere. As usually built they depend on the centri-
fugal action to accomplish the separation. While the different
makes will vary in their dimensions, the table on page 101 will
serve to give a general idea of the sizes used. These are built
either right or left hand.
A properly designed separator should not cause a resistance
of more than one velocity head due to the flow. That is, with
a velocity of 4000 feet per minute the resistance would be one
inch.
Friction Loss
A complete discussion on the loss in pressure due to friction
of the air passing through the piping and elbows will be found
on pages 115 to 120. For perfectly smooth piping we may con-
sider that one velocity head, or a pressure corresponding to the
velocity, is lost in every 60 diameters of pipes, but for planing-
mill work it is customary to use a factor of 55 diameters. That
is, with a velocity through the pipes of 4000 feet per minute,
which corresponds to a pressure or velocity head of one inch,

there would be one inch of pressure lost in every 55 diameters.
With a pipe 18 inches in diameter, one velocity head would be
lost in each 83 feet of length. The fan must operate at a total
pressure sufficient to care for all of the losses, and still leave a
pressure corresponding to the velocity desired. For the loss
in elbows see the diagrams on pages 118 and 119.
Standard and Slow Speed Planing -Mill Exhausters
For conveying refuse from wood working machines and
carrying off factory waste of similar nature, fans with steel plate
housings and overhung blast wheels are usually employed.
These fans are of heavier construction than those used for ven-
tilating, and are built double as well as single for use where a
double fan avoids unnecessary length of piping and elbows. A
description of these fans and capacity tables will be found
in Part IV, Section III. In the ordinary planing -mill the
refuse is sufficient to furnish fuel for heating and power.
The standard type fan is designed for large capacity, rather
102
EXHAUST SYSTEMS
than wherever it is necessary to buy fuel, a
for high efficiency, so
more should be used, and proper design jnakes it
efficient fan
possible to combine the features of high efficiency and slow speed.
The most efficient type of slow speed planing-mill exhauster has
a housing which is very large and narrow in proportion to the
size of inletand outlet connections, and will reduce the power at
least 15 per cent, while operating at about two-thirds the speed
of the standard fan. Dimension and capacity tables of these
fans will be found in Part IV, Section III.
Primarily, the speed of the fan depends upon the suction or
vacuum to be maintained at the hoods. To move shavings and
saw dust, a velocity in the piping system of from 3000 to 4000
feet per minute is the average requirement, which corresponds
approximately to 13/8 to 2 inches suction in the pipe near the
hood inlets, and a velocity head in the piping of from 0.6 to 1.0
inch. In addition to maintaining this suction at the hoods,
the operating pressure at the fan must be sufficient to overcome
the friction losses of the system. Piping friction losses, plus
and discharge losses, therefore equals
collector loss, plus intake
the necessary operating pressure of the exhauster.
Examples. As an example, take a planing-mill installation
having three 7-inch branch pipes, three 6-inch branch pipes,
two 5-inch branch pipes and one 4-inch branch pipe. Assume
that the longest run of piping on the suction side of the fan is 57
feet, that there are three right angle elbows in the same (radius
of elbows \]/2 diameters), and that the fan discharges its refuse
into a collector located 60 feet from the fan, with one right angle
elbow in this pipe.
Adding the areas of branch pipes, the diameter of the main
suction pipe will be 18 inches. Referring to the data on friction
losses on page 102 the loss in 55 diameters of pipe equals one
velocity head.
57 feet of suction and 60 feet of discharge piping {\]4 foot
diameter) is equivalent to
^ =78
57-1-60 -o diameters
J.
Referring to the curve of friction loss in round elbows on

page 118 it will be seen that the loss in each of the four elbows will
be 0.17 of a velocity head. For a perfectly smooth well built
system this would mean each elbow was equivalent in friction
103
to 9.5 diameters of pipe, but the general custom to allow
it is
ten diameters to each elbow. We will

then have the four elbows
equal to 40 diameters, in addition to the 78 diameters of piping.
Allowing the customary 55 diameters as equal to one velocity
head lost we have as the loss in the piping and elbows
r-
GO
=2.15 velocity heads.
Intake and discharge loss = 1.5 velocity heads.

Loss in refuse collector = 1 velocity head.
Pressure due to the velocity = 1 velocity head.
Total operating head =5.65 velocity heads.
Assume 4000 feet velocity required, which corresponds to
a pressure of one inch or 0.5768 ounces per square inch.
The necessary operating pressure of exhauster will then be
5.65X0.5768 = 3.25 ounces.
From the dimension and capacity tables, an exhauster hav-
ing an 18-inch inlet, or the 45-inch size should be used. If a
slow speed exhauster were chosen we would find from the table
on page 345 that for 3 oz. pressure the necessary speed would be
742 R. P. M., the capacity 6620 cu. ft. per minute and the power
required 8.97 H. P. But as the pressure required is 3.25 oz.
the accompanying conditions must be calculated from the
above factors. That is,
the speed will be 742a/|^ = 770R.P. M.
^^ = 6880 cu. ft. per min.
and the power will be 8.97^// A?5\ =10.10 H. P.
If a standard exhauster is used the speed will be 1245 R.P.

M.; the capacity the same as above; and the power 12.15 H. P.
The power as stated would be maximum, that is, the amount
required when all the branch pipes are open. In pattern shops,
all of the machines are seldom used at once, which means that
handled, with resultant reduction in power.

less air is
The capacity tables on pages 343 and 346 which have been
compiled with velocity as a basis, will be found more convenient
in computing speed and powers than the above method and for
104
EXHAUST SYSTEMS
most installations will be sufficiently accurate. Assuming the
same conditions as in the preceding problem, an example of their
use follows:
Length of suction and discharge pipe, 117 feet.
Length of pipe equivalent to four elbows equals
4X10 = 40 diameters.
40 diameters X 13^ = 60 feet.
Length of pipe equivalent to collector equals one velocity
head or 55 diameters.
55 diameters X 1 3^ = 83 feet.
Total equivalent length =260 feet.
The tables are based on an assumption that the system will
contain an equivalent of 275 feet of piping, and corrections for
15 feet will be necessary. For each 10 feet difference, the speed
must be decreased one per cent., or 1.5 per cent, in this instance,
and the power three per cent., or 3X1.5 = 4.5 per cent.
From tables, pages 343 and 346, the following is obtained:

Slow speed exhauster Speed 790 less 13^% = 778 R. P. M.
Power 11.1 less 4i^% = 10.60 H. P.
Standard exhauster Speed 1295 less 13/^% = 1275 R. P. M.
Power 13.3 less 43/^% = 12.70 H. P.
105
SECTION VI
MISCELLANEOUS APPLICATIONS
Fans are used for a great variety of purposes, many of which
have special engineering features which make it impossible to
lay down any easily applied rules for their installation. Even
where a standard fan is to be used, a full knowledge of all of the
features of the case are necessary before making a selection.
One large field for the use of fans, propellors and disk wheels
is in connection with drying and cooling work, a brief discussion
of which is given in the following pages. Fans and blowers are
used for many purposes requiring air under considerable pres-
sure, such as foundry and furnace service; blast supply for
forges; for sand blast machines, pneumatic tube installations,
mine ventilation, tunnel work, in glass factories, and many other
special applications.
For forge service either the volume blowers listed in the
table on page 335 may be used, or the pressure blowers already
mentioned, depending on the conditions. For exhausting the
smoke and gases from forges a pressure of from one to two
ounces is required. The blast is usually run at three to six
ounces pressure. Piping should be properly proportioned to
allow for friction.
Special blowers and exhausters, either low or high pressure,
are built for handling gas at gas works, or for removing acid or
other chemical fumes. These latter may be made of special
acid-resisting metals. For gas works the low pressure exhausters
range in capacity from 30,000 to 1,500,000 cubic feet per hour
at a pressure up to 15 inches of water. High pressure exhausters
range in capacities from 30,000 to 3,000,000 cubic feet of air.
In connection with blast furnaces a special gas cleaning fan
is used, in which the inner surface is kept wet by sprays. The
centrifugal force throws the dust particles against the water
covered surface of the interior of the fan, so cleansing the gas.
In the case of gas producers the same form of fan, or gas scrub-
ber, is used to remove the tar from the gas.
Forge Shop Equipment

Table on page 107 gives sizes of blast and exhaust fans for
School Forge Shops and table on page 108 sizes of blast and
exhaust tile.
106
FORGE SHOP EQUIPMENT
z S
is
a
U
tr,
u
o
a *
O -a
r
w -g2
(ti S"
^
s
o S
SIZE OF BLASTAND EXHAUST TILE
FOR FORGE SHOP EQUIPMENT
Sizes of
FOUNDRY BLOWER PRACTICE
Foundry Blower Practice
The air required per ton of iron melted has been variously
given at from 30,000 to 33,000 cu. ft. per ton. As it is almost
impossible to measure the air directly it is necessary to resort to
indirect methods of chemical analysis of the escaping gases.
By analyzing a sufficient number of samples, the amount of air
used in the combustion of the coke can be determined with con-
siderable exactness. The following weights and volumes of air
are required per ton of iron in different melting ratios.
MELTING RATIO
TABLE OF AIR PER MINUTE AND H. P. FOR CUPOLA SERVICE
Static Pressure at Cupola in Oz. Per Sq. Inch
10 oz. 12 oz. 14 oz. 16 oz. 18 oz.
5690 6230 6730 7200 7640 CAP.

1423 1558 1688 1800 1910 A. P. M.
7.4 9.7 12.3 15.0 17.9 H. P.
7740 8480 9170 9800 10390 CAP.
1935 2120 2293 2450 2773 A. P. M.
10.0 13.2 16.7 20.4 25.9 H. P.
10120 11080 11970 12800 13570 CAP.
2530 2770 2993 3200 3393 A. P. M.
13.2 17.3 21.8 26.6 31.8 H. P.
12810 14030 15150 16200 17180 CAP.
3203 3508 3788 4050 4295 A. P. M.
16.7 21.9 27.6 33.7 40.2 H. P.
15810 17320 18700 20000 21210 CAP.
3953 4330 4675 5000 5303 A. P. M.
20.6 27.0 34.0 41.6 49.6 H. P.
19130 20960 22640 24200 25660 CAP.
4783 5240 5660 6050 6415 A. P. M.
24.9 32.7 41.2 50.3 60.0 H. P.
22770 24940 26940 28800 30M0 CAP.
5693 6235 6735 7200 7635 A. P. M.
29.6 38.9 49.0 59.9 71.5 H. P.
26730 29270 31620 33800 35840 CAP.
6683 7318 7905 8450 8960 A. P. M.
34.8 45.7 57.5 70.3 83.9 H. P.
30990 33950 36670 39200 41570 CAP.
7748 8488 9168 9800 10393 A. P. M.
40.3 52.9 66.7 81.5 97.3 H. P.
35580 38970 42090 45000 47720 CAP.
8895 9743 10523 11250 11930 A. P. M.
46.3 60.8 76.6 93.6 111.7 H. P.
40480 44340 47890 51200 54290 CAP.
10120 11085 11973 12800 13573 A. P. M.
52.6 69.2 87.2 106.5 127.0 H. P.
CAP. of metal melted per hour.

is lbs.
A. P. M. is cu. ft. of air required per minute.
H. P. is power required to deliver air at pressure given with steel pressure
blowers.
110
PART III
AIR DUCTS
Under the subject of "Air Ducts," will be found detailed
information pertaining to the design of various duct or conduit
systems used for the conveying of air. The data relating to
pressure losses and friction in piping and elbows is based on
actual experiments and tests, and in many cases where required
information was not to be found, special experiments were made
to obtain data for use in this hand-book. The subject of the
proper proportions of piping in different systems, as well as the
proper velocity of air, is also completely covered.
Material of Air Ducts
One of the essential parts of any heating and ventilating
system consists of theducts or conduits used to convey the air
to the desired points in the building. These ducts may go under
ground, when they are usually constructed of tile, brick, or of
concrete. When the system is an overhead one warm air is
usually carried through galvanized iron pipes or ducts, the ver-
tical risers being either of brick or galvanized iron. A very
common construction is to run galvanized iron risers inside brick
flues. In any event the inside of the duct should be made as
smooth as possible, in order to avoid excessive friction and for
this reason iron ducts are generally preferred to brick or con-
crete, unless low velocities are employed.
The piping systems for industrial buildings and those for
public buildings are designed according to two distinct methods.
In industrial buildings the problem is chiefly to convey heat
units with as great an economy of power, material and space as
possible, while in public buildings there are the additional re-
quirements of freedom from noise and prevention of drafts.
In industrial buildings air is usually conveyed through one or
more main lines extending lengthwise of the building, the areas
of such pipes decreasing as they extend, to give a uniform
distribution of air throughout. On the other hand in public
buildings individual ducts are carried from the apparatus to each
room, so that it is evident the same method is not applicable
to both systems.
Ill
!

<J
e
1
s
i
3
: 1
o
FRICTION IN AIR DUCTS
iTfTt* OOmCC 0-H50 lOÔO
_iin-i
iÔ OOÔ
05'*'*
t-005
0050
CîCOW
OSCOOJ
t^
D
OOO --i-H
ÔO OOO
OOO OOO eOlNOi
i-KNCO
O^^
Oii-t
t^QO^
(NW*
(NCO-*
fNtÔO
tpt-OJ
i-l
M<MO iCO^ COIN

CO OSCO 11 t--C<D O500CO 0(NCO t^
.-H(NiO OCiî 00<^ '-IO-* -^
t^C*30 O5O00 t>>C0O 05
888 8oo g.3 OîM
Ci O) c^
CC(Nt^
r-cco
^ CO
QOOOiO
oo<Nr-
-^
Q0tÔ
(NOCiO
OOr-^
0(N<N
CCOCO
tô
lOQO^
^ ô> ^
^ o
OOO O^rH
OOO MINCC
- ->0
qo,c OOO 00'<t(N
O-iM
(NCOI^ (NOJtj op
o
io-*o
OfOGO
T-iCOCD
OOO
iOQri>
(NrÔ
C0(N0
CC-Ô COr-iO OC50
00CO<
t-
OC
E
OOO Oî-H
<*(t^r-i
OOO (MCOfO
OOO 0'-H(N CC*! t0>i-< CO
cocooo <ô OOSTt* OiCiO

ooooo 0<Mt^ 00
OTfO OCC<N
iCCicO <(N05 00tDO
^ to t^
o
OOO
OOO ^^(N
OOO (MM-*
OOO t>-i-icO
OOOIN 0
(A
oo-.
< 00>-iCC <*0'^ CC-HOO 1>C5 00
ootot^
Q. OOO
OOO OOO (NM-*
OOO .-i.-i(M ,-iCCcC (N'ÔO
OS-" CO
U. 1.0 coo CDCO'-H CCCOt^ Oi<N00

005t^ tÔSOO <N
OkO-H I (MOO TfOCO COO-H a>
o
z
o
(NO0-* <N-hO
CCl'iO
OOO
*OCO (NMCO
0<N>0
lO
$2
00 o
B lOOO
(N-hO
lOOJCO
i-HCOt^ <N05C0
00(N
OCCO
ooo
O'Ô bt
ai -Ôt^ COiOCO
(NiOOJ
u. OOO
oor rH(N(N
OOO OOO (NiMiO
-H(NC5
UJ ^^^ (N
8
u
>
o
?
o
u
a
u
0^
u
c^
^
c/J
UJ
a<
a
oog OOO goo OOO OOO
ui Jiy |o X^po|3/\ M^^ WOS 00. "5! III
2SS
MN 'jr^'T '^'^'
113
Pressure Losses in Air Ducts
Losses in piping systems are made up of two parts, dynamic
losses and friction losses. Dynamic losses are due to changes
either in direction or velocity of the air flow, and are composed
of loss at entrance and loss in elbows and connections. The first
is the pressure required to produce velocity in the pipe, and
may vary from 1 to 1.5 times the velocity head, i. e., pressure
corresponding to velocity, depending on whether the pipe is
connected directly to the fan outlet or through a plenum cham-
ber. It is expressed as a multiple of the pressure corresponding
to the average velocity produced in the pipe. Where velocity
in the pipe is the same as at the fan outlet this may still be con-
sidered a loss, in view of the fact that with a reduction of velocity
through a gradually diverging outlet to a larger area the differ-
ence between the velocity head at the fan and the velocity head
in the pipe is largely utilized by conversion to static pressure.
The other chief source of dynamic loss is in elbows, and de-

pends on the radius of curvature of the elbow and not on its size
or on the velocity of the air. This loss may be expressed directly
in per cent, of velocity head, and, with a round five-piece elbow,
having a center line radius of one diameter, the loss will be 25 per
cent, of the average velocity head. With a five-piece elbow
having a center line radius of one and one-half diameters the loss
will be 17 per cent., or only two-thirds that of the first case.
This shows the advantage of an intelligently designed system
and the possibility in power saving, for elbows may be of so
short a radius as to cause loss of an entire velocity head in each
one.
The second source of pressure loss in the piping system is due

to friction of air against the sides of the pipe. This loss will
vary directly as the length of the pipe, or as the square of the
velocity, and inversely as the diameter of the pipe. As length
is a fixed quantity for any system, the only factors subject to
modification are the diameter and velocity, which determine

the relation between power cost and piping cost.
As in the case of heaters, it is the usual engineering practice

to proportion piping arbitrarily, either from assumed velocities
depending upon the velocity of the air at the fan outlet, or, in
better engineering practice, by determining the velocity which
114
FRICTION IN AIR DUCTS
will give an assumed resistance considered suitable and within
the fan capacity. It is the best practice to gradually decrease
the velocity in the main conduit as the latter is decreased in
size owing to partial delivery of the air through the branch
outlets.
This practice serves three useful purposes:

1. A proper proportioning of the velocity permits a uni-
form delivery of air through all the branch outlets without
dampers and regardless of distance from the fan.
2. By a gradual reduction in velocity a considerable pro-
portion of the velocity pressure is usefully converted to static
pressure, thus largely compensating for piping friction.
3. It decreases friction in the smaller piping, where it would
otherwise be greatest.
Friction in Air Ducts

Resistance to flow of air through piping varies with several
factors such as velocity, roughness of the contact surface, or
obstructions such as dampers, deflectors, and elbows. This
resistance or loss in pressure is ordinarily expressed as the
equivalent head in feet of air; or in velocity heads (that is the
ratio of the friction loss to the theoretical pressure correspond-
ing to the average velocity in the pipe); or as the equivalent
pressure in inches or ounces. As explained on page 16 under
the subject, " Relation of Velocity to Pressure," one veloc-
ity head is the pressure corresponding to the velocity of the air.
The formula used by the U. S. Navy Department for loss of

head due to friction in a round pipe is
H=4f^V2 (47)
where H = loss of head in feet of air.

L = length and D= diameter of pipe, both expressed
either in feet or in inches.
V = velocity of flow through the pipe in feet per second,
= 0.00008 for first class piping.
f
Since in air measurements it is more customary to express

velocity as feet per minute, this formula reduces to
H = 0.000000089^ V2 (48)
For rectangular ducts the formula becomes

115
H =0.000000045-'-^
n
~\^ 1
(49)
^ ^
where H=loss of head in feet of air.

V = velocity in feet per minute.
i= short side of pipe.
n = long side of pipe.
L= length of pipe.
f = assumed as 0.00008.
It will be noted that the
loss in head as given by the above
formulae expressed in feet of air, while the head or pressure
is
ordinarily considered in air measurements is expressed either

in inches of water or in ounces per square inch. As shown by
Equation (12) page 18, the velocity head expressed in feet of air
is
^2^ times the head in inches of water. Taking a value of
d =62.31 as the density of water we then have

= ^, 62.31
^ ^12W-
1096.5
Vo = ^ (Seepage 18)
where Vo = vel. in ft. per min. corresponding to one inch of

water.
W
= weight of air per cubic foot.
From the above and from equation (48)
L / V \ 2
where h'=loss of head or pressure.

V = velocity of air in feet per minute.
Vo = velocity corresponding to unit pressure.
L= length and D= diameter of pipe in feet.
^= length of pipe in diameters.
/V \2
\^J~]
= velocity head or pressure corresponding to
the velocity.
A more general form of the above equation may be expressed
as
L /V \2
cdQ
, .
(51)
where C=a constant depending on the character of the pipe.

The constant C in the above formula represents the length
of pipe in diameters causing a loss of one velocity head. For
116
FRICTION IN ELBOWS
perfectlysmooth unobstructed pipe we may take C = 60, but for
fairlysmooth work such as the piping of a planing-raill exhaust
system a safe factor would be C = 55. For heating and ventilat-
ing work where there exist more or less obstructions in the form
of dampers, deflectors, etc., and where the piping is usually
swedged, we may consider one velocity head lost in from 45 to
50 diameters. For such systems we may usually take C = 45.
For tile or brick ducts C = 40 will meet the average conditions.
From the above we see that if the velocity in a 12-inch pipe
is 2832 feet per minute (corresponding to y^ inch) and the pipe
is 45 feet long, the loss in pressure will be one velocity head, or
Y2 inch. If, for instance, a pressure of ^

inch is required at the
outlet end of the pipe, a pressure of 1 inch must be maintained
at the entering end.
Friction in rectangular pipe may be determined by using
the tables on pages 156 and 157 which give the circular equivalent
of rectangular pipes computed to give equal friction for the
same air quantities. By means of this table the corresponding
diameter of round pipe may be found, and the friction loss
determined as above.
Friction in Elbows
The two diagrams on pages 118 and 119 of pressure loss in
elbows show respectively the loss by friction through elbows of
either round or square section. These curves are based on data
obtained at the testing plant of the Buffalo Forge Company, full
details of the testsand of the results obtained being described
in a paper presented before the A. S. H. & V. E. on "Loss of
Pressure due to Elbows in the Transmission of Air through
Pipes or Ducts."* The loss in per cent, of a velocity head is
given for elbows of different radii, the center line radius being
expressed in per cent, of the pipe diameter or width.
It may be seen from these diagrams that with R = 13^D,
or an inside throat radius of one diameter, fairly good results
may be obtained without making the elbow unduly long. Prac-
tically nothing is to be gained by making R greater than 2D.
It is also evident from an inspection of these diagrams that
if it can possibly be avoided an elbow with R less than one D
should not be used. Even with these an elbow in a square duct

will cause a loss of 17.5 per cent, and in a round duct 25.5 per
*Ain. Soc. Heating and Ventilating Engr8 1913, by Frank L. Busey.
117
FRICTION IN ELBOWS
cent, of the velocity head. In case R = 3/^D (throat of elbow
square but outer side rounded to a radius of one diameter) the
loss will be 95 per cent, for a square duct and for a round duct
75 per cent, of the velocity head. The loss indicated on the
diagram for R = D represents tests on elbows of two-piece con-
struction, or with both inside and outside made square. It is
evident that this is a construction that should never be used,
since in the case of the square pipe one quarter of the loss may
be saved by making the outer side round (R = 3^D).
From the curve for round ducts it will be seen that with an
elbow having R = 1 }/^Dthe loss will be 17 per cent, of a velocity

head. If we consider one velocity head lost in 50 diameters of
straight pipe this means that this elbow is equivalent in friction
effect to 8.5 diameters of pipe. If the elbow was in a square
duct the loss would be equal to 4.5 diameters (or widths) of
pipe. With a velocity of 4000 feet per minute through the
duct (corresponding to one inch pressure, or a velocity head of
one inch) the above elbow in the round pipe would cause a loss
in pressure of 0.17 inch and in the square duct of 0.09 inch.
For ordinary calculations one easy long radius elbow in a

circular pipe may be considered as equal in friction loss to 10
diameters of straight pipe. This is the factor given by N.
S. Thompson in his book, "Mechanical Equipment of Federal
Buildings," as applied to elbows having a center line radius of
1 }/2 diameters.
Pressure Losses in Diverging and Converging Nozzles and in

Orifices
It may be stated as a general principle in air flow calcula-

tions that "The coefficient of pressure loss is the square of the
reciprocal of the coefficient of discharge," i. e., the coefficient of
pressure loss, m, in terms of the coefficient of discharge, c, may

be stated as
(52)
(i)
and the loss in static pressure will be
pe = m(^)' (53)
where V = velocity of air flow and Vo = velocity corresponding

120
PRESSURE LOSSES IN NOZZLES
to unit pressure. Then I tt-) will be the velocity pressure.
Coefficients of discharge to be used in air measurements will
be found below. Coefficients for pressure loss will be

m for sharp edged orifice 2.78
m for short length (2i^ to 3 diam.) of pipe 1.47
m for short pipe on fan outlet 1.11
The coefficient m for converging nozzles having different

angles of convergence between the two sides will be
Angle of Convergence
Diverging Nozzle in Air Ducts
A diverging nozzle is used in an air duct when the area is
increased in order to reduce the velocity, or when blowing into
an enclosed space or plenum chamber. Any change from a
higher to a lower velocity is accompanied by a conversion from
velocity to static pressure, but inasmuch as there is always
some loss in making this conversion, the total pressure is not
the same after making the reduction in velocity. That is,
there always a certain portion of this converted static pres-
is
sure lost inmaking the change, and the efficiency of conversion

is never the full 100 per cent. As will be seen from the chart
on page 123, the efficiency of conversion depends on the per cent,
of slope of the sides of the diverging nozzle, the more gradual
the slope the less the loss in pressure.
The diagram on page 123 shows the efficiency, or ratio of
actual to theoretical velocity head
obtained with diverging
nozzles of different slopes to the sides. That is, while theoreti-
cally we should obtain an increase in static pressure equal to
decrease in velocity pressure we will really convert only a part
of this decreasing velocity pressure to static pressure, depending
on the slope of the nozzle. While theoretically we should have
/ Vi\ 2 / V2\ 2
P82 - Psl - PtI - Pv2 = - \^-^j (56)
\^^J
as a matter of fact we will have
p.-p.,=e[(^)'-(^)^] (57)
where e represents a factor depending on the^ proportions of the

nozzle.
For instance, ifwe have a nozzle whose length is five times
half the difference between the diameters of the two ends, that
isa length of five times the slope or a slope of 0.20, we will have
an efficiency of conversion of 51.5 per cent. In case the side
of a nozzle makes an angle of say 30 degrees, the slope will be
0.577 and we may see from the curve that there will be no gain
from the cone outlet. From the foregoing it may be seen that
length of a diverging nozzle should be made as long as the
case will permit in order to get the greatest possible benefit
from it. The length should be at least from five to ten times
the slope, giving a slope of from 0.10 to 0.20 or an angle of ap-
proximately 6 to 12 degrees.
122
DIVERGING NOZZLES
XN3D H3d ADN3IOIJd3

123
FAN ENGINEERING BUFFALO FORGE COMPXNY
Conversion from velocity to static pressure may be deter-
mined from the formula
and loss in pressure due to the increase in area will be
p, = (l_e)[l-(^)']p. (59)
where Pb= static press, converted from vel. press.

pt= loss in total press.
pv = velocity press, at inlet of nozzle.
e = efficiency of conversion.
Ai = area of inlet to nozzle.
A2 = area of outlet from nozzle.
On page 125 be found a diagram plotted from actual tests
will
of diverging nozzles, showing the per cent, of velocity head con-
verted into static head by using nozzles having different ratios
of length to diameter of outlet, as well as outlet to inlet area.
While these tests were made with round pipe, it was found that
the same relations held for those of rectangular section. As an
example of the use of the diagram we may select a nozzle having
a length equal to twice the small or inlet diameter and the outlet
area twice the inlet area. From the diagrams we will find that
a nozzle of these dimensions will convert 50 per cent, of the
velocity head into static head.
The line marked "Curve of Maximum Efficiency" indicates
the best ratios of outlet to inlet areas for different fixed lengths
of nozzle and inlet diameter. It will be noted that for the
different ratios of there is a certain ratio of outlet to inlet

Di
area that will give maximum conversion from velocity to static
pressure.
Example. Assuming a case where a diverging nozzle is to
be used on a 24-inch pipe in order to increase the diameter and
so decrease the velocity in the pipe, and owing to the limited
space the nozzle can be made only six feet long. Then the ratio
= 3, and according to the diagram we will obtain the

Di
maximum conversion from velocity to static pressure when the
area of the enlarged pipeis made 2.3 times the area of the 24-inch
pipe, or 36.5 inches diameter. In this case 57 per cent, of the

124
DIVERGING NOZZLES
^^
'
A,
-^= RATIO OFOUTLETAREA TO INLET AREA.
DIVERGING NOZZLES.
125
change in velocity head will be converted into static pressure
at the nozzle outlet. Then
A diverging nozzle on the outlet of multivane fans is often

used to advantage, giving a reduction of the comparatively high
velocity at the outlet of this type of fan, with a corresponding
increase in static pressure.
Theoretical Outlet or Blast Area

It is a well known fact that air (or water) under pressure in
passing through an orifice in a thin plate will not deliver the
volume indicated by the actual area of the orifice, because the
resistance of the orifice, depending on its character, will tend to
restrict the flow. It follows then that there will in each instance
exist an equivalent or blast area, differing from the actual area
in proportion to the resistance. This blast area of an air convey-
ing system is the area which would be theoretically required to
deliver the same amount of air at a velocity corresponding to the
total pressure resistance offered by the system.
Every point in a fan blast system has its blast area, which,
as stated above, is less than the actual area by an amount de-
pending on the resistance at that point. The blast area of the
entiresystem may then be computed, inasmuch as the total re-
sistance is the sum of the various resistances passed in series.
In case the air passes through parallel channels the blast area
of the system is the sum of the blast area of the separate channels.
Blast area based on the total pressure drop may be deter-
mined from
= A P M ^^
Blast area .

^^^^
4005 y total press, drop in inches
Representing the blast area of an entire system by At and
the blast area of the various sections by Abi, Ab2, etc., we will
have the relation
Examples. As an illustration of the above principles, we

willassume a case where a fan blows through a short pipe into
a heater and thence into a single duct piping system. We may
determine the drop in static pressure for the various sections of
126
PROPORTIONING PIPING
the system, and calculate the blast area of each by means of the
formula
Blast area = actual area
4 '
'static loss in vel. heads -h 1
Assuming the connection between fan and heater to have an

area of 4.9 sq. ft. with a loss of entrance of 0.1 vel. head and
0.25 vel. head lost in the connection itself, we find the blast
area of this section to be
Abi=4.9-. ^.
=4.24 sq. ft.
\0.35 + 1.0
The blast area of the heater may
be found by means of the
above general formula. With 10000 A. P. M. and a pressure
drop of 0.5 inches through the heater, the blast area will be
10000
rT= = 3.53 sq. ft.
4005l/0.5
If we assume 0.5 velocity head lost by entrance to the piping
system, and two velocity heads lost in the ducts and elbows,
we will have as the blast area of this part of the system, if the
main pipe is 35 inches in diameter
fifiS^/ 1
= 3.58 sq. ft.
\2.5 + 1.0
The blast area of the entire system may then be found from
Ab = , . .. / ? xa 7-1-^. = 4.68 sq. ft.
U-24/ ^ \3.53/
"*"
(,3.58/
Proportioning the Various Losses

In addition to loss of pressure due to friction in piping and
elbows, there is a loss of static pressure due to entrance to the
piping system of from 0.1 to 0.5 of a velocity head. In addition
to this, one velocity head must be maintained to produce the
required velocity in the system. In case the piping is connected
directly to the fan outlet this loss of entrance is frequently
neglected especially if is larger than the fan outlet
the piping
and is made cone shaped. This is the case in a draw-through
system, where the fan draws the air through the heater and
blows directly into the piping or ducts. Where the system is a
blow-through one, that is, the fan blows into the heater and the
air passes through and into the ducts, a considerable loss will
occur at the entrance to the piping system, depending on the
127
character of the layout. Where the piping is connected to the
heater casing by an easy cone shaped approach, the loss of static
pressure may be only 0.2 or 0.3 of a velocity head. In case the
fan blows through the heater and into a plenum chamber, from
which the pipes radiate to different parts of the building,
the loss of entrance to these pipes may be as much as 0.5 of a
velocity head. In any case the pressure required at the fan
must be one velocity head greater in order to produce the veloc-
ity desired in the piping.
It evident from the above that loss of power due to fric-
is
tion in a draw-through than in a blow-through system.

is less
Loss of pressure through the heater would be the same in either

case, and may be determined from the table on page 446.
In the case of the draw-through system, the sum of all the
pressure losses is to be deducted from the total pressure at which
the fan is operating, while with the blow-through apparatus this
loss is to be deducted from the static pressure.
In an ordinary draw-through system it is usually considered
advisable to keep the sum of all piping losses approximately
one-third to one-half, and the loss through the heater at less than
one-half of the total pressure. The balance is then available
for producing velocity. In case a system has been laid out and
the pressure loss is found to be greater than desired, the size of
the piping may be reproportioned by means of the following
formula so as to obtain the desired pressure drop.
C2 = Cia/p^ (62)
\ pi
where Ci = present pipe area.
C2 = required pipe area.
pi = present pressure loss.
P2 = desired pressure loss.
Thus, if we have made a layout
a system which it is de-
for
sired to operate at a total pressure of notmore than one and one-
half inches, and find that at the velocity required to handle the
air through the sizes of ducts selected the piping loss will be say
one inch, we may reproportion the size of the ducts by the above
formula. Assuming that a maximum pressure loss of 0.75 inch
is to be allowed instead of the above 1.00 inch, we have
^='=^^n/I1=^-'^<^^
128
That is, the area of all the ducts in the system must be in-
creased 15 per cent., or the diameters increased 7 per cent.
Then if the heater loss is found to be 0.60 inch, and the velocity
through the main duct is to be 1800 feet per minute, correspond-
ing approximately to 0.2 inch, the total pressure required at the
fan will be 1.55 inches.
Proportioning Piping for Exhaust Systems
It is recommended in the design of piping for an exhaust
system where no dampers are provided, to make the area of the
main pipe approximately 20 per cent, greater than the sum of the
area of the branch pipes at that point. This results in greater
uniformity of distribution than where the increase in area is
not made.
Where dampers are provided, as in the vent system from
buildings, uniform velocity through the system should be main-
tained by making the area of the main duct equal the area of the
branches. Where the exhaust ducts in a public building con-
nect directly to the exhaust fan, a velocity of from 1200 to 1500
feet per minute may be allowed in both branches and main.
Where the exhaust register or opening is placed near the floor,
a velocity of from 600 to 750 feet per minute should be allowed
in the register box. Velocity of the air should be kept uniform
throughout the entire system.
For public buildings where a plenum exhaust system (vent
stacks connecting into one large chamber, such as an attic) is
used and the air discharged from this chamber by means of an
exhaust fan, it is customary to allow a velocity of 600 to 750
feet per minute through the vent stacks and remove the same
amount of air delivered by the supply fan. When the air is
discharged from the exhaust chamber by some other means
than an exhaust fan, about two-thirds the amount of air sup-
plied to the building is ordinarily taken as a measure of the air
discharged, with a velocity in the exhaust stacks of from 400 to
500 feet per minute. In either case this makes the exhaust or
vent stacks the same area as the supply risers.
For industrial buildings, a velocity of from 1500 to 2000 feet
per minute may be allowed through the exhaust system, the
velocity being made uniform in both the mains and branches.
The actual velocities assumed in any case will depend on the
best proportion between the first cost and the operating cost.
A study of the relation between these two factors will be found
on page 130.
The design of the piping system as well as the size of branch
pipes required for a refuse exhaust system will be found discussed
under "Exhaust Systems," Part II, Section V.
Proportioning Piping for Maximum Economy

The subject most economical velocity of air through
of the
piping systems has been discussed in a paper* presented before
the A. S. H. & V. E. at their 1913 annual meeting, some of the
more interesting conclusions of which will be here given,
A decrease in velocity increases the size and cost of the air
conduit, but decreases cost of power consumed in overcoming
the conduit or piping resistance. From a point of economy the
question to be determined is what relation between power cost
and conduit cost, as determined by the velocity, will give mini-
mum annual total cost.
This relationship may be shown to be
or Vm = 0.7 Vo(^)^^ (64)
Where Cpo and Cwo represent respectively cost of power to

overcome piping resistance and an annual charge for interest and
depreciation on piping designed for an assumed velocity Vo;
and Vm is the relative velocity required for maximum economy.
Comparing these relationships with those obtained for the
heater on page 414, it is evident that they are almost identical.
It will be seen in this case that for maximum economy the
annual cost of power consumed by piping resistance should be
practically one-third of the annual interest and depreciation
charges based on initial cost of piping.That is Cp =0.335 Cwfor
maximum economy. This annual allowance on the piping system
for interest and depreciation may be assumed to be about 25
per cent, of the original cost of the installation.
While these lower velocities and consequently lower resist-
ance would require the use of large fans in order to operate at
high efficiency, considering the entire installation of heater,
piping and fan, the annual cost of power should be practically
"The design of Indirect Heating Systems with Respect to Maximum
Economy of Maintenance and Operation," by Frank L. Busey and Willis H
Carrier.
130
PR0P0RTI0NINJ3 PIPING
30 per cent, of the total annual allowance for interest and de-
preciation. If this allowance is taken at 20 per cent, as an
average, we would have approximately 6 per cent, on the first
cost as the most economical yearly rate to be allowed for power.
Practical Applications
For the purpose of illustrating the application of the fore-
going principles to a system of galvanized iron piping, different
cases will be assumed and results shown. A system handling
30000 cu. ft. per minute, at a velocity of 1950 ft. per minute,
will require a pipe 53 inches in diameter, or an area of 15.32 sq.
ft. These quantities will be taken as a constant condition, but
different arrangements considered in the system of piping.
Assuming one straight duct 200 feet long and 53 inches in
diameter, delivering all of the air at the end farthest from the
fan, we have two sources of loss
will to be overcome by the fan.
dynamic loss due to the velocity of 1950 feet per
First, the
minute (or one velocity head), and second, the loss due to fric-
tion, amounting to one velocity head in each 50 diameters of
length. The pressure due to the velocity of 1950 feet per min-
ute in the pipe (one velocity) will be 0.237 inch, water gauge.
The loss of pressure due to friction will be
"T-TTT =7: =0.905 velocity head

4.42 X 50
This loss expressed in inches of water will be
0.237X0.905 = 0.214 inch
and the total loss will be the sum of these two, or 0.451 inch.
In the piping system a part of the velocity is converted to static
pressure, hence the power calculated should be based on total
pressure with a corresponding fan efficiency of 50 per cent. At
a rate of $20 per H. P. yr., the annual cost due to the piping re-
sistance will be
Cpo = 30000 X 0.000324 X 0.451 X 20 = $48.
Around galvanized iron pipe, 53 inches in diameter, would
be made of No. 18 iron, weighing 2.3 pounds per square foot,
and would contain 14.2 sq. ft. per running foot. This would
make 32.7 pounds per running foot, or a total of 6540 pounds for
the entire pipe. Allowing 25 per cent, annually for interest and
depreciation on an initial cost of say 10 cents per pound, the
yearly allowance would be 2.5 cents per pound of iron. Then we
would have as the yearly allowance for interest and depreciation
131
Cwo = 6540 X 0.025 = $163.50.
From equation (64) we may determine that for the most
economical conditions the velocity of air in the pipe should be
Vm = 0.7 X 1950 /^^)^^ = 1670 ft. per min.
Assuming the case where 30000 cu. ft. per minute is to be uni-
formly distributed by a galvanized iron pipe 200 feet long, with
equal openings every 20 feet of its length, each discharging 3000
cu. ft. per minute, we will have an example of another common
form of installation. Referring to the chart, page 138, we see
that, if the first 20 feet of pipe is 53 inches in diameter, the next
20 feet carrying 90 per cent, of the air should be 51 inches in

diameter. Treating each successive section in the same manner
we may determine the diameter and weight of each section, and
will find the total weight of the piping to be 3922 pounds.
Then the yearly total allowance for interest and depreciation
on the piping system will be
Cwo = 3922 X 0.025 = $98.05
Loss in pressure due to friction will be the same as in the
first case considered, or 0.214 inch, but loss due to velocity will
be only 40 per cent, of the loss as calculated in the first example,

or 0.095 inch. The total pressure loss will then be
0.214 + 0.095 = 0.309 inch,
and the annual power cost at $20 per H. P. yr. will be
Cpo = 30000 X 0.000324 X 0.309 X 20 = $60.
As before from equation (64) we will have as the velocity for the
most economical operation
Vm = 0.7 X 1950 (1^)^'^ = 1575 ft. per min.
Proportioning Ducts for Public Buildings

In public buildings the sizes of air-conveying ducts from fans
or heaters to vertical induction flues, and the sizes of these flues,
depend upon the velocity of flow in such ducts and flues. The
essential factors in determining these velocities are:Limitations
of economical relative speed of fans from the standpoint of power;
limitations of air velocities on account of noise or by reason of
increasing friction as velocities increase; limitations of velocity
of inflowing air through registers into rooms; desirability of as
high a velocity of air as is permissible under the limitations
referred to in order to get as quick a conveyance of the warmed
132
PORPORTIONING PIPING
air as possible; and necessary initial and intermediate velocities
to overcome the resistance existing in each particular system
or case.
Register Velocity O
The size of vertical flues to the registers in the rooms i^^
determined by the maximum velocities allowable in avoidingCI
drafts and noise in the rooms. shown that the^
Practice has
best velocities for the registers should be from 200 to 400 feetC
per minute over the face of the register, depending upon theO*
size and location; floor registers from 125 to 175 feet. Velocity ^
in the vertical flues leading to the registers should be from 400 O"
to 750. Sizes of these vertical flues are determined largely b;^s-lJ
the size of register desirable. In general, the velocity in these
risers should be low, in order to obtain as uniform a flow as pos-'oS
sible over the register area. ^r
STANDARD SIZES OF REGISTERS AND RISERS

FOR PUBLIC BUILDINGS
Cu. Ft. of
Duct Velocity
The velocity in horizontal ducts leading from the apparatus
to the vertical risers is determined chiefly by the resistance of
the duct. In practice these velocities will vary from 700 feet
to 1200 feet depending upon size and length of duct, number of
elbows, etc. A designer with considerable experience may pro-
portion these ducts so as to give very uniform distribution with-
out going into any extended calculation. However, it is desir-
able to have a correct method as a basis. For the benefit of
engineers and architects we give here the method that may be
employed in the determination of duct velocities and sizes.
Allowing for Friction
The principal losses in piping systems for public buildings
are in the horizontal ducts where velocity is highest. Losses
in these ducts depend upon velocity, size and length of duct,
and upon the number with a considerable
of elbows, together
loss in pressure as the air enters the duct. An ideal system
should take all these factors into consideration and so propor-
tion the velocities that the resistance may be practically equal
in all ducts regardless of the length, etc. The system above
mentioned accomplishes this in a practical manner and at the
same time avoids any laborious calculation. For each duct a
factor may be obtained by inspection in accordance with the
following formula:
F = 2}i + ^+^ (65)
This factor represents loss by friction in terms of velocity head.

The first term, 23^, is approximately the number of times the
velocity head is lost by entrance to the pipe, entrance to the
vertical flue, and loss in riser and register. The second factor
represents loss due to length and size of pipe; L is length in feet
and W is approximate width in inches. The third term repre-
sents that proportion of the pressure lost in elbows, and N is the
number of long radius elbows. One square elbow considered
is
equal to two long radius elbows. In checking over the piping

layout the factors for the various ducts are first found as above
and from these factors the velocity in the respective ducts is
ascertained directly. In determining these velocities it is usual
to allow a loss not exceeding one-quarter of the total fan pres-
sure, which in practice usually amounts to about one-quarter
134
of an inch. The velocity corresponding to a pressure of one-
quarter of an inch is 2000, and since the velocities vary as the
square root of the pressures, the factor F and the velocity V
will give a loss of one-quarter of an inch if
Example. As an example of the above system we will

assume a case where the longest run of pipe L = 50 feet, = 18 W
inches, and the number of easy elbows N=4
We will then have
50 4
F = 2.5 + -^ +-=- = 4.0 velocity heads
Velocity which will cause a loss of one-quarter inch will then

be
V = ??22 = 1000ft. permin.

1/4
Proportioning Piping for Industrial Buildings Supply or Blast
System
In proportioning main and branch pipes in industrial build-
ings, the primary aim is to secure as uniform a distribution as
possible without the necessity of dampering, as well as to secure
economy of power and economy of material. It has been found
good practice in proportioning piping systems to decrease the
velocity in the main pipes as the air quantity decreases. As
already stated, this principle of proportioning has three advan-
tages. First: It utilizes the velocity of the air in producing
static pressure in the system. Second: By this means a nearly
uniform static pressure may be secured in all parts of the pipe
line, giving a very uniform distribution of air throughout. Third:
It reduces the friction in the smaller pipes, which would other-
wise be excessive.
135
Equations giving ideal distribution of static pressure for round
pipes are
Vi \ 40Di
^'
+^+1
5
(67)
and ^=./^</lLL+.A
^ +
^ i (68)
D2 \ C2 \ 40Di 5
These equations for rectangular pipes are
Vi 80 a b ^ 5
\
^ ^
or when the short side a remains constant and the long side b
changes
b2^^^ 1
bi Ci^ /(a+ b)Li^_N__^^

1 (70)
\ 80 ab 5 '
where Vi = an assumed or predetermined velocity in a

length of pipe.
V2 = velocity in the preceding length of pipe.
Li= length and Di= diameter both expressed in
feet of the length of pipe having the velocity Vi.
N = number of easy long radius elbows in section Li.
Ci and C2 = the corresponding air quantities in the
pipes Di and D2.
a = short side and b =long side of a rectangular pipe.
The application of these formulaeis clearly shown by a refer-
ence to the sketch on page 135, where velocity Vi is known and

velocity V2 is to be determined. The decrease in velocity
between V2 and Vi will then be sufficient to care for the static
losses of the section having the velocity Vi.
The following method of proportioning piping has been

carefully tested and has been found to give a satisfactory dis-
tribution when applied to the reduction in size of the main duct
with a series of outlets along its length, or to branch pipes of
equal length. It also facilitates the calculation of friction.
When the branch pipes are of unequal length a correction should
be applied as explained under "Equalizing Friction for Unequal
Length" on page 139. The principle involved is to so proportion
the velocities in the various pipe sizes as to give equal friction
in all air pipes per running foot regardless of their size. It may
136
easily be shown that the equation which relates the carrying
capacity of a pipe to its size to suit this condition is
Where di and d2
^= iW two pipes and Ci
are the relative diameters of
and C2 are the relative air quantities. As an equation in this
form would be difficult of computation, the diagram shown on
page 138 may be conveniently employed.
In using this chart commence with the main pipe equal in
area to the fan outlet, or larger, as circumstances may require.
All sizes are proportioned directly from this main pipe size. It
will be noted that the curve is plotted for per cent, capacity and
for per cent, diameter according to the formula for constant
friction per foot of length. For instance, if we have a branch
pipe wWch is required to carry 50 per cent, of the capacity of
the main pipe, we find the point on the curve which corresponds
to 50 per cent, capacity and which gives a corresponding point
of 76 per cent, diameter; that is, a pipe to carry 50 per cent, of
the capacity with the same friction per foot must have 76 per
cent, of the diameter, which may be easily calculated or be read
directly from the chart for various pipe sizes. It will be seen
that straight lines are drawn for pipe sizes from 20 inches up to
80 inches in diameter. Supposing the size of the main pipe is
60 inches in diameter, then following to the line of 60-inch pipe,
we find from the scale above a diameter of 46 inches, which is
the size of pipe which has half the capacity of a 60-inch pipe
with the same friction per foot. By this method the sizes may
be read off rapidly without any intermediate calculation.
Example. Let the main pipe from the fan be 48 inches in
diameter in the form of a straight duct having ten equal outlets.
The first section of piping is 48 inches, the second section has a
capacity of 90 per cent., the third section 80 per cent., the fourth
70 per cent, and so on. Corresponding to 90 per cent, we find a
diameter of 96 per cent, which for a 48-inch pipe gives us 46
inches for the second section. For the third section we have
80 per cent, capacity corresponding to 91 per cent, diameter, or
again following from left to right to the 48-inch line, we find a
diameter of approximately 44 inches. For the fourth section
we have 70 per cent, capacity with a corresponding pipe size of
863/^ per cent, of the main pipe and a diameter of between 41
137
138
'
PROPORTIOiNING PIPING
inches and 42 inches determined as before. For the last section
we have 10 per cent, capacity or 40 per cent, diameter, which
gives a diameter of between 19 inches and 20 inches.
The sizes of the outlets are not calculated by means of this

diagram, but are proportioned so as to give the desired velocity ta^
the air leaving the outlets. This velocity will be determined hy^
the size of the room and consequent distance the air Ls to be-*
carried, and by the required freedom from drafts due to high S
velocity of the air leaving the outlets. Outlet velocities vary- r^
ing from 500 to 1400 feet per minute are ordinarily used, depend-
ing on the circumstances. The most commonly used velocity'cx
at the outlet is about 1000 feet per minute.
2
As already stated, the above system of proportioning pipin g 7l
applies to the reduction in size of the main pipe where a series
of outlets are taken off, or to branch pipes of equal length.
qJJ
When these branch pipes are of unequal length, a correction
should be applied as explained in the next paragraph under ^^
"Equalizing Friction for Unequal Length." .5
Equalizing Friction for Unequal Length O
The above system of proportioning piping refers to cases
where outlets or branch pipes are of equal length. In case one*
^

or more of these branch pipes are of unequal length the shorter îv
pipes will tend to discharge more air than intended. It will
then be necessary to so design the various branches that the
frictional resistance in each will be equal, or adopt the common^ZS
practice of placing a damper in each pipe and partly closing it
in the pipes which deliver too much air. C-^
Q
The better way of equalizing the friction through a system .__
having runs of unequal length is so to proportion the different
qJ
runs that the resistance of each is the same. This may be t,^
accomplished either by using a smaller pipe and higher velocity O
*0
in the short pipes, or by making the long run of greater diameter
with a corresponding lower velocity and pressure loss.
The change in diameter required to accomplish this equaliza-
^
O
tion of friction loss due to unequal lengths of piping may be Qu
computed by means of the following formula,
di^ ^ ^/ rmsTi ,-2)

di \l2+45d2
where di and d2 are the diameters, and h and I2 the lengths of
the two runs of piping, both expressed in feet, as originally laid
out, and di' is the diameter di, corrected so as to equalize friction
139
140
between the two branches. The regular method of propor-
tioning piping as described on pages 135 to 139 results in
equal friction per foot of length, but formula (72) gives equal
friction, and therefore the desired distribution, for different
lengths of piping.
This formula may be readily solved by means of the curve
on page 140, which gives the fifth root of the various ratios
of (li + 45 di) to vl2 + 45 d2). This ratio, should be applied after
the piping has been laid out for equal friction per running foot
according to the method explained in the preceding section.
Example. Assuming a piping system has been proportioned

for equal friction per foot of length according to the method
explained in the example on page 137, and that there area num-
ber of branch pipes ten feet long and one branch pipe to a distant
room is 50 feet long. To carry the desired amount of air, accord-
ing to the method already referred to, we will assume the short
pipes are to be 15 inches and the long pipe 12 inches in diameter.
While friction in this long run will be the same per lineal foot,
owing to the fact that it is longer than the other branches, total
friction will be greater and air delivery will be less than desired.
Letting li = 10 and di = 1.25 as the length and diameter of
the short branch, and l2 = 50 and d2 = 1 as the length and diam-
eter of the long branch, we may determine the corrected diam-
eter di' by means of the factor obtained from the curve on page
140. We will have
li + 45di _ 10 + 56.3
l2 + 45d2 50 + 45
and from the curve (page 140) the corresponding ratio
^
di
=0.93 or :^'=^
d2 0.93
= 1.075
In this case diameter of the longer branch should be increased to
d2' = 12 X 1.075 = 12.9 inches.
Piping Layout
Values in the tables on pages 144 to 152 are taken from the
diagram and for rapid work may be found more convenient than
the curves. They give directly the diameter of the branch pipe
required to carry, with equal friction, any given percentage of
the air carried in the main pipe whose diameter will be found
across the top of the table.
141
In reducing from one pipe size to another the taper should be
13^ inches to the foot until the area is reduced to the size re-
quired.
The branches should leave the main at any angle less than
45, preferably 30, but it is not necessary to adhere to this,
rigidly.
Elbows of 90 should be made with a radius of 1 )^ diameters
to the center of the pipe. In mains having high velocity two
diameters is a better radius.
Outlets which discharge directly from the main or branches,
as is often the case in industrial buildings, should be made about
two diameters in length.
By the foregoing method of proportioning piping, it becomes
unnecessary to consider the resistance of each section of pipe
independently as the friction is constant per foot of length. It
is simply necessary to know the length of the longest run of pip-
ing in feet, number and sizes of elbows, and diameter and veloc-
ity in the largest pipe, as the loss is exactly the same as though
the entire amount of air was carried through the largest pipe the
entire distance. It is usual to make the area of the largest pipe
approximately equal to the area of the fan outlet.
Example . As an example of this method of figuring, assume
120 feet as the length of piping to the farthest outlet with a
main pipe of 48 inches diameter and with three reductions of
39, 30 and 20 inches diameter, each containing one 90 elbow.
We may then compute the friction in the following manner:
The main pipe is 48 inches or 4 feet in diameter.
120 feet is equivalent to i|^ or 30 diameters of 48-inch pipe.
1-48-inch elbow is equivalent to 10 diameters of 48-inch pipe.
1-39-inch elbow is equivalent to 10 diameters of 39-inch pipe or
11 X 10 = 8. 13 diameters of 48-inch pipe.
If X 10 = 6.25 diameters of 48-inch pipe.
11X10 = 4.17 diameters of 48-inch pipe.
Then the total equivalent length will be 30 -1-10 -t- 8. 13 -I- 6.25 -I-
4.17 = 58.55 diameters of 48-inch pipe.
The equivalent loss in velocity head will then be
58.55^50 = 1.17
times the velocity head in the 48-inch main. Further, there is
a velocity head remaining in the 20-inch pipe which gives an
142
additional loss evidently of ^%8 of one velocity head or 0.42
times the velocity head in the 48-inch main. This gives a total
loss in the piping system, neglecting the loss of entrance of
1.17 + 0.42 = 1.59
times the velocity head in the 48-inch main. If we allow a
velocity in the 48-inch main of 2000 feet per minute the corres-
ponding velocity head will be 0.25 inch. The loss in pressure
in the piping system is
0.25X1.59 = 0.398 inch.
Carrying Capacity of Pipes
Carrying capacity of round ducts at various velocities may
be found from the tables on pages 154 and 155. Capacity of rec-
tangular ducts may be determined from the table of equivalent
sizes on pages 156 and 157. Thus if we are to handle 20000 A. P. M.
at a velocity of 1800 feet per minute a round pipe 46 inches in
diameter should be used. In case a rectangular duct is to be
used the size may be found by selecting from the table on page
157 the proper sizes to correspond to the dimension 46 inches in
the body of the table. Thus we might use a 35X50, a 38x46,
a 42x42, or any one of a number of other combinations.
Three-Quarter Housing Planoidal Type "L" Fan
143
. I
COt^000500'-i<N(N(>4COM-<i<-<t-îCiOCOtCit^l^t^l>-OOOCOOOOOi050S
<t^000iOO^^(NC<l(MC0C<3T(<-*Tt<iOOOOir>CDCOt^t^r^t^0C000000
Ot>'G0000500'-HrH(N(N(NMCCCC-<*<-*Ou:)0OlOCC<CCD<Ct^l>t^tÔOOO
CO t^ OC OJ 05 o o KNWMCOCOCC-^rti-îOiOiOiCiOcDCOCOCCiCt^t^l^
<Ot>OOOS050000-HC<|C^(N(NCOW*-*-<f-<f<tOlClOIOOCDiCCOCCCO
tOt^00000505000'-^'^<-i(NC^<NCOC<5MCCW*Tj<TfTtiiCiOiCOiOCCC
cOt^tÔOOOOSCiOOO^^' i<N(N(NCCCCCO(rCCCTj<Tt' < * 0 iCi lO lO
<ICt^0C000J05OOOOl^^(N(^^(M(N<^^COCOCCCCM<t<J^<*^<!t<T}^<J^
tOtt*OOOOOOOSOOSOOOO^'-i^<M(N(N(N(N(NCCCCCCCCCOeC't'1<
OOt't^WOOOOOSOOSOOOOrHr^^-H.-KNWCîMfNC^lCCCOCCMW
C0il>t^l^000000050S0S05OOOOOr ' (N CJ (N W (N (N CO
tf)Dr^t^t>-OOOOCOOOOSOS05050500000'
CCwb-t^t^t^QCX00000505050505OOOOOOO
cocoi>tîîîôoooooooojaia>050>0500oooooo
cD50(Ct^r^t^t't*X'OOOC00000505050iOiO>0500COO
<occcocot^t*t^r^t^rôooooco5050JOiOi0505050>050so>
CDOCOi50t^t^t^t^t^t^tÔOOOOOOOOCOO00000>50S
iCOO<OCOOt^t^l>t^t>.t^t^t*00000000M00!00
!V |o %
144
DIAMETER OF BRANCH PIPES
00OOOOC3iOHÔSOC^JC5OCCI>X0500)CO"*'<**0CtÔOOiOC
<J<05
: rf IC -^ r- GC CO '
1<X^
<. (Ncccc'^'^'Oiccci^rôo
r'cnivot>.oeo'0 (src-riv
jfVJO%
146

(M C-J C-l CJ I^ !M CJ IM C
C^l(N M eC fO M CC M
(N CM 1^1 (>4 (M C^l (N C^ C
^ ^ ^ ^ ^ ^) Cv^ Cvj /;
M ^^ M (N M c^
(>) cî -:
I C^J (N C^ IN <M C
I (N M M (M (N C
Ci 35 C5 05 OS 03
I C^ "M C-l C^ <N C
CC'OCXCJCJSOSOSOiC
OOXOCQOOCXXOOOOidCiOC
I M (M !N (M 04 C
ii;t^t>.t>.1^0000XXXC(000O5C5OSO>O5a)C
CCOcD'.C:Ct^tÎÎ^t^XXXXXXXaiC5 0--OJOSOi050COOOOC^'H
ioocc'-c-^'x;-îu;r^t^t^t^t^t^xxxxxxxxxo>cs05ai05050oco
ic-oiccioioxw-wcocc'-c'^t^r^r-t^t^^-t^xxxxxxxxxososooi
j"^'i"îoiominicoic>o-X'0--o-^cccDxt^r>'r^t^t^t^t^t^t^xxxorx
cc?0'r'^'^'^'i"t"*"Ooiooioooo;ci:ccccccDcccccct--t^t^t^t^t^r^t^
C0C0MCCCCC5TfT}(TfTji'>*TtTjiTfHCiOOiO>OO>CiCO<CCCCO5CC',C<CiOI>-
ClC^lC^<M'NMMC0f0C0C0MM-^'<*"t'^'<t^Tl<'<tOOiCmnOOiOtOOcO!C
^(NCJC^C^(N(N(NM(NCOeCCOWC0WC0e<5MCC-*-*->tTji-*'l<Tj<'
iMiMC^<N(NC^liN^)iN(MCOC<5COeCC0C0COWeCCC'
ooooo- i(MM(M!M(NM(MM(NiM(NCOC5C5MCOM
,0000000000. I
M W CI <N IM (N (N (M IN (N M
Oi0>a30>a>a>o>ooooooooooooo<
oooso>o>ojo>ooa>oc50a>osoooooooooooooooo
IV JO %
147
I
"^ ^^^
'OOO--i^C'lC^lCCCO'*^
|(N'MCC CCCCfiCiCiCi-OCC:
Icccccc
: ^: 'T '+ Tt- lo I
! cc fc ^: re cc cc (
'Tl M 'M CC C! * Tf T} rf I
: CC CC CC CC CO CCI
J!V io %
148
IiOOCDtÔC00050-^^MCCM'^C>0Ot>.OOOOOSOSOO^C^4C')C5M'*'*0
l-fiOfOt^xxoscso^ (Mccfc-îoiocixr-xxocjco :-)Mcc'r'<r

jx>cocOwco;ct>.tt^t>-t^t^t^t^t--t^t^t^t^t^t^t>'Xxxxxxxxx
I'1<>00:Ot^t>.XOSOO'^-HC^MM-^'îOiOI^r>.XOSOJOOCCC^lC>)
-^-J5-.0:0C0^<Ct^t^t^t^t^t^l^t^t^t^t^t^t^t^t^t-I^XXXXXXX
|rC'îO'0---X;t^Xi030^ <MMW*-<*"00tC?Ct^XX0SOOC^'^lNN
|-^-.c-.c--c-^-^y;->c--ct^tî^t^t^t^t^t^t^r^t^t^t^t>.t^t^t^xxxaoo
|^JW<*0OXt>-t^XO>C500 C^l!MW^'^'OiO:Ct-t>-r>.X00500<-<W
|:cx!;c-^'.c.c-^'^X'^-.ct^t>.r>.t^r^r^t>.rî^t^t^t^t^t^t^t^t^t>.xxxoo
|c^MM'<t'<l<iO-^t^t^XC>C500'-iMC^CCeCiO00;?CcCt^XXCiOSOO.
I-HMCOW^-<<<OCCt^r^XXOSOO^'-iC^lC<3'*'^Tt<0OCCtCt^t>-XX03050
IO^MCCCCi-*iOLO-i'.Ct^XXC;00-^^0)f<3TCCO'<tOiOCCCC!CXXXOJOS
lo-H -^
c>^^^^:T}<lOlO.t^^^xxc-05CC^c^c>lcceott^ClC;c^t^^^xx
I
'.c -sC -.o X -^ -c c; -c :c :c w -c cc -^ t^ t^ (^ i^ t^ i^ t^ t- 1^ r^ t^ r^ t^ t^ t^ r^ t^
-.c
lOsO'- MMCO-^'OiO'^:CtÎ^XCiC5CO'-< C^lfCCOW^îOiCCOOt^t^

losOO -<MfCOC-*-*'00-^t^t^XXCSOOO^?)C^'Me<0'^'*>0<fl;CtOC
I
o lO
-^ -^ -^
-x; -.c -.c X X o ; ; ; X
IS t^ t^ t^ t^ r^ t^ t^ t^ t^ t^ t^ t^ t^ t^
-X: ->c -.c
1x050500 C')ccec-*-îo-':cccct^xxo:ooO'-<'^<NcîeoeO'^-*c"5cc
X X cc cc X t^ t^ t^ t^ t- 1^ t^ r^ t^ t^ i- i- t>- 1*
I
lO lO ic -c xXXXXXXxxX
|t^X050500-<'MMCC1<'1<0OXXt^XXCiC5 OlCÎCCeC'^'^'O
I
iC lO lO iC X ' X X X X X X '-C X X CO X X X X X t^ t^ t- 1^ t^ t^ t^
1> t>- 1>- t> t>-
|t^t^XOJO>00-<iNINMCC'*-<t>OCCwt^t^XX05000 CJC^COW*"^
|iOOiO0OXCC?CCCXXXXX?CtCXCCCCXXOt>tt>-l~-t^t>-t^t^l^t>-t^
lot^t-XXOJOO WINeCfC-^-îCi-TXXXXXOOOO C^WMCC
lO c o lO lO o X X X X o X X
I
X X x X X X X X X t^ t 1^ r^ t^
--C --c -.c t>- 1-- 1>-
IiOCCt^t^XXOOO-^C-lMC^lfC-^-îCiOXr^t^t^XXOOOOO-* MM
iO0OiOiO0'OXOXXXXXXXXCCXXXXCCCXCCt>-t^t*t^t^t>-t<-
lioioxt^r^xxooo -^MMccccîOioxccr^t^t^xxososco M
i|ioioiocoo":)ioxx'-cxxccxxi:cccxc;xccxccxx!Ct^t>-r^rr^
|Tj<io0'>cxt^xxoi0500 Mccfcfîoioxxt^r-t^xxoaoo.
iiOiC0iCiCO'0i0OiOXcCXXtCtC<C!C50XC0CCXXCCCCXCXXr>'t^t^
i|w*ioioxxr>-xxOiOsoO'---McicofC'*'<t'Oioxccr^xxxxososo
i|ir5i0ioicio0'0'ni0io0'-ccc'-cxccxxxxxxxccxxccxccx!ccci^
IMW^'OiOX'-Ct^XXOOCO'- CIMfCCC'J'-îOiOiOXr^t^t^XXOSW
lM'CM1'^'-'5!CCCt>.N.XX05C>CO MMCCCC^'^'^'CXXCCN.t^XOO
|iCiO>0OiOiO'OkO'00OiCiOOXXCCCC5CCOXCO<CCC5CCXCCC0<C
l-'MCOMTTTfiCXCCt^t^XXOOSCO ^^WCCW*^CÔ<CCDr^^
|lOÔ^ClCÔÔÔ^ClCÔOlC^CCÔCOCOCCtf>CO?C0:!C<ClC5C;cCCC0
iMMM-^'îOiOCCtDt^XXOSCTiOOO --MMC0C2M-*-*inCJCCCD
|iC"500'0000iC0000iCi0CCCOXXXCCXXXXXXC^CCX
I
I
O - M M M -^ 't lO 0 X t^ X X O C: O O O O -- - M M M M >* -* lO lO JC
|oO--i--MM'*-*Ou:iXt^t^t^XXX<5OO-;.-'-MMW0'*;*;*jC
I050SO-'--^'>'MMC5^'4<CXXt-t^XXXO'-Oi020C '-^'IC'lCCfC'^^
ic 'C-C'^, c^ "~
^ ^ lO lo o >flo lo lO lo lO ico ic o o lO o lO lo '-': "-C X *-.
'lx0J05OO---MMC<3'!f't>0i0X'X;r^t>>t>'XXX0505CO <-" ^^'^^wcc

,|.^^^,/3,/3,/5U500>COiC0000C000OiCCkCXCCCCi:CX!CX!^5C
ilxxo>0500-''-MMMW<f'*"0'0'-oct^r^xxqooooo-;'-M2i
tciooioococoxccxx<c
,
|!^:^^!;^;^sjiotooiQtoicto>ctofltoioiQiotQ
i|:DI^XXX05000 MMMCCM'*^0CiOCtCXt^tXXXOOiOO<
ilxxr^r-xxooco Mcocc-*'^'00'flcr-t>-xxxoaQ
149
l(N(NiM(N(N(
I C^ M (M (N (M (N (N (
I C^ (N(N CM <N (M <N (N (
coooooooos0505oo5a>05C5(
I C^l (M (N <N (N
t^ t^ 00 00 00 00 GO 00 00 OC X X 05 C5 C5 Ci O Ci C; C- Ci C5 C5 <
t^t^t^t^t^t^t^t^t>.tôcxxxxxxxxxxxxo5ai0500iC". 05050503
coo'X)-Xi->cx:ococot^t^r^t^t^t^tî^t^t^tî^t^xxxxxxxxxoc
ioio>oicioooiO;ccocci)?oocccccccccccc^t^t^tî^t^t^r^t^t^r>-t^t^
Tf^T^T}<-<^'<tU5iOOiOiO00OiOiC0CiC>COiC<C5CCCCCCCC?CCCCCCCCC
CCeCCCf0MMT}(Tj<TtTtTj<TjtTj<Tj<Tt<-*-<J<T}<T}<Tf<l<ifliCOiOiCiCiCCiOiOOC
MC^CêOCCCOCOCCMCCeCMCOCCCCCCCOCC^rfrf'^TfTj^TjiTjiTj'TtTfrJiTfrtrf
i<NiNiN(NiNiN(N(NM(N(NiN(NM(NeOMWCCeOCOMCCeCOCCCMe<5CCCOfO
(sl. M C^l IM (M (N (N IN (N <N C) M (M (N (N (M W
!v J" lo |vo>o>ot>it^t>i>.tt>ir>it>.r>.i>.ooooQoaoQOooaoooooooo>>a>ovo*Ovo^>
150
rN|tôoooosos050JOOO'H,-(^(N(N(Necoorow'<*<'*-ôooc(C(ccDt^t^t*
i-^^^fNMNN
*iyV' io >0'0>orNr>.t>.t>rNt^t>.t>rNr>iaeQOaeacooooaeo.>>a>o>o>'0>
151
ICCC0CCTj<r)<iCOCDCb-l^C0000J0JOO
)00O5O5OSO3<
>!V 10 % " wfl * U5 >o rN 00 >(
162
LAYOUT OF PIPING
153
CARRYING CAPACITY OF PIPES

This table specifies the diameters of pipes required for the
passage of stated volumes of air at given velocities. The column,
"Cubic feet of air per minute,'* indicates various quantities of
air to be moved per minute. The figures at top of table give the
velocities in feet per minute at which the air is to be moved, and
the figures in the body of the table state the required diameters
of pipes for the passage of the volumes mentioned at the given
velocities.
^-i


-r ci-':^ i^ re rj c r. c s; cc Ci :c c CJ x c-i <;
-^' t>I
X si c ?i re -T -c
' x x' c; d c ci ci t-i r-i '
5: 5s XX c re X c ir X "T q
I-; "t c :c ri x ^ r-^
c; t^ x' si c ~i re ce t -^ x x si c c ri
r4~jM:)xrefer;?;:erererer:r:r:r:T^'r'r'^
i-e r-^
c^ ?< re ce T-i c X o "T q t-; q c< ^qmx

q c; r x '' j 'i; <*<
o ' t>^ x' ss

!'
d d ci re' :e' -r id d t^ x o d C
i-': r-^
?c t^ t^ q i: ^ ri c X q ~i q X re c q r: q -T -^ t- ?i x ce
't^xx'ddd
cqqX ce q X 1; X -^ ce q ^ -^
t>; -e ?> ?i -i; t>; t>- t>.
^ d d t^ x d d d ri r ce -r t
ri ce -r d t^t^v.'X. o i i-e i-e
; -r ce c^ q t>. i^ ?j q t-. ri q i-e -^ X -T q -r q q

T Le d d r^ X i>^
q X X q -?
X -^ 5i X ce q :s ~J q T q -r q T C; T q
c^j q t>. 'T
d ci
d ri
d 71 ri
^ -r
n ci
' d d x' d d d c ce ce -r t'
J ce' c' t-^ dd
ri ri ri c^ n r< n re ce ce ce ce ce ce ce ce ce ce ce ce
' ' cî i-e i-e
t-rixcecericsii^'"?<xcr)x-e i-ecexrexcexrexcex
rj 1^ ?) i^ -c I
-c ce c r>. ce ( :c ce t^ CI t^ T* t^ ~j r^ CI t^
l:ii2^;i?!?i?3;^^5^f^^^?i^?;gS;:;:îHH;^^
a :.
Q^
CIRCULAR EQUIVALENTS OF RECTANGULAR DUCTS
00
Sizes and Weights of Piping
Conduits through which air is conveyed may be either round
or rectangular, depending on conditions to be met, and the terms
piping or ducts are used for either cross section. The accom-
panying table gives the gauge of iron generally used for galvan-
ized iron piping in heating and ventilating work, and also the
weight per lineal foot of such pipes. For method of proportion-
ing piping see page 135.
WEIGHT PER LINEAL FOOT AND GAUGES FOR GALVANIZED IRON PIPES
ORDINARILY USED IN HEATING AND VENTILATING
i
SIZES AND WEIGHTS OF PIPING
The figures to the right of theheavy line indicate the gauge
of iron ordinarily used in heating and ventilating work. Thus
a pipe 20 to 29 inches in diameter would be made of 24 gauge
and a pipe from 30 to 39 inches of 22 gauge iron. The dotted
line indicates the gauge to be used for planing-mill or other ex-
haust work. Other gauges than those indicated may be used
under special circumstances, but those shown as above repre-
sent good average practice. The small table on page 158 gives
the same information regarding weights of iron in tabular forms
as explained above.
In the table on page 162 will be found the weights of black
steel pipe of various diameters and gauges in pounds per lineal
foot, and also the material used in square feet per running foot
for the different sizes. The tables on pages 163 to 167 give
the weight in pounds per lineal foot of rectangular galvanized
iron ducts. These tables also show the gauge of iron ordinarily
used for the different sizes of ducts. Thus a pipe say 20X50
inches would be made of No. 22 gauge iron and would weigh
17.5 pounds per lineal foot, while for a pipe 20X70 inches No. 20
gauge would be used, and the weight would be 26.3 pounds per
running foot.
Niagara Conoidal Type " N " Fan
159
WEIGHT OF GALVANIZED IRON PIPES POUNDS PER LINEAL FOOT
Diam.
WEIGHT OF GALVANIZED IRON PIPES
WEIGHT OF GALVANIZED IRON PIPES POUNDS PER LINEAL FOOT
Diam.
of
Pipe
WEIGHT OF BLACK STEEL PIPES POUNDS PER LINEAL FOOT
WEIGHT OF GALVANIZED IRON DUCTS
t*Tj< t^Tjl -^t*
o^coiocDoqpcocco
^23S!?5^050^t^C^H005C^COO
oqoMTj.cqx-;ioo5Tj<QoouV.-2ioo
r^ X QO 00 oo' 00 05 oJ OS d d ^ î
-<'
cî eo
C<3'*-1Ttt'>Ji-*T*iOidd
ooqq(NTi;oqco?:oiSoifla.coS
t^ r^ 00 GO oo' 00 oo' d d d d ^' ^' -J cvj cj
CO ec ^ ^ ^' Tji Ttt o lO lo

J^S^S3S!5'^'^'*'^^'^^'fl05CJcoo
C0iOQCqC^Tj;-^îOO5-*Xct)t>.i-io
O :C GO O <
CO i-O 00 ^ 0 t^ r^ r- 00 00 00 00 d d d d d ^' -. ci w
CO CD CO '*' <1< Tf< Tt" Tt OO
-Heo.ooqqcvji;ooeot^(N'5oioosJ2
COeOCOCC^'TTjtTjdOi.'J t^ t>. t^ t^ 00 00 00 00 dddd '
-.' -4 1 :
u -jcoiccqcoO'îooq-^
0^
' cc ec co' co' co' *' *' >*' -^ lo
z
o
O CO iC -C X O CO :c O
CO co' co' co' co' -*'
r^q^-eoiocoq-ôocorîMcooict
N o co' >*" Tji"
LO o CO r^ r>.' N.' r>.' oo oc x d d d d Ji "
iot-q-.coiOX(M'.c-ô^x?ih.
O (A
c^ CO CO CO CO co' co" t' Tt <^'
5C -X) CO t^ r^ i^ t>.' 00 00 ddddd ' -h"
!i S! ^ '^ "^ <="^

oooo.-<cooscpcoco
"^t^
O <n ci CO -^ *" q ic q ^ CO lo q
r>. cc t^ oj 'S 5 lo Tj< OC'
rJ x x x d d c d
<n' co' CO co' Tji'
cc -o
o '^ t>^ r>.' ' -.'
u.
<
u "scqoop-
.-. . lO OS -t X (N
z c^' c4 w CO co'
_
c6 co' CO *' . . .
d(coco(t^b-t^x x'ddddd^'
.
.
CJ M M (N CO CO CO CO *' -*'
xqc^jiflt-q-îoo-^xcohlwSo
locococoddt^t^'xxoo'dddd-^"
^_ CO 0 -^ 00 -^ O O
QC <
qxqMiot^qcooq(Nq--ioq-ôc
c^' M
(N fi i>i CO CO CO CO -"t
o d d d
"3 u5 !>' b-' 00 x' ddddd
.--.coiopoqococcc
M cs ci m' csi c^' CO CO CO ^'
o>
""
I
<i;
q X q c> lo r>. -< lo o "* X 00 P- iM ^
>oiocddddt>.'t>:x' 00 x'dddd
K' (S(SN(SNf<S
163
N ea
> es
SI
g
o
0>00 U500 O
*
_
ocodod
^rH^ ocoios
^^^ oioic)
r^^(N
ddo
(NCM(N
-i^'w
(N(N<N
cic6e6
(N CM Cs(
*
W
(NiOr-; q<NiO t>;q(N Ct>;q (NiOt>. 0(NiO I>Wb: Sb-M t2
cooco
^^_ i^t^t>^
,-<,-(^
t>.'o6o6
,-(^rH
ooodoi
^rH,H
oJoioi
.-( ^ rt odd
<N (N (N
dr-i'^"
M (N <N (Npico
<N (N IN
m
(N
q<Niq
to in ifi io ioo o loio
t^q(N "5t>.q iNiot^ oo-iui tôcM oqio ooo lo
o?oco
^^^ cot^t^
rHrt^
i^'n.*o6
^^^ oooo'od
^,-Hrt
oJoJd
,-1,^^
ddd
.-i<N(N
di-n'^
(N(N(N
ddco
(N(M(N
m
(N
15^55 .Îf'^ "' "^ o lo lo loicio >oioo

t-q(N oN:q if?
iNct>; qiNo i^q<N otô db-w b-c^S
ift
S
lococo
,-<^rt
ocor^
rH^,-i
i>t>^t>^
T-,^,-,
060606
,-i,_(,-i
ocdd
rt,-i,-(
ddd dd
IMC^(N
,-i,-i(N
r-I i-Imcm'
(N(N(N
m
(N
iot>.q qtcq
loiod
(Nior>. ooc o
<OCO t^t^l>- tÔOOO ooooc osdd ddi-H .-<(n'(n CO
'-!.-<.-( (NC^(N <N(N(N (N
loioid ddd dh-'t^ t>-'tô6 060606 ddd d<
O q(NU5 l0t>;0
Of
loqio oicq
"Old 10 ddt>^ 00 05 oJ oido r^ T-ôi
,-H,-lr-l rH<N(N (N (N (N
Q M
U
No JO^kO
b.q(N
10
'o^^q
lO
c^^î^
10 U5
qiNo
10 >o
tôw
10
iotN.q
lOiOiO
<N t>- (N
lOiCUS
t^ CM t^
kCi
CM
<*iu5d ididd CO '0(6 N^t^^t^.' ts!o6o6 06 06 05 ddd
CM
d
CM CM M
r-i i-H
M
CM
1-J r-<'
>
oa Tt* *< lO 000 COCDCO COt^t*
>0t>;0
NiN.'o6
qico
aiaio
lOOiO
d ' -<'' ^''
o
cm'
.-Hr^CM CMCMCM
U. 3 o 10 10 o 10 o 10
cior>. qcMo t^qcM lOh-o t^CMt-^ CMt^Cl rs.
O ^ Tj< Tj( ioioo ujdd ddt^ t-^t^t>^ oc aiai (6(6 ^' ^
O --lrH,-C CMC^JC^ CM
U.
< t^qcM qcîio ic q 10 q lo o

Tjtidid lOiOCC COCOCO 06 dd d
-i.-(r-(
d -H
CMCMCM
t^qcM otv.q cMiOb. qî>o t^qc^ 't^q CMb-CM r^csih- e^

^^^ ^<*io
co^Tjt
^^^ ioio>c
.H^rH
cocod
rt^^ dt^r>^
^^^ t-tô6
,-H,^^
o6o6d
^^^ Cbod
^CMCM
i-I
CM
0 lOiO lO iO0 0 lOiO

ct~>;q c^>">t^ q*^"? ^q'-^l "?'^q cMot^ ou? ooio o
ncc-^
^^^ >*Tj<T}<
r-.rt,-4
ieii6i6
^^^ i6c6(6
,H^^ ddt^
.-HrH^
t>it>it-^
^rH^ ooocd
,-l^rt
ddd
^CMCM
-<
C^l
ip
c^jiot>.
lo _ to
qcMiq
lo
t>:qcM
o tn
ot^q
lo
eîor^
lo
qô KSS
10 m ooio mmiA
nScm !>:
coecco *<Tj<nJ îdui dicd ddd t^r>^^ t>ô6o6 ddd d
0 lOiO lO lOiO lO lOiO
qriio i^qcM ot>.q cM'Ot>. qcMio >^qcM >ooo o>oo 10
ccooco co^'^ Tt*Tfio loicid ddd dN^h- Nio6o6 doid d
0>0 ^>O00 ON^ 0

WJ'^
N
'OtNOO
.So 3 00O>
i/s^CN CMÔ
<S(S<S <SNM JfJ0 m
165
g
Q
<
o
z
:
TtirjiTl<
^.od
^JiTjt'^
--- ---
<C^QO eOOliO OtOiN 00r}<0 "CiNOO -"fOiO OC!(N 00-*O<

conn
"5 d 50 ^'^'^
^2?2 2n:2J i?2^ ^""I'd -- -

coccco COCO'* ^ *
Tj( *** ***

9=i <W05 OOO WO0t< OcO(N OOtJho iOOCO (NXTfO

JSJS^ ^^n ?'^" ^^^'
^J^f;; ^'^'i^ ^^5" ^-i^5
<*;o) <M.oqco OJoq cocmoo -^cxd woo-* Oioq q(N-*

rtiiOip
eocoeo
<r>C05;
cococo
N-'oOOJ
cocoeo
OJOO .-I(N(n' COCOtjJ lO ic <C C N.' I^; 00
CO-'*'-* -*-*-* *-**
q*q
cc*o
qc^QO ccqio ooin oo*o co(nqo -*oio o<C(moo
coMco
^-ô h-tôo (âid d^'(N cvjcceo -*ioio dccN-V
coccco cococo coco* *** *** TfC'ji'* ****
qq<N oocoq loqq (NOO- OCOM 00 -*q loqqw

io>od co't^V-' oooo) do. (N(NCO co-*o o dd
cococo cococo cococo t>^
o
r>;Coq . - "*.9'^ ^QOW qiOO D(NOO -*0(C (NX* OiCOO
Mcoco
(n'coco
CO CO CO
*ioo
cococo
cp<t>.
cococo
r^cooJ
cococo
ddd
CO** *** coco*
^'csIcm' uiin^'^,
*** ^^^^
N ^
tU
S
z a q*o
i
rHi^co qiNoq coqio qq(N x*o ;d(n *OkOO
I cîMco' co*"o 1255 5^5^00 aJoJq" dôi
*** oinn *diod
cococo cococo cococo cococo coco* *iti^ ****
o
ni io^_t>. coq- qqiN XCOOl tOOCC w X -* q q cj X-*OiO
l odd
-iricsi COCO' Cd r-i (N (N *' d o
cococo cococ CO coco -*** *** Co' eo'
qio^ t^^noi '^,c>'0^ woqco anno (N -*oco (nx*o

o-hc^ oinn *-'5o <o<r>.* i>.'xc> ojdd ^r^ri ^w^-^.
CCCOCO cococo cococo cococo COTOM M** *^^ ^^^J^
*qio 'î^co q-*q qc-ix coquo qq<N oc*o ccc^x*
d d -H im' m' co' CO* * d 1222 t:to ddd d^'iN
cococo cococo cococo cococo cococo *** ****
coco*
c^'co'co*
oq-*q MO?"* ooM xcooi coo

^.-i^^. (noo-* o<c(n
S^?
cococo ;:;?^2J
cococo S2S22I
cococo
12'o
cococo
db-'N.'
cococo
xdd dd^'
*** cide
cococo **5
MCO-* q-o-H r>.coq -*qq
oi dd d -h' c^'
cîxco osioo cocjx *ooM
oic^co cococo
-*qiO
oor:?
-;t>:co q*q
eo*io
noo eooic oom x*oi
eocoeo JL??
cococo
in<6(6 Nih-'x oJdd d^'(N(N
eocoeo eoeo cococo coco* ****
ot3
-OQO (s-*o aor<
s ;^;^?5 ^J?;? ;s j?5^ 5S?s
167
PART IV
APPARATUS
The essential elements embodied in most installations using
fans, more especially those for heating, ventilating, or similar
work, are the fan, heater, ducts or piping system, and some form
of motive power for driving the fan. In this section will be
found complete data relative to the performance and dimensions
of fans, heaters and engines, together with detailed directions
for making fan tests. Data on the performance and dimensions
of cast iron heaters are also given.
SECTION I
FANS
Fan Design
Centrifugal fans may be roughly divided into two classes,
those having rotors with straight radial blades, and those having
rotors with blades curved with reference to their direction of rota-
tion. Curved blade fans have quite diverse characteristics,
depending on whether they are curved forward or backward
with reference to their direction of rotation. The mathematical
theory of the radial blade fan is very completely and clearly
discussed in Prof. Carpenter's book on Heating and Ventilation.
In any centrifugal fan there are two separate and independent
sources of pressure. First, pure centrifugal force due to the ro-
tation of an enclosed column of air. Second, kinetic energy
contained in the air by virtue of its velocity upon leaving the
periphery of the fan rotor. The amount of centrifugal force
imparted to the air depends largely upon the ratio of the tangen-
tial or rotational velocity of the air leaving the periphery of the
rotor to the tangential or rotational velocity of the air entering
the fan at the heel of the blades.
When the flow of air through the rotor of a fan is partially
obstructed the centrifugal effect in the rotor produces a com-
pression corresponding to the centrifugal force, which is known
as static pressure. On the other hand, the kinetic energy of the
air leaving the periphery of the rotor must first be converted
largely into potential energy in the form of static pressure before
168
FAN DESIGN
being serviceable. This conversion from kinetic energy or veloc-
ity into potential energy or static pressure is ordinarily accom-
plished in the scroll formation of the fan housing. A still further

conversion is often secured, where the velocity leaving the outlet
is high, by means of a diverging nozzle on the outlet of the fan.
The principle covering the design and application of such nozzles
is discussed on page 122.
STRAIGHT FORWARD BACKWAR!)

TYPES OF FAN BLADES.
The accompanying diagrams represent the parallelogram of
forces for the three general types of blades, the first a straight
blade, the second a blade bent forward and the third where the
blade bent backward. The line U2 represents the tip speed
is
of the wheel and the line V2 represents the radial velocity of the
air leaving the tip of the blade. The diagonal line R then repre-
sents the actual velocity of the air with respect to the fan casing.
The amount of total pressure developed by a straight blade
fan may be determined by means of the following formula:
(U'-U;)-f- MU'-(l-M)V'-(NV.)2
(73)
yl.
where p = total press, developed by fan.
Vo = velocity of air through inlet.
Vp = vel. corresponding to unit press.
V2 = radial vel. of air leaving tip of blades.
Ui= lineal vel. at heel of blades.
U2 = lineal vel. at tip of blades.
M =per cent, velocity pressure conversion in fan scroll.
N= ratio of actual to effective area of inlet.
It will be noticed from the two diagrams of curved blade
wheels that when the blade is bent forward an accelerated veloc-
ity will be obtained, while with the blades bent backward the
opposite efifect will be the result. This explains how it is pos-
sible to build a fan with a small wheel, such as is used in the
multiblade type, and obtain the desired pressure and velocity
without using excessive speeds. By curving the blades forward
a pressure greater than that due to the peripheral velocity is
obtained, as indicated in the diagram.
The velocity of the air leaving the tip of the blades and the
corresponding velocity pressure is greatly in excess of that or-
dinarily required in the piping system, and at the same time
the static pressure is too low. By enclosing the wheel in a cas-
ing having a properly designed scroll, this velocity is reduced,
and a part of the velocity pressure is converted to static pres-
sure. Since the static pressure due to the wheel varies as
the difference of the squares of the rotational velocities at the
periphery and inlet, it is evident that the shorter the blade the
greater must be the dependence on the scroll-shaped housing
to obtain the desired static pressure. For this reason the pro-
per design of the housing is of greater importance in the case of
a short blade multivane type of fan than with the older styles.
There are frequently cases where a fan is to be direct con-

nected to a high speed unit, where the corresponding pressure
obtained would be greater than required. In this case the back-
ward bent blade is used, since, as may be noticed from the dia-
gram, a pressure less than that corresponding to the peripheral
velocity is then obtained.
The standard steel plate fan is essentially a straight blade

fan, as compared with the later styles of short curved blade
multivane type, although, as just shown, when the tips of the
blades are bent either forward or backward the fan will have
different characteristics from one with straight blades. This
fan as ordinarily built does not give as high an efficiency as the
multivane type owing to the fact that it is designed for large
capacity rather than for high efficiency. But if these long blade
fans are built according to special design they may be made to
give greater efficiency than may be obtained from the curved
short blade fans. This calls for a tall narrow fan with the inlet
diameter smaller than that used on the standard fan. It may
170
FAN DESIGN
be readily shown that there a certain diameter of inlet that
is
will give maximum economy of operation.If the diameter is

increased the loss by impact at the heel of the blades is increased
as the square of the diameter, and the loss by entrance is de-
creased as the fourth power of the diameter. The opposite
holds true in case the inlet diameter is decreased.
The proper size of the fan inlet depends on the cubic feet of
air per revolution handled by the fan. It has been determined
both mathematically and experimentally that the most efficient
diameter of inlet is given by the simple relationship
D, -^'1 (74)
where Di = inlet diameter in feet.

Q = cubic feet of air per minute.
N = revolutions per minute.
C = a factor determined experimentally, and is
practically a constant for all ratios of inlet diameter

to wheel diameter.
COMPARATIVE EFFECT OF BLAST-WHEEL PROPORTIONS UPON THE

EFFECT OF STRAIGHT BLADE FANS OPERATING AT THE
SAME CAPACITY AND PRESSURE
ratios. It will be seen from the second and fourth columns
that the height of the fan increases rapidly while the width
decreases. This means, then, that these special high efficiency
fans are tall and narrow, which naturally makes them more ex-
pensive than the ordinary commercial steel plate fan.
These special tall narrow fans are frequently used for induced
draft work, partly because the narrow wheel makes a shorter
over-hang on the fan bearing, and partly because they may be
operated at lower speed and are therefore more suitable for
direct connection to steam engines. A table of special induced
draft fans will be found on pages 328 to 330, giving the size of
engine and dimensions of fan for various boiler capacities.
Pressure Characteristics
The relative performance of the different styles of fans may

be shown by the performance curves given for each style in
Section III. The diagrams on pages 214 and 215 or on pages 224
and 225 are typical of the straight blade fan, while the diagram
on page 276 shows the performance of the Niagara Conoidal fan,
it being typical of the forward curved blade style.
In making the tests on these fans from which the diagrams

referred to were computed, the fan was operated at constant
speed with varying sized discharge orifices. Horsepower, pres-
sure and capacity readings were taken and the efficiency cal-
culated. The horsepower, pressure and efficiency were then
plotted to capacity, giving the performance curves of this test
fan. This set of curves shows the power consumption increasing
with the capacity, but not at the same ratio. The efficiency
increases to a maximum and then decreases to zero. The rating
of the fan may be taken at any point adopted as a standard, but
the desirable point will generally be determined by the point of
maximum efficiency.
As will be noted from the diagrams already referred to, the

relative pressure developed by centrifugal fans will differ for
fans having straight or curved blades. A further comparison
of the two types may be made from the diagram on page 173.
The two curves indicate the ratio of the total pressure developed
to the pressure corresponding to the peripheral velocity of the
wheel, when operating at different per cents, of the fan's rated
172
FAN CHARACTERISTICS
;^?
r
I
o o O O
Q
00
ID 00
A1I3013A 'Id2d 01 ONIQNOdSaadOD

S93^d 01 SS3dd "IVIOI dO 0\1VH
173
capacity. Thus we see that at rated capacity the straight blade
fan gives a total pressure of 87 per cent, and the forward curved
blade type gives 208 per cent, of the pressure corresponding to
'the peripheral velocity.
When the Niagara Conoidal fan is operated up to 50 per cent,
overload the total pressure increases, but when operating at less
than rated capacity the pressure decreases. Just the opposite
holds true in the case of the straight blade fan. From this we
note that if a forward curved blade fan is intended to operate at
a certain pressure and capacity and if for any reason, such as
resistance greater than expected, the quantity of air handled is
less than the fan's rating for the speed maintained, the total
pressure will also be less than that specified. With the straight
blade fan just the opposite holds true, for as the capacity is
reduced the pressure will increase, at constant speed.
Care should be taken in the selection of a fan with forward
curved blades in case it is to be driven with a motor. If for any
reason there should be a tendency to operate at over capacity
both the air quantity and the pressure will increase, which may
overload the motor in case sufficient margin has not been
allowed.
Special Types and Features

There are numerous special types of fans intended to meet
various requirements. Some of these, as for instancefoundry
and forge shop blowers and exhausters or planing-mill exhaust-
ers, are adaptations of the ordinary straight blade fans. There
are other types, such as the disc or propeller wheel, that differ
essentially in the principle of their design and operation. These
special types will be found described under their proper heading,
together with their capacity and dimension tables.
The steel plate fans may be divided into two classes, namely,
blower or exhauster, depending on whether they have two inlets
or only one. With the double inlet fan, or blower, we have the
total inlet area divided between the two sides of the housing,
therefore each inlet may be made smaller in diameter for the
same size of fan wheel, and so approach the more efficient type
of fan as indicated in the table on page 171. For this reason the
steel plate blower is a more efficient fan than the single inlet fan,
or exhauster. Either may be made full housing or three-quarter
174
FAN CHARACTERISTIC
housing, depending on the size and requirements. The multi-
vane fan as ordinarily built is a single width fan with but one
inlet. These are also built in a double width style, with two .
inlets,being essentially two fans placed back to back. They O

have the same characteristics as regards full or three-quarteiC^
housing, as also the angle of discharge, as has the steel pla-t. o ^ .
exhauster. C
Both the steel plate
built with a bearing
and the multivane fans are ordinarily
on each side of the housing. In the case of
^
C
the exhauster, where any substance is to be handled that would^Q
be injurious to the bearing located in the intake, they are usually 1i^
made with an overhung wheel, a pedestal supporting an extra O
bearing being attached to the back of the housing. Special''-'-^
exhausters, as for instance planing-mill exhausters, are built ^^
with an overhanging wheel in order to avoid obstructing the^^
inlet of the fan. Induced draft fans are built with an overhung ^
wheel to avoid drawing the hot gases over the bearing. In any C
fan handling hot gases the bearing attached to the back of the L-
housing should be of a special water cooled type to avoid heating. QJ
The regular discharges of fans and blowers are designated ^
as top or bottom horizontal discharge, up or down blast, and*
special, which are described by giving the angle of the discharge
from the horizontal. The hand of a fan or blower is determined i^
^
by the side on which the pulley or engine is located. Standing
facing or nearest to the discharge outlet, the fan is right or left*^
hand, according to whether the pulley is on the right- or left-hand C
side. C-
Horsepower
Each cubic
of Fan
foot of air per minute moved
against a total
Û
pressure of one inch water gauge, equivalent to 0.577 oz. per
square inch, or 5.19 pounds per square foot, represents the
^ *
expenditure of 5.19 foot-pounds of work. We then have as the j

theoretical expenditure of energy in doing this work
c 1Q
q
335^
= 0.000157 H. p.
and also
1
6370 A. P.M.
0.000157
That is, with perfect efficiency, it will require 0.000157 H. P.
to move one cubic foot of air per minute against a pressure of one
175
inch, or 6370 cu. ft. minute moved against one inch
of air per
pressure will require one horsepower. Assuming a fan efficiency
of60 per cent, will give 0.000261 H. P. per cubic foot of air per
minute per inch total pressure
H P - ^' ^' MX0Q00^57X total press, in inches

total efficiency
TT p A. P. M.X 0.000157 X static press, in inches

static efficiency
where total or static efficiency refers to the efficiency based re-

spectively on the total or static pressure. The ratio of total to
static pressure will remain constant for any style of fan at rated
capacity but will vary for the different types.
In the case of straight blade fans we may determine the horse-
power for any given air delivery by assuming twice the pressure
corresponding to the peripheral velocity with a corresponding
efficiency of 100 per cent. This will give approximately the
true horsepower regardless of actual pressure or efficiency ob-
tained. Thus if we have a straight blade fan delivering 30000
A. P. M. at 230 R. P. M. with a fan wheel 833/^ inches in diam-
eter, we will have a peripheral velocity of 5040 feet per minute,
and twice the pressure corresponding to the peripheral velocity
will be 3.17 inches. Then the horsepower required by the fan
will be
H. P. =30000X0.000157X3.17 = 14.9.
Relations of Total, Static and Velocity Pressure

In fan work a certain static pressure
air is delivered against
or resistance of thesystem and in addition has imparted to it a
certain velocity at the fan outlet. This velocity is dependent
on the amount of air required and on the area of the fan outlet,
and the velocity pressure expressed in inches corresponding to
this velocity may be determined from the formula (see page 20)
Pv = I )
= inches of water
where the term velocity refers to the velocity of the air through
the fan outlet in feet per minute.
When it is desired to express the velocity pressure in ounces
per square inch the following formula should be used:
176
FAN CHARACTERISTICS
(velocityN 2
=oz.persq.m.
-5273 ;
The total energy imparted to the air is composed of the static

pressure of the system and the energy of discharge correspond-
ing to the velocity pressure or velocity head as it is termed in
hydraulics. The total pressure is the sum of the velocity pres-
sure at the fan outlet plus the static pressure produced, and is the
pressure upon which the performance and efficiency of the fan is
usually based. In the case of an exhaust system, the static head

on the fan should be taken as the difference in static pressure at
the inlet and outlet of the fan. The method to be used in making
these various pressure determinations is fully explained under
"Fan Testing/' Part IV, Section II.
The ratio of static to velocity pressure at the fan outlet is
very important in fan engineering. This ratio varies as the
capacity of the fan is varied at constant speed, and bears a defi-
nite experimental relationship to the efficiency of the fan. The
rated fan performances which represent the most desirable con-
ditions of operation are based on certain relationships of static
to velocity pressure. For instance, the rated performance of
the Planoidal fan is based on a relationship of
-2^ = 3.88
Pv
The performance of the Planoidal fan with reference to this

ratio is shown by the diagram on page 215. As an illustration
of its use we will assume an 80-inch Planoidal exhauster operat-
ing against a static pressure of % inch and delivering 12000 cu.

ft. air per minute.
Since the outlet of this fan is 5.54 sq. ft. (see
table on page 207) the velocity at the fan outlet will be
12000-^5.54=2170 ft. per minute, and the velocity pressure
will be
-=^=0.294 in.
The ratio of static to velocity pressure will then be
P'- Q-75 o^g

-"^-^^
Pv
~ 0.294
From the diagram on page 215 we find that with the above
ratio, the fan will be operating at 111.5 per cent, of rated
177
capacity, with correspondingly increased power consumption and
lowered efficiency. Further examples illustrating the applica-
tion of these diagrams will be found under "Selection of a Fan"
on page 182.
For certain kinds of work a low velocity in the piping system
is desired, while in other cases it is necessary to maintain a high
velocity. In case a lower velocity is required in the piping

system than that maintained at the fan outlet, the area of the
main pipe should be gradually increased. This is termed a
diverging cone on the fan outlet, and if properly proportioned
the loss due to the reduction in velocity at this point will be
reduced to a minimum. Rules for the design of cone outlets
and their effect in increasing the static pressure will be found
given under "Diverging Nozzles in Air Ducts" on page 122.
The Relation Between Pressure, Velocity

and Air Density in Fan Work
For low pressures, as in fan work, we may consider that the
pressure varies inversely as the absolute temperature and directly
as the barometric pressure. The volume of the same weight of
air is directly and the weight of the same volume is inversely
proportional to the absolute temperature. We will then have
^ = -f and^=-iorp = pX^X^
p lo p b
'^
lo b
(77)
where = pressure at absolute temp. T and barom. b.

p
Po = pressure at absolute temp. To and barom. bo-
T=
absolute temp, of the air in deg. Fahr.
b= barometric pressure in in. of mercury.
Then in order to correct any given pressure reading p, at
temperature t and barometer b, to the corresponding pressure
for standard dry air at 70 F. and 29.92'' barom. we will have
^ 530 ^ b 0.075 p ._g-

^^^^
Po=PXli60TT^ 29:92'- ~^r~
Since at constant capacity and speed the power consumption
will vary as the pressure, and the pressure varies as the density
of the air, we will also have
H P -rH P
H.P.-(H.P.)oX Wb _ (H.P.)o
i60+i 29.92
-:53o-X - 0.075
W ^^"^
Thus, if a fan is to operate under some other condition than

standard air, corrections can be made for pressure and horse-
178

FAN CHARACTERISTICS
power by equations (78) and (80) respectively. For illustration
seeexample 5, page 188.
In a centrifugal fan working under a constant orifice condi-
tion and at known air density, the theoretical velocity and
pressure developed each bears a definite relation to the peri-
pheral or tip velocity of the fan wheel. That is, the air velocity
at the fan outlet and capacity is directly proportional to the
peripheral velocity and fan speed, and the pressure developed
varies directly as the square of the peripheral velocity and
therefore as the square of the fan speed.Since the horsepower
is proportional to the product of the pressure and capacity,
the horsepower evidently varies as the cube of the fan speed.
These combined relationships may be expressed by the follow-
ing formula:
N'W = N'W
H.P.

p
-^
Po
N^Vo
=- 5
and
,
(H.P.)o NjWo
r
(81)
where N = revolutions per minute, and W = air density.
Laws of Fan Performance

In the selection and operation of fans, the size, speed, capac-
ity, horsepower, and pressure each has a fixed and definite rela-
tion to the other, which may be expressed as follows:
For a given fan size, piping system, and air density
Capacity varies directly as speed.

1
Velocity varies as speed or capacity.

2
3 Pressure varies as the square of the speed.
4 Speed and capacity vary as square root of the pressure.
5 Horsepower varies as cube of the speed or capacity.
6 Horsepower varies as (pressure )^'^2
For a constant pressure and at rated capacity

7 Capacity and horsepower vary as square of the size.
8 Speed varies inversely as size.
9 At constant pressure the speed, capacity and horsepower
vary as the square root of the absolute temperature.
10 At constant capacity and speed, the horsepower and pres-
sure will vary directly as the density of the air and approximately
inversely as the absolute temperature. Thus increasing the
temperature from 50 to 550 practically cuts the horsepower
and pressure in half if the speed and capacity remain the same.
179
It has been shown that the pressure loss or resistance (ps) of
a given system of air passages varies as the square of the air
velocity and therefore as the square of the air quantity, (Q).
Also, since the pressure produced by a fan varies as the square
of the speed (Law 3) under constant outlet conditions, it is evi-
dent that the air quantity delivered by the fan through a given
system of air passages will vary directly as the speed, (Law 1).
The relation between pressure loss and the air quantity passed
through a given system may be termed the coefficient of resist-
ance (R) of that system in the formula ps =R Q^, and R = -^
Also on page 125, it has been shown that each system has an
equivalent or blast-area (A), where the pressure loss may also be
/ Q \ 2
expressed as ps = (
,. ) and where Vq is the velocity corres-
\* 0^^/ / 1 \ 2
ponding to unit pressure. Hence R= (
j
If the coefficient of resistance or blast-area of a system to

which a fan is applied should be changed while the fan
is oper-
ated at constant speed, then the air delivery will undergo a

change which is complicated by the fact that the fan pressure
also changes with the quantity of air delivered, as shown by
the pressure-capacity characteristic. This is best illustrated
by the accompanying diagram. In the first case suppose we have
a system whose coefficient of resistance is Ri = 142/l0^ and in
the second case the system is changed so the coefficient of
SELECTION OF A FAN
resistance is reduced to Rg =42/10^. Curves Rj and Rj show
how the pressure losses will vary with the air delivery in the
respective cases, and curve P shows how the pressure produced
by a 90-inch Planoidal exhauster at 330 R. P. M. varies as the
air delivery. Itevident that the air delivery will be Qi for
is
the first case and Q2 for the second, with corresponding static
pressures pi and pj.
Selection of a Fan
It is well known and capable of demonstration in practice as
well as in theory that of two straight blade fan wheels, the one
having longer blades gives greater pressure, and that curving the
blades forward in the direction of rotation increases the pres-
sure, the converse also being true.
It is not a fact that a fan with forward curved blades is on
that account any more efficient than one with radial blades;
the two types have radically different characteristics, and each a
field in which it excels; with the short forward curved blades
good efficiency requires a greatly increased number as compared
with the few blades of the radial type familiar in the steel plate
fan.
In both types the need of careful design does not end with
the proportions of the blades; the design of the scroll or housing,
the area and position of the outlet and the diameter of the inlet
are very important factors.
As explained under "Pressure Characteristics," the per-
formance at other than rated capacity of the older style straight
blade fan is entirely different from that of the curved blade
multivane type. With the straight blade fan the pressure drops
off rapidly when operated at overload, but increases when the
fan is operated at less than rated capacity. In the case of the
multiblade fans, the static pressure is greatest at normal load,
and decreases at capacities either above or below this rated
point.
Thus we see in the case of a system where a uniform air
quantity is desired, whether for heating, ventilating, forced
draft or for drying processes, the steel plate fan will come nearer
giving this uniform quantity in spite of variations in resistance,
throttling effect of closing dampers, and similar conditions.
On the other hand, it is sometimes very desirable to be able
to throttle the capacity of a fan without increasing the pressure
181
and velocity, as for instance, if one wing of a building is closed
off and it is not convenient to change the speed of the fan, the
steel plate fan would deliver an increased amount of air into the
remaining part of the system on account of its increased pres-
sure, while the multiblade fan would be more sensitive to the
increased resistance and would show only a slight increase in
velocity through the ducts which remain open.
In general, the multiblade fans, of which the Niagara Conoi-
dal is a type, require less space than steel plate fans of equal
capacity and efficiency. Another important advantage is the

fact that the higher speeds of these multiblade fans make them
more suitable for direct connection to motors, or at least give
better pulley ratios than may be obtained with radial blade fans
of equal efficiency.
After determining the style of fan to be used, there are two
things to be considered in its selection. It must supply a defi-
nite amount of air perminute and it must supply this air at
sufficient static pressure to overcome the friction loss of the sys-
tem, which should be calculated or determined with reasonable ac-
curacy. The performance of the fan under this condition is
determined from the capacity tables.
Example 1. Assume that it is required to deliver 16500
cu. ft. air per minute against a static pressure of 0.95 in., and
determine the size of Planoidal exhauster to be used, with the
corresponding speed and horsepower. From the diagram on
page 214 we see that at rated capacity the Planoidal fan gives a
static pressure which is 79 per cent, of the rated total pressure.
Then the total pressure corresponding to 0.95 in. static will be
1.20 in. From the table on page 208 we see that we may
choose between a 90-inch fan operating over capacity or a 100-
inch fan operating at less than rated capacity.
A 90-inch Planoidal fan has an outlet area of 7.10 sq. ft.
(table page 208) so that the velocity through the outlet will be
16500^7.1=2325 feet per minute. From formula 21 on page
20 we have the pressure varies as the square of the velocity, and
from formula 13 on page 18 the velocity for dry air at 70 F.
corresponding to one inch pressure is 4005 feet per minute.
The velocity head or velocity pressure at the outlet of the
90-inch fan corresponding to a velocity of 2325 feet per minute
will then be
pv = (2325^4005)2 =0.337 in.
182
SELECTION OF A FAN
The total pressure against which the 90-inch fan must operate
will then be the sum of the static and velocity pressures, or 0.95
+ 0.337 = 1.287 inches, ratio of static to velocity pressure
and the
will be 0.95 -^0. 337 = 2.82.
Referring to the diagram on page ^
215, we find a point on the bottom scale corresponding to a ratio O

of static to velocity pressure of 2.82, and from the intersectioi(LJ
of a vertical from this point with the curves above we mayj^^
determine the relative performance of the fan from the scale on CI
the left-hand edge of the chart. Thu^ we find that the fan will
be operating at 108 per cent, of the rated capacity, and require
^
C
104.5 per cent, of the rated horsepower. ,5
As already stated, the fan required to deliver 16500 A.
is ^ P.
M.; so that 16500-^1.08 = 15300 A. M., the rated capacity
P.
^
of the fan at the speed used. From the table on page 208 we see"^
that this fan will deliver 14890 A. P. M. at 334 R. P. M. anc^^
require 6.65 H. P. According to the relations given on page 179
the speed varies directly as the capacity and the power as theOJ
cube of the capacity. We will then have for a rated capacity of ^
15300 A. P. M.
^
= 334 = 344
Actual speed
(jfl^)
R. P. M.
g
^
Rated power = 6.65 = 7.22 H. P.
(|||^) ^
But as already stated, when operating at 108 per cent, of the
rated capacity, or 16500 A. P. M., this fan will require 104.5 ^
per cent, of the rated power, so we have G
Actual power
will
= 1.045 X 7.22 = 7.55 H. P.

^
The pressure developed will be 0.95 inch static and 1.287 inches q-
total. ^
In case a 100-inch Planoidal fan should be selected for thi*^
service it will operate at less than the rated capacity, and the ^
power required may be determined in the same manner as for ^
the 90-inch fan. The outlet area of a 100-inch fan is 8.75 sq. ftQ,
so the velocity at the outlet will be 16500-^8.75 = 1885 feet per
minute and the corresponding velocity pressure will be (1885 -^
4005)2 = 0.221 in. Since the fan is required to develop 0.95 in.
static, the total pressure against which it will operate will be
0.95-1-0.221 =1.171 in., and the ratio of static to velocity pressure
will be 0.95-^0.221 =4.30. From the diagram on page 215 we
183
find that with a ratio of static to velocity pressure of 4,30 the
fan will operate at 97 per cent, of the rated capacity and require
98.5 per cent, of the rated power.
As 16500 is to be 97 per cent, of the fan's rated capacity at
the speed used, we have 16500-7-0.97 = 17000 A. P. M. as the
rated capacity. From the table on page 208 we note that the J
100-inch fan at 300 R. P, M. will deliver 18370 A. P. M. and^
require 8.20 H, P, To obtain a rated capacity of 17000; A. P, M
it will be necessary to reduce the speed and consequently the
power to
Speed = 300 =278 R. P. M.

(j||^)
As determined from the diagram, the power at 97 per cent,

capacity will be 98,5 per cent, of the rated, so we will actually
require
0.985X6.50 = 6.40 H. P.
Thus we see that while the first cost of the 100-inch fan will
be greater, the cost for power will be less than with the 90-inch
fan.
Example 2. A case frequently met in the application of
fans is where the resistance against which the fan must operate
is different from any of the pressures given in the fan capacity

tables.
We will assume that 12000 A. P. M. is required at 0.20 in.
static resistance. What size of fan shall be used and what will
be the speed and horsepower? If a fan should be required to
operate at rated capacity at a speed corresponding to the 0.20
in. resistance, we may select a Planoidal fan from the table
on page 210. We note that the lowest pressure given in the

table is %inch, but from the ratio given on page 179 we may de-
termine the speed, capacity, and horsepower at 0.20 in. as
follows: Since the capacity varies as the square root of the pres-
sure, and we require 12000 A. P. M. at 0.20 in. the correspond-
ing capacity at % in. will be 12000 (0.375 ^0.2)^2=. 15450
A. P. M.
We see from the table on page 210 that a 120-inch Planoidal
exhauster at %
inch static will have a rated capacity of 16030
A. P. M. at 152 R. P. M., and 2.62 H. P. We would use this size
184
SELECTION OF A FAN
and operate it at slightly over the rated capacity to give 16450
A. P. M. at H inch or 12000 A. P. M. at 0.20 in. The rated
speed of the 120-inch fan at 0.20 inch pressure will be 152 (0.20
^0.375)^2 = 111 R. p. M. and the rated power will be 2.62 (^)^
= 1.02 H. P. When operating at 16450^16030 = 102 per cent,
of the rated capacity, the power required would be 1.08 H. P.
In order to determine the performance of a smaller fan under

these conditions we will assume that a 100-inch Planoidal ex-
hauster operating at over capacity is to be used. The outlet
area of a 100-inch fan is 8.75 sq. ft. in area so that the velocity
through the outlet will be 12000-7-8.75 = 1370 ft. per minute
and the corresponding velocity pressure will be (1370 h- 4005)^ =
0.118 in. The ratio of static to velocity pressure will then be
0.20 -4- 0. 1 18 = 1.7, and from the diagram on page 215 we note that
with this ratio the exhauster will operate at 118 per cent, of
rated capacity and require 110 per cent, of the rated power.
If 12000 A. P. M. 118 per cent, of the fan's rated capacity

is
at the speed required to meet the assumed conditions, we will

have 12000^1.18 = 10150 A. P. M. as the rated capacity. The
table on page 210 does not give the speed and power required for
10150 A. P. M., but does give 182 R. P. M. and 1.81 H. P. for
11140 A. P. M. Since the speed varies directly and the power
as the cube of the capacity we will have for 10150 A. P. M., 182
(10150 -^ 11140) = 162 R. P. M.,and 1.81 (10150-^11140) = 1.28
H. P. as the rated speed and power. As already found from the
diagram on page 215 with a ratio of static to velocity pressure of
1.7, the power required will be 110 per cent, of the rated, which
gives us under the assumed conditions 1.28X1.10 = 1.41 H. P.
when delivering 12000 A. P. M. against 0.20 in. static resist-
ance at 162 R. P. M.
We see from the table on page 210 that a speed of 182 R. P. M.

corresponds to a static pressure of 0.375 inch, and as the pressure
varies as the square of the speed, the pressure for 162 R. P. M.
will be 0.375 (162 -=- 182)^ = 0.296 in. That is, although the
resistance of the system is only 0.2 in. and would call for a 120-
inch fan, we may reduce the initial cost by using a 100-inch fan
operating at a speed corresponding to approximately 0.3 in.
with but 30 per cent, increase in the power consumption. Where
the fan is to be direct connected to an engine and the exhaust
185
steam used in the heating coils, this additional power is of little
or no consideration.
Selection of a Niagara Conoidal Fan

Example 3. The Niagara Conoidal fan may be selected
either from the static pressure tables on pages 232 to 273 or from
the total pressure tables on pages 228 to 231. The total pressure
tables, like the tables for Planoidal fans, give the performance
of this fan at its point of rating only. The static pressure tables
give the performance at other than the rated capacity, and give
the speed and power required on both sides of the most efficient
point. The tables on pages 274 and 275 indicate the efficiencies
obtainable under different conditions of pressure and outlet
velocity with these fans. Thus we see that there is one point in
each pressure column at which the fan will give the highest effi-
ciency. In the selection of these fans it may often be found
expedient to operate at other than the most efficient point.
When selecting a fan for use in a public building it is advis-
able to use a velocity of about 1800 feet per minute through the
fan outlet, with a maximum allowable velocity of 2200 for such
work. For industrial installations, where higher duct velocities
are the rule, outlet velocities up to 4000 may be used, without
varjîng greatly from the most efficient performance.
To illustrate the use of the static pressure tables we will
assume that it is required to deliver 17000 cu. ft. of air per
minute against a pressure of one inch static. By an inspection of
the corresponding tables, we find that we may use a No. 6 at
419 R. P. M., 6.59 H. P., and an outlet velocity of 3200 feet per
minute; a No. 7 at 332 R. P. M., 5.19 H. P., and an outlet veloc-
ity of 2400 feet per minute or a No. 8 at 291 R. P. M., 4.86 H. P.,
and an outlet velocity of 1800 feet per minute. For use in a
public building the No. 8 should be selected, but in case it is
desirable to use higher duct velocities and absolute quietness of
operation is not essential, either the No. 7 or No. 6 may be used.
Example 4. A common case of variable resistance in a fan
system of heating and ventilating is where a fan is selected to
supply a definite amount of air, and during the winter this air
is drawn through the heater, but during the summer the damper
to the by-pass is open so that the air may be drawn through both
the heater and by-pass. As shown by the tables on pages 446
and 447, the resistance due to the heater will depend upon its
depth and the velocity of the air through the clear area. From
186
SELECTION OF A FAN
page 457 we see that under average conditions, we may assume
two velocity heads lost due to the by-pass. Assuming a case
where the heater is four sections deep with a velocity of 1000 feet
per minute through the clear area, we find from the table on
page 446 that the resistance will be 0.382 in. Allowing a loss
of 0.24 in. static in the piping system, the fan will be required
to operate against a static pressure of 0.382+0.240 = 0.622 in.
or ^ in. This is under normal working conditions when the
by-pass damper is closed.
We assume that a Planoidal exhauster is required

will first
to deliver 25000 A. P. M. under the above conditions. With
this type of fan at rated capacity the static will be 79 per cent,
of the total pressure, so that with a static resistance of 0.622 in.
the corresponding total pressure will be 0.622^0.79=0.787 in.
or approximately in. %
From the capacity table on page 210,
we a 130-inch Planoidal exhauster at
find that inch static %
pressure has a capacity of 24150 A. P. M. at 180 R. P. M. and
will require 6.57 H. P. As this capacity is within a few per cent,
of that required, it will be taken as the rated condition.
According to the data given on page 457 the resistance of a

standard by-pass is approximately the same as that for four
sections of Buffalo heater, so that in the case assumed when the
by-pass damper is opened the effective area will be doubled.
Since the loss due to the resistance varies as the square of the
velocity and the velocity is to be reduced to }4, the resistance
for the same air quantity will be (3^)^ or H
of what it was when
all the air passed through the heater. That is, with the by-pass
damper open, the resistance at the heater will be one- fourth of
0.382 or 0.095 inch.
With the damper in the by-pass open the static resistance of

the system will be reduced to 0.095+0.240 = 0.335 in. provid-
ing the same air quantity is handled and it will be required to
determine the results obtained under this new condition. The
area of the outlet of the 130-inch fan is 14.85 sq. ft., so the
velocity through the outlet under rated conditions would be
24600-7-14.85 = 1685 ft. per minute and the corresponding
velocity pressure (1685-^4005)2 = 0.177 in. With the by-pass
damper open the static pressure based on the same air quantity
is 0.335and the ratio of static to velocity pressure will be
in.
0.335^0.177 in. = 1.89. From the diagram on page 215 we find

187
that with this ratio the actual static pressure will be 65.5
per cent., the capacity 116.5 per cent, and the power 109 per cent,
of the rated as given on page 187. That is, we will have
actual pressure =0.622X0.655 =0.407 in. static.
actual capacity =24150X1.165 =28150 A. P. M.
actual power = 6.57X1.09 = 7.15 H. P.
Correction for Temperature

Example 5. A case frequently met in selecting a fan is
where the air to be handled is specified at some temperature
other than the standard of the fan tables (70 F.). For instance,
a "B" Volume Exhauster is required to handle 5500 cu. ft. of air
per minute at a temperature of 600 F. against a pressure of two
ounces. What size exhauster should be used and what will be
the speed and horsepower? This fan is to handle the air at 600
while the capacity tables are based on air at 70.
As explained on page 179, if the speed and capacity are kept
constant the pressure and horsepower will vary inversely as the
absolute temperature. Thus an increase in temperature from
70 to 600 doubles the absolute temperature (1060-7-530 = 2)
and if we handle 5500 A. P. M. at 70
select a fan that will
against four ounces have the same capacity at the same
it will
speed against two ounces when the temperature is 600 and the
power will be half that given in the table for four ounces. From
the capacity table of "B" Volume Exhausters on page 335 we
find that the nearest size to that required will be No. 8, operating
at 1420 R. P. M. and requiring 10.20^2=5.1 H. P.
Another exam^ple illustrating the effect of temperature would

be to assume a fan is delivering 3500 A. P. M. at 1000 F. against
1 3^ oz. pressure with a speed of 920 R. P. M. What will be the
speed and capacity of this fan at 500 and 2 oz. pressure?
The relative pressure of the air at 1000 and 500 is given by
the ratio of the absolute temperature, or 1460 -^ 960 = 1.52.
That is, if this fan handles the same volume of air at the same
speed, due to the change in the temperature the pressure devel-
oped will be 1.5X1.52=2.28 oz. But this fan is required to
operate at 2 oz. instead of 2.28 oz., which calls for a lower speed.
We may see from page 179 that the speed and consequently the
capacity will vary as the square root of the pressure, so the speed
at 2 oz. will be 920 (2 -=-2.28)^2 = 708 R. P. M. and the capacity
188
SELECTION OF A FAN
willbe 3500 (2 ^2.28)^'2 = 2700 A. P. M. Thus we see that this
same fan will deliver 2700 A. P. M. at 500 against 2 oz. pressure
when operated at 708 R. P. M.
Correction for Altitude

Example 6. The following example represents the calcula-
tions required to correct the fan performance to a sea-level
(29.92 inches barometer) basis. Required a fan to handle 40000
A. P. M. against 0.5 in. staticpressure at an altitude of 5000
feet with a temperature of 70. As the fan tables are based on
air at sea-level, we will have to reduce the above specified pres-
sure to a sea-level basis. From the diagram on page 25 we note
that the relative pressure for this altitude
is 0.835, so that a sea-
level pressure corresponding to a pressure of 0.5 in. at 5000

altitude is 0.5-^0.835=0.6 in., or approximately %
in. The
horsepower required to operate the fan will be 83.5 per cent, of
the rated horsepower as given in the capacity tables for ^
in.
static.
We find from the table of Niagara Conoidal capacities on

page 248 that a double No. 8 fan with an outlet velocity of 2200
feet per minute would answer the requirement at ^
in. at
238 R. P. M. and 8.66 H. P. At 5,000 altitude the power con-
sumption will be 8.66X0.835 = 7.23 H. P.
Another example illustrating the correction for altitude
would be to require an induced draft system for 1600 boiler H. P.,
at an altitude of 5200 feet with flue gases at 550 F. and a static
pressure of one inch required. From the diagram on page 25
we find that the factor for 5200 altitude is 0.83. The corres-
ponding pressure at sea-level will then be 1 -e- 0.83 = 1.20 in.
From the table on page 325 we find that a 150-inch Planoidal
exhauster operating at 305 R. P. M. will have a capacity of 63110
A. P. M. at 1.25 in. static with a temperature of 550 and re-
quire 34.0 H. P. With an allowance of 32.4 A. P. M. at this
temperature per boiler H. P. this fan will be capable of supplying
draft for 1950 H. P. of boilers at sea-level. The air required
per boiler horsepower at an altitude of 5200 feet will be 32.4 -^
0.83 = 39.0 A. P. M. so that the draft capacity of this fan at 5200
altitude will be only 63110^39.0 = 1620 or 1950X0.83 = 1620
boiler H. P. The power requirements at 5200 altitude will be
34.0X0.83 = 28.2 H. P.
189
SECTION II
FAN TESTING
Itfrequently necessary to make a test on a fan installation
is
in order todetermine the quantity of air being delivered. There

are several methods of making this determination, depending
on the degree of accuracy desired, the object of the test, or the
conditions under which the system is installed. The velocity
and quantity of air delivered by a fan or flowing through a duct
may be found by meansa pitot tube, an anemometer, a con-
of
verging nozzle, an or a short length of pipe.
orifice, Each
method may be especially applicable under various conditions
and a selection should depend on the object and accuracy
desired.
The most accurate method for ordinary work is to use the
pitot tube, either in an air duct or in connection with a converg-
ing nozzle attached to the fan outlet. The anemometer is espe-
cially useful in determining the velocity and quantity of air enter-
ing a room in order to properly proportion the air distribution
in an indirect heating or ventilating system. An orifice or a
short length of pipe is frequently used in connection with test
work, where a permanent piece of apparatus is desired.
The Pitot Tube

The is an instrument used for making velocity
pitot tube
measurements of a current of air, the principle of its action being
shown by the accompanying diagram. As already explained.
190
FAN TESTING
when there exists a flow of air due to a certain pressure, a part
of this pressure, termed the velocity head, is transformed into
velocity, while the balance, termed pressure of static head,
serves to produce pressure. a bent tube with an open end
If
be inserted in an air duct, as at B in
figure on page 190, with the
open end facing the air current, a pressure due to both the
velocity and static head will be produced in this tube. This is
the total or dynamic head, and the amount can be read on an
attached gauge or manometer tube. If, instead of a bent tube,
a straight tube be inserted as at A, the difference in levels in
the manometer tube will indicate the static head or pressure.
The velocity head or pressure may then be determined by sub-
tracting the static from the total manometer reading.
The an instrument combining
pitot tube as ordinarily used is
the tubes just described, as shown at C, the outer ends being

connected to the two legs of the same manometer. By this
means the subtraction is made automatically and the difference
as shown by the gauge is due to velocity pressure only. These
tubes are usually combined in some form as shown in the follow-
ing figure.
Static TOTfKU
pnss.
'NC HOLtS
rN-N
Care should be taken to have all of the connections made

tight, especiallyon the static side, as a very slight leak here will
cause considerable error. The small holes as shown above in
the static tube should be about 0.02 in. in diameter.
The greatest difficulty to be encountered in air measurements
is in obtaining accurate static pressure readings. Many differ-
ent forms of static tubes have been used, with varying degrees
of accuracy. Some of the more common forms are shown on page
192. Charles H. Treat in a paper on "Measurements of Air in Fan
Work"* gives the results of his efforts to check the accuracy of
some of these forms. He found that tube D was fairly accurate
so long as it was set exactly parallel to the air flow, but the open
tube E held at right angles to the air flow gave readings as much
* Am. Soc. Mech. Engra., Dec. 1912.
191
DIFFERENT FORMS OF STATIC T/P5,
as 50 per cent, too low. The total pressure, or impact tube, as

shown in the sketch representing a pitot tube, will give practically-
true readings, so that if the static readings are too low the corres-
ponding velocity pressure readings will be too great.
The Gebhardt tube uses a static tip which has the end
beveled as at F, in order to avoid the suction of the air flow across
the end of the tube, but Prof. Gebhardt states that further ex-
periments are necessary to show whether any fixed angle is
applicable to all velocities. A static tube as shown at G is fre-
quently used, having slots on the sides of the tube, the Taylor
pitot tube being an adaptation of this form. Mr. Treat found
that a slot % in. long and 0.01 in. or less wide in a 3^-inch tube
gave good results, but it is advisable to cover these slots with

a piece of fine mesh wire cloth.
fi>ii>rrftn>tmfff/f//. '
^>ffi//ni>/t/ijir
'lini I rjjn firm J i i , j-r->
fKPf^fôMCD f=-Of^M or STATIC Tl F'
The tube here shown is the most approved form, and is

static
the one recommended by the A. S. M. E. for fan testing work.
It may be combined with the impact tube to form a pitot tube as
already shown. Charles H. Treat found that a static tube of
192
FAN TESTING
this form with clean holes 0.02 in, in diameter in ^-inch tubing
Vs'j in.thick, gave static readings accurate to within less than one
per cent, of the pressure due to the velocity. A hole Vie in. in
diameter in this tube gave readings considerably off, while a 2-
in. slot Me in. wide gave velocity readings approximately 10 per
cent, too low. Covering the slot with wire cloth improved the re-
sults obtained. A ^-in. slot 0.01 in. or less in width gave fairly
accurate results. The tube of the above standard form gave
fairly accurate results even though as much as five degrees out of
parallel with the air flow.
A very complete series of tests have been made by W. C.
Rowse* in which he compared different forms of pitot tubes with
the readings of a Thomas electric meter. The author used a
pitot tube similar in shape to the one already described, and
found that "of the various forms of static openings in the pitot
tube itself, very small holes in a perfectly smooth surface give
the most accurate results. Slots give erroneous static pressures
and beveled-ended tubes for obtaining static pressures are not
reliable."
A convenient form of gauge for use with low pressure is the
ordinary Ellison differential draft gauge. Mineral seal oil
should be used in the Ellison gauge, but it is so graduated that it
gives the pressure directly in inches of water, without any cor-
rection. The mineral seal oil as ordinarily used for this purpose
has a specific gravity of 0.8284.
The theory of the pitot tube is thoroughly discussed by
Frank H. Kneeland, together with a study of some of the differ-
ent forms, in a paper recently read before the American Society
of Mechanical Engineers.**
Having determined the velocity head as above explained,
the actual velocity may be calculated approximately by means
of the formula
V = 1096.5\^gr
'W^ (84)
where V = velocity of air in ft. per min.

p = pressure in in. of water.
W= weight of air in lbs. per cu. ft.
"Pitot Tubes for Gas Measurement" Am. Soc. Mecb. Engrs., Sept.,
1913.
**"Some experiences with pitot tube on high and low air velocities"
Am. Soc. Mech. Engrs., Dec, 1911.
193
FAN ENGIXEERING BUFFALO FORGE COMPANY
The and pressure will be found
relationships between velocity
on page 16, from which we see that
we have dry air at 70 F.
if
and 29.92 inches barometer, we will have =0.07494, hence W

the approximate formula (84) becomes
V = 4005l/^
Where the pressure p is expressed in inches of water, or
V = 5273t/^
where p expressed in ounces per square inch.
is
The above formulae are only accurate for low pressures, and
should not be used for over 10 inches of water. For more accu-
rate work or for high pressures the following formulae should be
used. As a matter of ready reference the table of velocity for
various pressures as given on page 21 will be found convenient.
Capt. D. W. Taylor in a paper entitled "Experiments with
Ventilating Fans and Pipes"* gives the following exact formula
for the pitot tube:
where Vi= velocity in ft. per second at a point where the

pressure =p in lbs. per sq. ft.
p = pressure in lbs. per sq. ft. at any other point.

2= velocity in ft. per second.
W= weight of air in lbs. per cu. ft. where press. =p
y= ratio between specific heats of air under constant
pressure and constant volume = 1.408.
g = acceleration due to gravity in ft. per second.
The above formula has been presented in a simplified form
by Frank H. Kneeland** as follows:
Vi=4046.16A/^^^^^(l-0.355k + 0.202k2-0.137k3) (86)
where
=(^-)
The values given above are for a temperature of 70 F., a
barometric pressure of 29.92 inches, and a humidity of 70 per
cent.
Society of Naval Arch, and Marine Engrs., 1905, p. 35.

"Some experiences with pitot tube on high and low air velocities'
194
FAN TESTING
From the preceding it is seen that the velocity at 70 F. and
29.92 inches barometer due to one-inch pressure is 4005 feet per
minute and the velocity at any other pressure may be deter-
mined from the above relation. That is, the velocity varies as
the square root of the pressure. For any other temperatures
the velocity may be found by inserting the proper values of
W in formula (84), or from the ratio of the absolute tem-
peratures or barometric pressure, since at constant pressure
the velocity will vary directly as the square root of the absolute
temperatures and inversely as the square root of the barometric
pressure.
These formulae maybe considered sufficiently accurate
say up to 6000 feet per minute. Above
for ordinary velocities,
that velocity and for very accurate work, various corrections
should be made. These corrections, based on the experiments
of Capt. D. W, Taylor are discussed by Mr. Kneeland in the paper
already referred to.
Use of the Pitot Tube in an Air Duct

For fan testing or in ventilation work the pitot tube may be
used to determine the velocity, and hence the quantity, of air
flowing through a duct or pipe. The tube should be inserted at
a point where the duct is straight and the flow undisturbed. In
testing a fan the pitot tube should be placed from 10 to 20
diameters from the fan outlet with the point directly facing the
blast. The air pipe should be the same diameter as the fan
outlet.
The velocity pressure as shown with the tube in the center of
the duct will be higher than the average, and will vary at differ-
ent points from the center to the sides of the duct. In order to
obtain the true or average pressure it is necessary to multiply
the velocity pressure reading obtained at the center by the proper
coefficient. Various authorities give a coefficient of from 0,81
to 0.82 for circular pipes, by which the velocity pressure readings
taken at the center of the pipe should be multiplied to obtain the
corrected average pressure. Consequently the velocity based
on the observed pressure readings may be multiplied by the co-
efficient 0,91 toobtain the corrected average velocity.
For more accurate work it is better to make a traverse of the
pipe and either determine the coefficient for the case in question
or take the average of all of the readings. Where the duct is
195
rectangular it may be divided into a number of small squares or
rectangles and a reading taken in the center of each. Then the
average of all of the velocities corresponding to these pressures
will give the true velocity in the duct. In case the pipe is round
itsarea should be divided into a number of concentric zones or
rings of equal area, and four readings taken in each area, read-
ings being taken horizontally and vertically across the pipe.
The position of each successive point may be found by divid-
ing each ring into two equal areas and adding one of these to the
sum of the preceding areas. The radius of this resulting area
will locate the desired point.
Expressed by means of a series of formulae, these points may

be found as follows:
(87)
6.2832
+
R2 -4 a3.1416 (a-^2)
(88)
2a-|-(a-^2)
Ra
V 3.1416
(89)
R4 ^
/3a + (a-?-2) (90)
3.1416
Where Ri, R2, etc., = the distance from the center to the points
where the readings should be taken in each successive ring.
a = the area of each zone or ring.
PIPE TRAVERSE FOR PITOT TUBE READINGS

Distance From Center of Pipe to Point of Reading in Per Cent, of Pipe Diameter
in
of of
Equal
Traverse
Readings
Areas
No. No.
FAN TESTING
The location of the points on a traverse where readings should
be taken are shown in the accompanying sketch. The table on
page 196 is based on formulae (87) to (90) for laying out a traverse
and will be found very convenient for that purpose. As an
example of its use we will assume that a traverse is to be made of
a 24-inch pipe, twelve readings to be taken. One reading will
be taken at 0.204 X 24" = 4.9" from the center of the pipe; one
at 0.353X24" = 8.46" from the center, and one at 0.455X24" =
10.92" from the center.
noo tBOO tfOO ZOOO

VELOCITY
An example out a traverse and finding the average
of laying
velocity through a round duct is illustrated in the accompanying
figure drawn from test results. Twelve readings were taken as

shown on the diagram, the points being laid out according to
the table on page 196. The velocities were then computed for
each point and the average velocity for each area plotted as
shown, these points on the lower and upper half of the plot being
the same. A curve drawn through these points indicates the
velocity at the edge and at the center of the pipe, and these
points should be used in calculating the average velocity.
The Anemometer
The anemometer is used in many cases where the velocity
of the air low or extreme accuracy is not required. It is more

is
frequently used to determine the velocity of the air leaving a

register or air vent than for testing a fan, although it may be used
for either purpose. An anemometer should be frequently cali-
brated, and when used in a current of hot air the bearings of the
197
instrument are liable to become dry and the readings affected
by friction. Such an instrument may vary as much as 10 or 20
per cent, from the true reading.
The space over which the velocity is to be measured should
be divided into a considerable number of smaller squares and
the velocity readings taken before each square; the average of
these readings gives the air velocity in the duct or pipe. Another
method frequently used when taking readings before a register
or outlet in wall is to take a series of readings along the two
diagonals of the openings, each reading being taken during an
equal interval of time and similar distances from the center.
A special committee, appointed by the American Society of
Heating and Ventilating Engineers to draft a standard method
for measuring air velocities at supply openings by means of an
anemometer, reported January 23, 1913, as follows:
FIRST: The openings
shall be divided into equal rectangular
areas,no side of which shall be over 10 inches long except-
ing where this would require more than ten readings, in
which case the opening shall be divided into 12 equal areas.

SECOND: Readings are to be taken in every case at the
center of every area.
THIRD Readings are to be
: one-half minute duration, the
anemometer being held at the register base or in the plane
of the opening.
FOURTH: Where the diffusers are used, a total area is to be
computed on the basis of the periphery of the diffuser.

FIFTH: The average of the readings are to be considered
as the average velocity at the opening. Where negative
velocities are found, they are to be deducted in arriving at
the average velocity.
SIXTH:
In computing volume, the net area of the opening is
to be taken, the volume to be considered as the product
of the average velocity and the net area of the opening.
In case the anemometer is held two inches from the register
face, no deduction should be made for the register mesh.
The Orifice
An orifice in connection with some form of testing apparatus

is frequently used for fan testing work, usually as a part of a
permanent testing plant where the air is blown into a large air-
tight box and escapes from the box through an orifice. A co-
efficient of 0.600 applied to the velocity is commonly used with
this apparatus. Professor Durley describes in Vol. 27 of the
Trans. A. S. M. E. a series of tests where various sized orifices
198
FAN TESTING
were attached to the end of a gauging box and a set of coefficients
determined for the different conditions.
The orifice may also be used for measuring the air delivery
in connection with compressed air systems, where the air is
under a pressure of several atmospheres. A convenient form of
apparatus for such use is here shown, in which the air passes
through an enlargement in which the orifice is fastened.
^TMCRMOHCTCH
PKCssuRc C'-uac -"YvT)
SX^X^SS^ S^
"""""^r^f/>/f>l
As shown in the sketch, bafflesshould be provided on each

side of the orifice. The on the
pressure, P, should be taken
leaving side of the chamber, the temperature being taken on
either side. The drop in pressure, pv, between the two sides of
the orifice may be measured either in inches of water or of mer-
cury by means of a manometer connected to the two sides of
the chamber. The inner ends of these tubes should enter well be-
tween the baffle and the orifice plate. Different sizes of orifices
may be used in the same apparatus, according to the pressure
carried, diaphragms with openings from 1 to 2 3^ inches in diam-
eter being a suitable range for a chamber having a 4-inch inlet.
The equation for this apparatus would then be
q = iooai/Ft^ (91)
where =
Q cu. ft. free air per min.
A = area of orifice in sq. ft.
P = absolute pressure in lbs. per sq. in.
T = absolute temperature deg. F.
pT = drop in pressure in inches of mercury.
199
Orifice at End of Pipe
An may be used on
orifice the end of a length of pipe for meas-
uring the air discharged, as shown by Fig. a. The coefficient
for such a case is
0.60
(92)
""V'-(i:)
where C = coefficient of discharge.
Ai =area of pipe.
A2 = area of orifice.
The quantity of air delivered by a given static pressure in
the pipe may be determined by
Q = 1096.5 CA2-v/^ (93)

\ W
where Q = cu. ft. of air per min.
C = above coefficient of discharge.
A2 = area of orifice in sq. ft.
p= static pressure in inches of water in main pipe.

W = weight of in per cu.
air lbs. ft.
For values of W see table on page 17.
0.60
c=
>(5J
FIG. a ORIFICE AT END OF PIPt
0.8Z
A, A2 C=
i^r
FIG b SHORT F\FZ ATTACHED TO END OF
LARGER ?\PE.
200
FAN TESTING
The orificemay be replaced by a short length of pipe as in
Fig. b, in which case the coefficient of discharge becomes
0.82
4-& 2
(94)
The above coefficients have been verified experimentally and

found to be adapted to measurements of air under low pressures.
Short Length of Pipe
A short length of pipe (preferably three diameters long)
connected to a box or plenum chamber into which the fan dis-
charges is frequently used instead of an orifice for fan testing,
and for several reasons makes a better arrangement. It is used
on the outlet of a tight box into which the air is blown by the fan,
the air escaping through the short pipe. The static pressure in
the box is carefully noted, it being a measure of the fan per-
formance. The box leakage, if any, should be determined.
A coefficient of discharge of 0.825 should be applied to the
area of the short pipe to determine the true effective area, or to
the velocity of the air. If required, a traverse may be made of
the pipe with a pitot tube and the coefficient determined for any
special cases.
The quantity of air discharged may be determined by means
of the formula:
Q = 1096.5 CAY^
C = coefficient of discharge.
A = area of pipe in sq. ft.
p = static pressure in inches of water in the plenum
chamber.
W = weight of air in lbs, per cu. ft.
For values of W see the table on page 17.
The Converging Nozzle

A method frequently used in commercial work for fan testing,
or for testing a special fan before its installation, is by means of
a converging nozzle attached directly to the fan outlet. The
pressure produced by the velocity of the air is measured at the
point of discharge by means of a pitot tube, placed at the center
201
CO
I u
CVJ
u
o
z
UJ
00
u
>
\ z
o
D u
o
II liJ
o
z
î
fj5 g> r^
^fflfo
in
CTi o>. QQ 00. 00. CO.
lOdVHOSia JO 1N3IDIJJ300
202
FAN TESTING
of the nozzle outlet. Proper correction must be made accord-
ing to the accompanying curve
of coefficients of discharge for
converging nozzles. This curve is based on coefficients as given
in Merriman's Treatise on Hydraulics.
To illustrate the use of the converging nozzle, we will take

the case of a 40-inch planing-mill exhauster at 1780 R. P. M.,
blowing through a converging nozzle having the inlet and outlet
ends 14 and 9% inches square, with sides sloping at an angle as
here shown. The outlet area of the nozzle would be 0.61 sq. ft.
2fe
^ = 0.0963G = TAN. S^'aO'
The angle ofconvergence of the cone outlet would be 11 and

from the curve on page 202 the corresponding coefficient
is 0.94.
The pressure on the fan will be taken at 3 inches and the ;
corresponding velocity will be 4005 "l/s =6950 ft. per min. At this
velocity through the outlet of 0.61 sq. ft. the fan would handle
4230 A. P. M., but the actual quantity handled will be
4230X0.94=3980 A. P. M.
Coefficients of Discharge for Air Measurements

Various coefficients are used in the calculation of fan per-
formance or in air measurements, and their derivation and appli-
cation will be found fully discussed under their proper heading.
The following summary is given merely as a matter of conven-
ience, but the factors should not be used without first having an
intelligent understanding of their proper application. In case
of special requirements it may be found necessary to modify the
given coefficient accordingly.
203
Coefficients of Discharge for Air Measurements
Coefficient for sharp orifice in thin plate ------ 0.600
Coefficient for orifice at end of pipe - - -
nRI)
0.82
Coef. for short pipe attached to end of larger pipe
For explanation see page 201.
Coefficient for short length of pipe _._ 0.82

(blowing from plenum chamber)
Coefficient for short pipe on outlet of fan (see page 201) - 0.95
Coefficient for round pipe (pitot tube in center) - - - - 0.91
Coefficient for converging nozzle (see curve page 202).
Coefficient for diverging nozzle (see curve page 223).
The quantity of air to be measured may be calculated by

means of the formula
Q= 1096.5 CA-y^
where Q = cu. ft. air per min.

C= coefficient of discharge.
A = area of pipe in sq. ft.
p= static pressure in inches causing flow of air.
For values of the weight of air in pounds per cubic foot for
various atmospheric conditions see the table on page 17.
The coefficients given above are to be applied to velocity,

capacity, or to the effective area of pipe or outlet. The proper
coefficients to be applied to the pressure readings will be the
square of the ones given above. Thus the coefficients for pres-
sure in a round pipe varies from 0.81 to 0.82. These are to be
applied to the pressure readings of the pitot tube when taken
at the center of the pipe or duct. This coefficient for round
pipes is based on test data, but should be decreased for pipes
below 12 inches in diameter and increased by a small amount
for pipes above 24 inches in diameter.
204
SECTION III
FAN CAPACITIES
The following chapter gives the capacity tables and perform-
ance curves for the various styles of Buffalo fans. These are
divided into the following divisions: Planoidal Exhausters,
Planoidal Blowers, Niagara Conoidal Fans, Turbo - Conoidal
Fans, Induced Draft Tables, Miscellaneous fans and blowers.
In each case the corresponding performance curves follow the
capacity tables. The tables show the rated speed, capacity,
and horsepower for fans operating at the different pressures
stated, with the exception of the static pressure tables of the
Niagara and Turbo-Conoidal fans, which give the performance
at other than the rated point.
Use of Performance Curves

In connection with the steel plate and multivane fans, as
well as several other styles, are shown relative performance
curves, based on actual tests. The scale on the lower edge of
each diagram reads per cent, of rated capacity, while the left-
hand margin reads directly in per cent. The capacity curves
show the relative horsepower, efficiency and pressure at any
capacity in per cent, of their respective values at rated capacity.
Thus we see from the diagram on page 214 of Planoidal Exhaust-
ers, if the fan is operated at say 80 per cent, of the rated capac-
ity, the horsepower required will be 87.5 per cent, and the total
pressure 116 per cent, of the rated values as given in the capacity
tables. The efficiency will be 6 per cent, greater than at rated
capacity.
The use of these diagrams for the analysis of fan perform-

ances has been fully covered under the subject of "Fans" (Part
IV, Section I) and their application in the selection of a fan may
be found on page 182, together with the practical examples
explaining the various calculations involved. The relations
between static, velocity, and total pressure will be found on
page 177.
205
Combination Fan, Heater and Engine Tables
A series of tables giving combinations of fan, heater, and
engine for various duties will be found in Part IV, Section VIII,
following the examples on "The Selection of Apparatus. " The air
capacities given are based on the assumption that an average

maximum value for the total pressure, in case of an installation
in a public building such as a school or theatre, will be about one
inch, and for industrial installations about two inches. A series
of heater sizes is given for each fan, and a selection should be

made on the basis of allowable velocity through the clear area.
For public buildings the larger sizes should be used, and for
industrial installations, the smaller. The depth of the heater
willdepend on the temperature range to be cared for, and may
be determined from the heater tables on pages 418 to 431. The
low pressure engines were selected on the assumption of 20 to
25 pounds and the high pressure engines on 80 to 100 pounds
steam pressure at the throttle. The engines are all suitable for
direct connection to the given size of fan.
Pianoidal Type "L" Fan Direct Connected to Class "I'

Vertical Cylinder Below Shaft Engine
206
PLANOIDAL EXHAUSTER CAPACITIES
u
O
_
Is
s 1
U H
U On
H z :
< O o^
J O <^
^ O o
S I
o S
Z Q.
< E
a. H
0.0
a
>
Zj o *>
<:'^
S =
o I
<
u.
u.
00
u.
O
IS
Jo
u
lo-
ll
ooo
oqjq
o "too rood 00 iC-*
CO 'J' Tj<
ooo
COTtO
OOO ooo ooo ooo
OOOO oioo 8SS ooo ooo
ooo OOO
OTf<^ O-i-'l* oooo ooos CÔCiCO .-H|>00 occo
50^00 COTfCO OP3 00 <tooo
t^00 OÎN
lOOCO loooeo MOO) 05 00^
OC^JCO
Q0O5O5
OiO^
iCCOOO
00O(N
iMfO
>XCD
(NOOiN
-ÔCXD cc-<tco (MOO OOt^CO 'COM
(M10 05 (MCDO
(MO IN
Q.O lOOO
CO ooo
OOO
I>.iO00
OOO ooo ooo
lOOO lOOO ooo ooo ooo ooo
fCOXN CO<M rl0C'O o<-<o
u 00O<M lOOt^
.-<(M<N COtÔO
ccooco MO OOCCiO
.-i.-(C<5
eoict^ OCOM
Q. .-|(N^^
>
OSiOOO OtÔ Tjlt^CO (MiOOO
(MOS-^ tÔSCO 00 CO OS ccc COÔ
.-I OS 00
*C0O i0>00 oo
coq p oqp
' M CO OSlMiO
Ot^(M
OOîO 00(MO
.-i(M(M (MCCCC
ooo
rococo
ooo ooo ooo
ooo iOOOO iOOO
CO 00*0 ooo
ooo S8 8S
U UJ 5 05O (N^CO
OOOiM
tcoso
(MOOS
cocot^ iOOSiO ooo coo
(MCOOO
1-1
uE
CO-* CO 00 00 OS 1-HCCt^ oscoo
o lOCOt^ oso^ COTfttO tÔS.-H
rH^(M
z
iOOr(< COOOO
r^coo
)00
)iOfO
00 -^-^
CO coo
lOO-* C0CO--H
<*<.-iOs
OSCS^
cctco
0(MOS
ÔOO
?3o5CO 'OS 00 t^COCO TjtTjICO COCOIM
-J e;
18 88^
So 00 (Mt^
CCOS^
eoOO
ftÔ
Ocot^
CO CO OS
M(MCO
C8IA
z ^
OiMiO
oco-*
ooco cOTjto OlO-H O^'*
iOCO^
OSCO'<}<
ocor^ (NOSt^ O300t>.
^00 o >*'C^.-i ggs t>.coo COCOCO C^^(M(M
(NcOt}< TfOCO
t>;t>;q CO 00 CO 00 xo
c^'ioos' (MO'd d(DiM
(NCO-<** TfCCOO 0(M10 tÔCO
<NC^(M (MCOCO
1= 888 OOl^
888 coxco
888
OOOSi-H
5<s
O-Ô COCO''* CO(NO OC^CO (MO
COOS'* osoc^ osr^to eoôs hc
cooc *** COCOCO eoco(M W(M(N
)-êo r-iox oox OOO OOP OOO
pw ^.M/o ooo ooo
>qos
C^ CO
"ico-H
lO h- O
i^'.
CO -- ^' ^' ^'

qpp
Tf
"^"5
CD OS
,-<
opic looc
-H
irj ^'
(XJ o' t^ d 00 0
< a:
1-1 KMCO *(
^^^ O-Ô
OC0--0
C^C^JCO
lOiMOS COCt)<
irtCCt^
COI'-'J"
e
ooo ooo ooo ooo
MJOrN OOaO <SfO
<SÔn .^.
.-. (S (S(S<S
213
^
PLANOIDAL EXHAUSTER CURVES
Motor Driven Planoidal Type "L" Fan
Left-Hand Bottom Horizontal Discharge Planoidal Type "L" Fan

with Overhung Pulley
216
PLANOIDAL BLOWER CAPACITIES
o
<
uu
>H
< UJ
^%
mO
^'^
go
^S
u
s
^
< IN
J u.
Is
I!
Sio
ca
OO
u^
<
Is
II
u
o
is
si
a. U
> u
O
C5tf
^ <
^1
.J
-J CK
c^ Z
u. %
U
53
U
a.
bu
g|
<
q
-J 03
!^
<
U.
u.
CQ
FAN ENGINEERING BUFFALO FOUGE COMPANY
223
PLANOIDAL BLOWER CURVETS
225
Niagara Conoidal Type "N" Fan Wheel
Motor Driven Niagara Conoidal Type *'N" Fan
226
NIAGARA CONOIDAL FAN CAPACITIES ,
Niagara Conoidal Capacity Tables

The fan capacity tables on pages 228 to 273 refer to the single
inlet Niagara Conoidal fans. It will be noted that there are two
sets of these tables, the first based on total pressures of from %
to 4 inches, and the second set for static pressures. The tables
on pages 228 to 231 give thespeeds, capacities, and horsepowers for
the different sizes when operating approximately at the point of
highest total efficiency. Under these conditions the static
pressure will be 77.5 per cent, of the total pressure as given in
the table. These are termed the rated capacity tables for these
fans, arid are similar in character to those given for the Plan-
oidaland other fans. Double width fans with two inlets give
double the capacities and horsepowers given in the tables.
Thestatic pressure tables on pages 232 to 273 give the capac-
ities,speeds and horsepowers of these fans at static pressures of
0.25 to 2.5 inches of water, and with velocities at the fan
outlet of 1000 to 4000 feet per minute. Thus we have the
performance not only at the point of maximum efficiency, but
at both sides of this point on the performance curve. It will be
noted that the peculiar performance of the Niagara Conoidal
fan gives a wide range of capacities at constant static pressure
with but little variation in speed and but very slight change in
total efficiency. The and 275 show the total
tables on pages 274
efficiency at various pressures and outlet While it is
velocities.
generally advisable to operate a fan at or near its most efficient
point, it may frequently be necessary to make a slight sacrifice
in efficiency in order to meet special conditions.
Particular attention should be used in public building work
to keep the fan outlet velocity about 1800 feet per minute in
order to insure quietness of operation, with a maximum allow-
able velocity of 2200 for such work. For industrial installa-
tions where higher duct velocities are the rule and absolute
quietness of operation is not essential, outlet velocities as high
as 4000 may be used. For practical examples in the use of these
tables see page 186.
227
s
z
o
H fa
Q
Z
o
o
z
o
^
u
o
<
Oi
UJ
uu
Q S
z o
3 or
^<
z ffl

< Qi
U
^
U H
U
z 3
^<
z ea
g|
H O
Z
"-^
"
tf)
Z (S
< ^
Ô
i<
^ >
< f^
Of H
< <
<
NIAGARA CONOIDAL FAN CAPACITIES
u
Q
Z
o 15
o
o
z
o
<
Qi
U
U U
Q ^
z o
^ <
z ffl
u u t/3
a.
*"
<
<
t/3
C^
W3 O
< E
o o
^ z
""
z
u S
a. e:
:r Z
<<
z <
o
<
<
a
a,
ENGINEERING BUFFALO FORGE COMPANY
-J
< oo
z <
o
<
o
a t^
z < !::
<
<
&
<
Q Oi
z u
< H
U
si
< ^
z z
<l
mi
<
9 o
o '^
o
<
1
a
^Q
z z
< <
%-<
<
<
Engineering BUFFALO forge company
as
NIAGARA CONOIDAL FAN CAPACITIEi
f/)
AGINEERING
/
BUFFALO FORGE COMPANY
<
<^
u o
tn
as
< r
z Z
< <
u.
it.
8
<
<
<
c

"-j^ oo5co ooiM i^"-; owoo t^cciN
uiio
,1^
loiocd cdcot>^ tâSd
,-t^^
d^<N -^cdoJ
,-i^,-( ,-i.-<r-i C<(N(N WC^CS
-^050
050000
CO^-*
001>-b
OQOCO
t>-;C0
o>o
tO00
ocoeo
0<DCO
^^^ ^^'J< ^^^ ^t^ -^^^
05W0 fÔM 0 05^> <0 05 p0.

Oi-H^' i-<(N(N (Ncieo cCTf'"** cdr>^<
COiOO TfOJiO COOiO ^^05 OOOJO >OOiC

^êo
*** C0CSJ(N
-"TTf"*!
OiC^T-^
<*<<*<}
.-HÔ
'<<'fTt<
OO^
Tf<f<*'
i-hi-hC^
^f*
oJdd c.-<'^' ^'wed *iot>; oi'*'
ooco
?o
^
* ÔO oco-*
WOO'*
OOOS
*
<*"*<
05W
roccco
Tfeoo
0000
eocoec
eo-<*oo
000000
coeoeo
-ôqc
osoo
*<
050 U5'*-<<< e*5<N-*

^_ CO 9 1^ o^ ". ^ '-^ Oi M ;q 05 o OS CO ^_
*_ 0(N
t^t^; tJt^t-J x'odoo ' OJ OS oi d ^' M * d 22^
lO^ OO-^tOS O-Ô OiO -"ifW* coooo

00 t^t^tO OCOtO 000 I0i00 105CC0
CC CO CO CC CO CO CO CO CO CO CO CO CO CO CO CO CO
00 "I CO COX'* oooo -^?o

CO
05 o^^p wop "ôsc^^
lodd ddt>^ tN^t>ô6 oo'ood
coco CO CO CO COCOCO COCOCO CO CO CO CC CC CO CO
lOiOO ooo t^tôo
OOO *
00*0 O0S05 OCOSO -Ô OC** -I
OS0505 05X
5)?IW WMM
XX05
(NCSM
OS050
(NiNCO
O Ol
CCCOCO
CO
COCOCO CO
OM* O <M OS O
OC^X
X
CCCC^
o

<-" c
oo <(N WC^(N CO * * o t^ X '.C c

2
ddd OOO OOO OOO OOO OOO oo.
OOO OOO
eo'ro
CO '.CO
OOO
co-.co
OOO
cocco
OOp OOO
c<cm cceo
osxxr^(C m** COM oxt^ tc*co
r^xoi o
iioo cc*<D xgs coct* c^
m5ico COCOCO
sss sss S| sss sss ps

ONt III
>00
.
f5'Vi
Otac N<S NNN i>OaO
Ci'^'^
tSNN f^**ir^ f^r^-^
240
tioo coo ot^t^ oooso
<JtiN.-H OOO ÎN-* tÔSCO OOOO

(N(NIM IMCSCS (N(N(N IM(NC^ (NM(N
05t* loeoM Mcsco ccooo

(N(N<N (NiM(N (NNCO TOW* 0C0
<N(N CS(N(N (N<N(N (N(N(N (N(N(N (N(N(N
(NINCO CCM -^-tiO lOCOi t^tÔJ OOi
CCMCO O-*^ oos

<NCS(N
OOO
a
u o
E?
O OS
^5
z 2
a. <>
> o
it
< o
<
00 ^0rH OfÔ ON50 coooo
CO coco
OOS 050500
<to 00 iC*
CDCCCD
COOOO
^CO CO CO CO CO CO CO CO CO CO CO COCO
q o oi M t^ "^J. MOO CO Ot}<oj OS-*

r-Jêo
C<M(N
r^coo i^coo 00<OCt^ CO 1^00 cîoo

I0l0>0 Tt* Tft Tf< COCOCO COCÎC^
COCOCO coco CO COCOCO CO CO CO COCOCO COCOCO
CO coo (N05'* CO 00 CO
lO iC co' 00 odd dcot>^
T-ll-tC^l
OcOCO OtîO CÔOO ccr-o

COCOCO COCOCO
r-HÔ
CO CO CO
oo.^
COCOCO
2SS
COCOCO
*2 'I "*. '^ 9 "^ 05Tj<q qoo^,

t-I ci c^' c^" CO CO co-*io id CO 00
OlOJiO COÔO COiOrf cocêo lOOOTtt

00 00 00 00 00 00 00 00 OS
coco
CO oO'^ occc closes owt^ *<*
oc odoJoJ oJd^' (Ncico
0^
<05CO MOO oq;q?o .-oco CONN.
05 IN oc>>o
ccco
fcm COCÔO
coco
coco CO CO CO
oco
CO CO CO
1^
Q U
Z H
< UJ
u o
O CO
> ^
^Q
Z Z
< <
u.
J "^
s ^
o
<:
<
NIAGARA CONOID A L FAN CAPACITIES
a
FAN ENGINEERING BUFFALO FORGE C OMPAN^
O
2 m
<J<-Ht- COOt-
2Si:
OWt*
tî sss
^-"tCO coo>
^^5
rîoio Tti-^Tj^ loict* ooosec Sfc

t3j
5 ^'i-'* ^^^ Trtt"* "TfTfio
"^SS
rt"* M0S5O O?C00 0<*0> C-*. C^*.**. '
do
iszi
^^'(N
-ri .iri M co-*o
,-<i-H^H
t>ô6oJ
,_i,-(,-i
r:Ss
c^cMW SJJ^S
coco^
t^co
cow
lotw
wcceo
coec*
cocow i2l2S
eoccco 222
^2;* SSS
222
gw* ooo xoso e^t-M 0500* T.;

tôool. oioo -^^'^ |2 2?::^ ^
Q locoeo Nr^C^ CCeCiO tÔO IN iC OS ^ 0>

uj S"" M<NC^ ^c^'N c^=^" f^^ '*'^
la i
I
CI ClTfiO OCMt^ i/5<-i'
;: 28 S2^ ^^2 2|g g gg

Cz
a. '>.
5;3 ^ So g<N ^X. t^^^. -
)<** ^*oio cot^r^ ccOi'
it !* tôsio
J55S& ^X^ 2SS S?2^ 05<NCC

OCjW
Or^M. 00
(N(N(N CCW'* lOtOO t-OT'd "f^SS ^M
i>.t^t^ acc^ ^coos -I'^t^

t^t^t^ S00 SoOQO 0S0>0>
iooo (NOO OMO> ooo-<

.2 -5
8 oc^-* C4kOt* OWX> occw ecec(N
ooo C4(NCJ
odd <6d<6 OOO OOO ooo ddd
OOO OOO OOO OOO

(N--0
OOO oco
--om
ooo
O00< lOiN'- ecoo> OCCQO
OOOM ;CO><N
N0 0^
T*<tÔ
ciooo (NiOOO
9-3 * xcco>
5^ s.

sss oeo Ov
o >Ct>iOO
O
riM
li^
262
<0 OOiN W<*00 iiOXi ^05 OJ :C
X> tOCOtO tOiO>0 U3-^-^ rliTf^ yfi-^'^

(N<N (N(N<N (N(N<N (MC^N M(NN ^^^
ecêo N. COI^W 0050
tÔJ<N CC t^ rfy-i
!N(N(N (NM(N MiNiM C^MCl
n c^.osoc '^!0"5 "^."5* omos ôqr^ oji-h

< 'H
CO
MoicO
cccow
rjJiO
coeoco
t>;o6oJ
coccoo
'iN0
-^Tt"-^
O'lo'^'
ooc
OCC'<f'
cot-oo
o u
Z H ooo coo
>OJ '-oec-* "*< .'^^ (Nrrio (Noj
O 00
(N<N (N(N(N (N(NeO MM rt W^ TfJoS Sti
3g :^
OO
SSS ;=;SJi:
22 3!^
*'5* z;"^<*
OSO50C
osooos
000000
o(n
o>oo
oco
o
o>eo o>oco OOJQO '^CJ'-< ostooo cot<-io lo

^ oo
H
^Q --.C^l

M(NM
'-hm* c6n-^
MCl^l
lo'r^'od
(N(N(N
oj'^'ifl
?1MM
o'o-^'
^TJlo
o6
O
Z Z
< <
u. . ;3IS2 ;;J5^ If?M'? co<N(m mw*!" r*OT< oj
0000 QOl>^ t>.t^N- t^t^h- t^r^t^ n-qooc oo

9
o>-*^ r^t^ Cecoo r^cvj OC4IO lO
,-i^rt rt^rt C^04C4 wcieo w

<
0^ "i^?
'0"l
rîo
OiOiO
lO'^Tt
lOiOiO
w*io
I000
lOh-o ^têc
<:
ifliC DCOt^
o
<:
3RSi22S
0
Oe^'*
'poio
(NiOt^
côo
OM<
omoj
OJCÎOO
ooo^h
W(N oo<
ô<
'111'l'^^. ^^.^ cocow 00^^ locot^ oqcB(
ooo ood do'd ddd ddd d<oci dd'
ill ill "^ -S ^ S - S2

ii sSS tofoto ^^
^^^ usioio
"- ||2 $2
t^t^t- oooooi So 1-!mw
III
râe

IP iNf<5^

(NN
i>oao
r<sr

(S(Sr<
pg
Ofs^
rur^r^
l
oaoo
fsnn
263
Z u
O
(/}
< r
o u
_ z
& e^
bo
< o
Z <
o
<
o
a.
J 0^
ia|
E?
<r
z z
u.
^
< o
o <
<
<
a
<
a
ot*" oidci coo't>.' oJ--J>o zi^^

t>.t>. t-OOOO QOOOOO 000505 OO'-'
^(N 0t^0 lOCÎi-H OSOOtO lOMW ccc;

CîN
(N(N
(N^
<N(NC^
I i-Hi-Hi-H
(N<N(N
ooo
CM (N (N
ooo
(N(N(N
ooc
(N(NC
i-it>.o ro0(N "S'-ioo ^'^.'^ P*.^ '^.^.'R

id doc oidiN eodto odo"d rjodd dt""^
0U50 iOitO i00 Ot^t 0000O5 0>-<CM
p eqco?o 05<No oO'*'*. '^.*^.'~! *.^9

côoo
*^.''?'*
id cdt>ô6 oii-H(N c6t6t^ 05<-Htf5 ojgj;-^
tH tJ^'^tJI rlt00 lOfJiO 000 tÔOW OOCM
oco "NfNfN iM.-ioq pi>cq icoot^: --i^.M P<-<

cdt^ odo'd -hwco iddoo ocmoo id cm .-I i-Icm
coco COCO"* l<'<i<-<*< Tj<Tl<Tji C00 COt^W 050
r
270
o
1
<
>
a!
2
z
^^
o ^
O Z
<
8^ ooo
ocO?i 666
0(N (NOO 0<NC5
28
OOO OC<3 O3!N00 CCC(N
cot* OOCXiX XCSOS
Ho. c^cscî c^CMC^ ecwcc cow CO
o o
Z H
8g
< <
<
1N30 U2d
276

TURBO-CONOIDAL FAN CAPACITIES
Turbo-Conoidal Capacity Tables
For high speed, high efficiency fans suitable for direct con-
nection to motors or turbines, see Turbo-Conoidal capacity tables
on pages 278 to 319 inclusive. These fans have speeds for cor-
responding capacities and pressures which are nearly double
those of the Niagara Conoidal of the same size. The dimensions
of the housing are identical with those of the Niagara Conoidal.
Complete and separate tables of capacities, speeds and horse-
powers at various static pressures and outlet velocities are given
for each size of single inlet fan as in the Niagara Conoidal
tables. This enables the engineer to select a fan for a fixed
direct connection speed and for any condition of static pressure
and capacity.
It will be noted from the performance curves on page 320
that the pressure builds up rapidly with decreased capacity and
increased resistance. In this respect it is in direct contrast with
pressure capacity characteristic of the Niagara Conoidal.
Iôr public building work where extreme quietness of opera-
tion is essential the following may be taken as conservative
conditions of operation of the Turbo-Conoidal fans:

At 1 in. static pressure, 1600 outlet velocity.
At ^ in. static pressure, 1800 outlet velocity.
At ]/2 in. static pressure, 2000 outlet velocity.
At 3^ in. static pressure, 2200 outlet velocity.
For exhausting and for systems blowing through heaters,
may be considerably increased.
these velocities
For industrial work any desired outlet velocity may be used
for static pressure up to 6 or 7 inches.
Double width fans with two inlets give double the capacities
and horsepowers given in the tables.
277
cineo
^^Tf
c^Tfx>
00 -O'*
'-"*
MOC^l
t-ceo
^XX
-.coJCM
CC^
loc^jco
CO C3 ^ ^^^ T}<iO0 0-.s;c '^-^l^ t^
WCÎC^ WC^M WCâC^l M!NC1 C^WI^
NCîN WMeC
OCMC-l T}i w o ^j * ;c X c c^ Tf X >*<

Ceoic t^ O CO OS C^ -C CS M OJ
rj-Nco
<J<
cccc-^
C^ CM IN
^^f
C^ M
C^l
--C
lO lO lO
C^J
:c t^
M C^ M OJ -.C
(N
--D
cc oi o IN << r*
' ^' N c4 (N N I
CtX CC * -^ X O M Tf C N X O M
N "3 XN
-i.X
N X X
Tf Ci LO C-l0 0 CO
CiOO
NM N N CC CO CO Tf -^ Tf 0 C
IN (N M N ?l N N M M M !N
C^l I
)0 eot^'t
|00 0>'-|'*
Ir^ ^' (N N
C^!CN
cox OSIN-.COS
-Ô-^ NXO CO^X
OWX
CO-XO
t^XX XXOS C5C50 COC
N N N N N CI
-H NricO
CI CI N
COfJ*
N CI CJ i
CI CI CO
FAN ExNGINEE RING BUFFALO FORGE COMPANY
<N c^4iNiM M<Neo mccco TjiTjiiO 0D <0
O5C0t^ (NCOO t>.t>.I^ t^rô MOOJ OJt^co

ooo < w*iri tôji C0O00 OlN-<*< CC'""
~
cor^r- tt>-r>. t>-t>.oo oooow a
288
^ -H(NC<1 (NCOCO MCOCO Tjl Tji' Tji
t>-050 MO00 0(NiC

l^l>b. 000000 00 05 05
00iC(N
Tjjjqoq
CO CO CO CO CO ^
r-osco r^-'^fN 050505 ososcî Tt<tô coosim t^'-oo

^(M-l"
<*'**
lOtÔS
TfTfTt*
0(N-<J<
lo^tiio
X^
lOioto
COiOOO
irxcir!
CMiO
r^t^t^
r^Moo
toooo
ZU
t^CÎ-i -(COiO QOrHTjt OlTftO t^TfCO <N(NtJ< I^Ôt*
q^(N coitio Doqo (N-^ kot-oj '^wio t^<N>:
^,h',-I ^^"^^ r-JrHr-t (N m' (N (N (N <N CO CO CO COrJ"'T^'
co^^'* 05co^ cocooi <M<N-^ t^noi >ooco Oicor^

0CO5 00O(N
So iMfNCO
C^T}<X!
CCCOfO CO'*'^
lOtÔS
^Tj^rji
i-iTt<:C
iCdO
o: (M-*
iOCO
:OCÎt^
'Ot^t-
u^
os
WX-*
tôoos
cocoeo
o-i(N
ccor^
eo'i'io
oeot^
tôoOJ
(Noo3
^<N'<*<
(N-ho
ooop
o-Ô
ci."-;
a, ^^^' ^^^'
>'^ rt'^,-; (NciiM' im'wco* coco'^
bQz
t^CÔS GiOiOi OX-^t^ OCOOi "ÔCC iMt^CD lOOSCO
COiO'i
^^^ XO'M lOtÔS OlTf'X) OSfN-* t-ÔiOX 10050
luu: -hM'M <N(NC^ COCOCO CCf* Tf'I'iO 10CC50
r
r
TURBO-CO NOIDAL FAN CAPACITIES
oq-<c<5 o^ 00 "<; 05 q
00 00 OS ^ oi oi CO Tj( o
opo
tÔOOS
wô
O'-iW
^^<
WîO osMt^
I00005
"COO
'N^
ectco
lOOO-^
OOO ^rHi-H i-l-l^ ^^1 C^CIN W(NM
Tfiom iooc t^tôc 00OSO3 oo.
!tJ<00 t^
'cot- C5N0
r>;Oeo
CO CO CO CO ^ '
|oo tfi<c<o r-t-oo ooojo om'co
51^00
oo-^
tôoos ocoo ooô ooo> ôo ^(N
Mco'* ccrôo oir-ic^ -"tioco oco^ ntoo)
050S05 OiOiOi 05 05 OS OS OO OOO O .^ loooec osoot- oo^D eoco^ t*eo^ i-tos

*.'~:^. "^'^^ ^."9*? '^. "?* ^. ^ ôstj< OSes'-"
i-I(NCsi (Noico' 00 coco' ^"rJ<Tji lOiOcO ^(Ct^ tÔOO
TjtcO^ OOSiM
lOCOQO
COi-icO
OS-HIN
OCO^ t^T}<0 CCO-H OS'tOO
INCO-* TfiCr^ 00O(N COiOt^ OOOJiO
t>.t>.t>. N.t-t>. t>.OOQC 000000 OOOSOS OSOSOS OSOO
^^^ dc^N wcoeo eo^'* ôio icoo tâoi
OSiNiC
COOOS
t^MiO
^
OC^JCO
- -'Ô
r^TfO CCCOOS
<cco<
OS-^M
CCMO cCiO^ os<o
^^CC- TJ<COOO OS^CO '*QOC^
t>-h>.t^ t>.t--t>. tÔC 000000 OOOSOS OSOSO
'2J
>OM n!9 C3100 lONO oo(N loosio 000^
-' rt<tO00 OC^0 t-OCO tOOS(N lOOON COMC4
IBJOl JOJ ppv > ^^-^ (NlN(N MCOCO eOeO't ^TfiO xo^t^
U!i4/ J3d Jiy

000
oscor^
OC
*00C^ coo-* ^lOOS
5jt^^
(N
toe ^ ... OSCOt^
^ OSCOt^
C0-<<
01.-IC
ld'no-|f3BdB3| iOiO0 Ot
uivv J3d Id
p

N
;s
(S<SM (SN(S (NNN r5r<Jf*S
291
Kg
*"
E
Q. E
o z
2:
^^
Z
O
^t*t*
OW0
MOM
OJ'^Ô
^^Tji
OM_0q
NO^
''5CÔ>
coot*
<D-Vr-i
N.^M
o>t*o
t>^t>^t>-' b-'oo'oi OSO'O ^'im'(N MT}<d iCt>.'cJ
t* lOcoM ooor* tôoo (Ncot* ocot* octêo

S; gS SoS SS 5 ^22 21:^
f
u
U
a.
> <^
boz
?^
ei.
*
rn
a.
a.
w3
f*5
ioaio <N. ilMCO 0t>.0 CS<0 'fiOOi
M

(N-.C(N OS05M OStÔS iOUSO oeco
t>.oqc^ "5^'* *^( '*Ot>; "^-lOJ t>.;cio <jii*<0
choice cc'cc-** -^'loo dt-^r>^ ooaJcJ d-IiN to'^'o
000505 Or-KN CCt-QO

lotôo
050'-i
ooOM <ror^
(Not0 CCt^W
CO ro cc '*<'* ^ îO0 OtOiO coo
1
E
05CD-.0
osc^jio
XM05
oqcô
O-*--
q-<tc5
COOt^
"ô/o
OiOC0
'-.oc'O
O
w^05 oON;r>.
^"mm" (n'mcc ^'^'^'
dcc t>^t>io6 05 <-<c<icc
(/3
riCOiO tÔ-H COiCX

I400 COtÔO 050^ cotôo o'M co<^lO
CO CO CO CO ^^ ^^* îfliO iOoC
(MC0 05
OSOOt^
(N !N' Co' CO CO ' <1<00 t^t* XOiO d ^ m'
05t^C <0>Ci *iCt* ON-* tÔr^
-H C^ CO ># "5 'X> t^05 '^(NCO "I'lOCO oso ^cc-*
CO CO CO CO CO CO CO CO CO ^^^ Ttt-*-* 'J'Tj'iO coo
OiCOO CO500
05 *q ^ *-^ OOS
^^.-1 N<N(N cocoec ^^yci id d t>^ bJ 00 os" 6d^'
t
CO '^CO
OSO'^
NCOCO
< o Kfi
-^
%l
:st
!<N-H 00<N 000 C^lOO 0(N>0 05ICO

)(N-* CDOOO <Ncr^ OCO OSNiO
IBjox JOj ppv s ,^^ 1-1 c^ N(N<N eocoeo CO <<* TfiOiC
OOO OOQ ooo ooo ooo ooo

o>iOO o --r^M ooeooo
coo>i<
oo
iCO
uivv J9d Jiv <t^00
lOiOtO OOOSOJ OO-" (NM *iCCO
jj-noXpBdBO O'HC^
C0-*iO (CtÔO CCtÔO 05O
c^coco
oo
o o
e oo
o
Nfs-"*

- >
^roe
fo-*i
-r<s <sr<s
OOO'O
^ '~;*.^. ô*"! '^^''^^^ '-;''5 05 -ôsio ^0'*
CO ^'^'"3 otô6 oJO'-H eoi<io t>-'o6o Noos
rt ^,-i,-i ,-n-(rt i-iC^(N <N(N(N (N<NC5 CCWCO
O 0^t* 00O5O (MOOO t^jo-it

050t>-
-êccs
00050
.-i,-io
r-lCOlO
CO COt^t^ (^t^t^ t^t^t>.. r^tôo oooo
(N f^.
5i P 00 05 OS lorô ôoco t*oo
^' ^' cd
iO0-<*< '^COW (NCCCO

lOioco corôo i-i(NCO Tt<iOCO t-05i-i
CD CO CO CO CO CO tî^t^ tî^t^ t*tôo
;C CO CO 00 05 "H * O * "5 CO
~^
lOOOl dO'-t ÎMM TjJidcd t^CCd 1-H C^' TJH lO 00 ,-h"

i-H.-H(N (N<NiM C^<NCO
Ot^-<t .-hOOO CO-*-"* COMIN .-IIM.-C (NN-<t

Tt<iOCO t>-00O5 0<-iC^ CCOt-
COCCCO CCCOCO t^t^t- t*N.l^
a
*
<
t
u
Q i-
< S
H CO
oo
CO t-M^ i-io
I00
cDiNoo
Ot^t>-
loôo
0OO5OS
ceco
Oi-iC^
coc^co
C^TflO
CC
U5
M'<l">r
lOOiO io<oo ooio oioo CCCO
a
o
H
H Ofi
< u
< f)
^ z
a N
z<
^ %
St;
o <
z
o
(J
6
00
=3
a,

&
T URBO-CONOIDAL FAN CAPACITIES
1
a
a
f<9
320
INDUCED DRAFT FANS
Induced Draft Tables
Induced draft tables are given for both Planoidal and Niag-
ara Conoidal fans, with gases at 300 and 550 Fahr. The 300
tables are to be used in case the gases are passed through an
economizer. These tables give the boiler horsepower that will
be served, together wilji the speed, cubic feet of gases handled
per minute, and the po\^ required to drive the fan, for different
sizes of fans operating at various pressures measured at the
breeching of the boiler. Thus, if we operate a 100-inch Planoidal
fan handling gases at 550 at 355 R. P. M., and 0.75 inch static
pressure, we will be able to develop 670 boiler H. P. In case we
speed up to 502 R. P. M., with a pressure of 1.5 inches, we will be
able to develop 950 boiler H. P., or an increase of 41 per cent.
The power required to drive the fan will increase from 7.06
H. P. to 20.0 H. P.
Special Narrow Induced Draft Fans

The tables of special steel plate fans to be direct connected to
Buffalo steam engines and used for induced draft work will be
found especially convenient when selecting apparatus for this
purpose. The first column gives a number, which refers to that
particular combination of engine and fan. It will be noticed
that these are narrow, tall fans, of the special high efficiency
type, operating at such a speed as to make them suitable for
direct connection to steam engines.
These fans are to be operated with a static pressure of 1.69
inches at the breeching of the boiler, with gases at 550, and
develop 50 per cent, overload on the boiler. These combina-
tions are so selected that a peak load on the boiler of at least
100 per cent, may be carried for a limited time, and approxi-
mately 50 per cent, overload all of the time. The engines are to
be operated at their normal rated cut-off and at the speed indi-
cated, the steam pressure required being indicated in the third
column. The above is based on the assumption that under
average conditions a pressure of 0.75 in. at the up-take will be
required, when operating at normal rated capacity.
321
O
O
O
to
PLANOIDAL INDUCED DRAFT FAN CAPACITIES
O
O
in
ii
o z
u <
is
u.
u.
:^
CQ
M
PLANOIDAL INDUCED DRAIT FAN CAPACITIES

O
O o^ t>A c^. >o O) 00 'X>
QOCJd
00 o>^
r-eoos;
ocsco

000
csco
000 000
coo CO 000
O X t^ ^
C; Xi^
OCOC'l
CM<N
100 ^*
w n
iC o
-HOOO
(NCC*
> Qo ^cio ootô
CO po ro
b e3 ^
O .-"T^'OOO
^_ ,.-w. ^
000
www-
1000
C^JOS^
OiOO
T}MO
lOiOO 000
05 000
ioo>o
O
OiOO
C!CU5
o eoxcc têcx
C> C) ccccco o '^ o
r= C-l
:
el
H <
t =
io
INM
u.
looo
000 000
t^ CstÔ ^MCO
cî r>.
o o
tôw
>. coot^
MOX
iO ooo
t^
*co
lOOL':
lOO'O
coo toco
Ci0
lOt^X CM-*
,-H t^ lO "3 00 N O^ "^ *?

lOtÔ Wcod Mt^'-j
COW*
,-i,-H(N (NCMCO
ooo ooo 000 000

ooo COOÔ 000 000
Tf^cD OiOO 000
.
OJ005 ooooio Oi-to cog
(NOOO 05CDO ^00 .-<(NTf CCO'*
C0"*iO tôio CSCDCC
iO>t^ 0000 i-iCC*
OCOO 000 Cs Oeoio OOi<N CCCO'^ OKN

*oo t-,TjicO
(N(N(N
"-lOOS
(N(N.-i
oot^<o
i-Hi-i,-!
OOiGO
ooo OOiO 000

00000
OiOO
O'*'^
coo
O'-i'*
(NO! CO '* O I> Oi --H CO CDOO"-! OX<M COOTf
,-i.-t(N (M(NCO CO-*-*
00
c <
OOOCO lOCOiO * I O lO N
CO t- q ^_ CO CO (CO CO
,-H ^' im' (m' CO Tji
000 000 000 COO coo OOO 000

lOXfO 000 ooo 000
.-i!N03 OCO (NOOO
(MM05
COX-* ÔO'^ OX:t^
00000 .-H035 COiC i-KN'*'
(NCO-* COtÔS
(Ncoro
iCMiM -^-lO oO'-;'-; (M--IO

COOO NCD^ i00 -HOOlO CO'-'O
OJOOt^ OiOiO '^^CO CO(N<N C^HM(N
i00>0 lOOO lOOiO 000 in 00(NOO

loioo
CCCCOO
000
(NOOiO
00!NiO OS'Ôi rI<t>"-H 00 CD
^r^^ ,-((N(N C0-*CO COOGO
(NINCSI COCOCO
C (M IN CO ^ t^ M 0 Tft ^ CO
t^05^ fOOOCO CO <N (N CO ^_ 1^
"^ ,^^(N' lO <C t^ 06 OJ O (NCO-
000 CO000O CD000

t>-0
000 000 000
(NlOO 10 M o r^ CO *
50
c^ô .-iCOt^ -'*C^ CO 'too coo W (N'-iO
(NC0->1< if5CDI> OSC^CD OiOO COC0(3S l>iOC0
OCDt^
(NMCO CO-*-*
*iOC0 OSOOt* (NI>CO 10 05

O 10 -H
iOtHtJi
00 "N 00
COCOM
10 (NO
IN(N(N
IU50 0>0>0 OiOO

000
000
COCO-*
000 10 100
lOS^* 00000 CO t'Jl'Jit
It-((N NCOiO cOt-OJ
TTlftlA >OtN00 v
326
NIAGARA CONOIDAL INDUCED DRAFT FAN CAPACITIES
O
O
I
.a
n
11
<
Of
Q
Q
U
CQ
111
|5
SPECIAL STEEL PLATE INDUCED DRAFT FAN CAPACITIES
HI
Us
a u on
i-.s
Boiler
P.
Overload
Developed
50%
H.
Planoidal Type "L" Fan Direct Connected to Double-
Vertical, Double=Acting Engine
Double Turbo-Conoidal Type "T" Fan
331
BUFFALO CONE WHEELS
The cone wheel is a style of fan frequently used where the

resistance to be overcome is moderate, or where it is merely
required to exhaust the air from a chamber or to exhaust from

a series of ducts into an attic or out of doors. It may be used
in many cases where a disk fan is ordinarily installed, and will
give better efficiency than the latter. The efficiency is, however,
lower than that obtainable with a wheel enclosed in a housing,
so that it is generally advisable to use a standard steel plate or
multivane fan. This is especially true if it is necessary to
operate against any considerable resistance.
The table on page 333 gives the cubic feet of air per revolution
at free delivery, as well as the performance under various pres-
sures. The air H. P. under free delivery may be calculated by
H. P. =0.00026 X cap. X press, corresponding to the peripheral
velocity expressed in inches. This should be increased to cover
belt and bearing losses.
332
CONE WHEELS
,i4 r5^(N co-îo t^c
ooo ooo ooc

oioic o*5o ooo
CCCOO ooo-* 00 CC (N (M 05 --5 O 00 >5
(N(NM <1<iOt* Ot^
tÔOfM
ÔO
t/3 OC 05 CC 05 ICIMQO
z COCOfN
o :j I
H C5 05 C oqq Lc (>i --C

Q ^ (N 05 i6 --I oo
z (M 00 iOt--X
o
coo ooo OOO ooo coo
OiOiO ooo lOOO
o O Ci O cooco o> 00 a o; <M c^
^
z t^ 00
CO lO t^
<N
O OJ O CiCO>-<
.-icêo
O "CO
rr lo ci;
i-i
05 01
lO
*
o O 00 :c iC
IM
XO
'^ ooc-*
eocsO
OOCC 00 t--
(N.-lr-l
CO<N<N
^ ttl
U H
(M_ t^ CO --; q
00 qqiq ^Gc-* o<Na>
Oio ^ --< (N CO CO -^ >dt>05
^ < ooo ooo

< CQ ooo ooo ooo lOiCO ooo
looo C0!NO
looci lomo t>-00iO
Of (MtÔO COiÔ
X o CO(N<N
coo
(/}
M Tj. -^ OC O CO iC
co-*io tô^
a
ii I
mCC-X)
SiAa>
UJ-^CC
t^XX
-^ct^
COCO(N
(MINOS
IOCC05
(NC^1.-i
"^-tS
t-ioec
11^
'^SJS
r-coi
1-I.-I
I
SiS C5CCIO (N'*'* co-ô

^^ oo^
HI Er^, ,^
t-rHio ooio--; q^.q
1
P Z
(NC4C0 00 -^d odc'fN
B < oooCO ooo

lO lO c^.o
ooo
coo ooo
iC o lO
i-i:c .--
X o lo <M :r '^ cjc'-co actX)
CCCC!N * X 'H
X w lO CCXO
c^co-*
oS^ xQM
ooco oco cccq C'~q
^5 XO^ r^St-
(NM^
iCTOM
^r^--
oxx
if ^^CO CCMfN r^
u.
CQ
It* T-i-teo ^ c X q lO lO
U. IrH wcêo ôco oico'io
o
ooo ooo ooo COO COO
OOOCO CXO iCCO C>-';C CCJO
w?oS o-îM --- ccx C;5"C^
uotô ^cox coa.^ c^c.
?5co^
COXIN
05(NX
CO COIN
Ot*X 05D^ N<-" >000

^
gJlot- OCO-H (NOC0 OCW^
Ot*7H
^,^<N C0-*O
JiV Jd *"3
ovOM OCT "ONin NOW

N
3ZIS, r*ir*i^ Ti'0 t>0O v
333
BUFFALO " B " VOLUME BLOWERS AND

EXHAUSTERS (TYPES BB AND BE)
"B" Volume Blower "B" Volume Exhauster
The table on page 335 shows the range of pressures

and capac-
ities forwhich these blowers and exhausters are designed, either
for producing blast for forges and furnaces, for removing smoke
and fumes in small ventilating installations, or for conveying
dust and refuse from emery and polishing wheels. Data on
application is given in another section. For forge blast a pres-
sure of 3 to 5 ounces at the fan is sufficient, and for removing
smoke a suction of 2 ounces at the fan is usually employed.
For exhaust conveying systems either "B" volume exhausters
or planing -mill exhaust fans may be used according to the
nature of the material handled. With either type extra
heavy blast wheel construction neutralizes the effect of abrasive
material, while acid gases may be handled by blast wheels of
cast iron, lead, copper, monel metal, or other suitable acid re-
sisting material.
334
VOLUME BLOWERS AND EXHAUSTERS
?5Sg
U
OQ
o
z
<
QQ
tfi
U
a
>
H
(/}
E
X
u
zo
< c
^ 2
2^
ml
o
>
BUFFALO STEEL PRESSURE BLOWERS

(TYPE P)
Steel Pressure Blowers are designed for relatively higher pres-

sures and smaller capacities than "B" Volume Blowers and while
they may be used for the same purposes, are intended especially
for supplying blast to cupolas, furnaces, and forges requiring
air pressure of from 6 to 14 ounces per square inch.
Steel Pressure Blowers for pressures as high as 16 ounces are

also built in two stages.
The table on page 337 gives capacities and horsepower

required for these blowers at various speeds and pressures.
Table on page 110 gives special information regarding require-
ments for foundry service, and table on page 338 describes
method of choosing blower and laying out piping in forge shops.
336
STEEL PRESSURE BLOWERS
Vi
2
O
Q
Z 00
O
o
z
(^
o
>
IS
l
u<
Qoa
to: .5
Pi
fiO o ts
O cd
c/}
afrn
35/
t/3p
UJH
-J a.
ug
o
<
u.
u.
ea
u.
o
BUFFALO STEEL PRESSURE BLOWERS (TYPE P)

APPLICATION TO FORGE FIRES
Static Pressure in Ounces
>>
STEEL PRESSURE BLOWERS
33
BUFFALO STEEL PLATE PRESSURE BLOWERS
(TYPE R)
Pulley Driven Motor Driven
These are high efficiency fans, and are usually designed for
pressures up to 14 or 16 ounces for cupola or furnace service,
but may be built for pressures up to two or three pounds per
square inch. The speeds for cupola work are suitable for direct
connected motors, which are usually mounted on a sub-base
built as a part of the fan. In large units the bearings are fre-
quently mounted on independent pedestals, and the fans driven
through flexible couplings, while for smaller sizes the fan wheel
is overhung on the extended motor shaft.
It is possible to obtain, with a fan of this type, higher effi-

ciencies than with standard radial blade steel plate fans, or with
any form of multiblade fan,
CAPACITIES OF BUFFALO STEEL PLATE PRESSURE BLOWERS
%
STANDARD PLANING-MILL EXHAUSTERS
BUFFALO STANDARD REVERSIBLE PLANING-

MILL EXHAUSTERS (TYPE M)
Pulley Driven Motor Driven
Standard Blast-Wheel for Buffalo

Steel Plate Mill Exhauster Blast-Wheel for Stringy Material
341
BUFFALO STANDARD PLANING-MILL

EXHAUSTERS (TYPE M)
These fans are ordinarily used for conveying materials and
for removing shavings and factory refuse in general. Instruc-
tions for application are given in another section. The con-
struction is similar to steel plate ventilating fans, but with special
proportions and of considerably heavier material to stand wear.
The fan wheels are always overhung, and the bearings of extra
size. These fans are made single and double. The table below
gives speed, capacity and horsepower for various pressures for
single fans.
CAPACITIES UNDER NORMAL WORKING CONDITIONS

Total Pressure in Ounces
s
STANDARD PL AN I NGjjH LL EXHAUSTERS
PLANINQ=MILL EXHAUSTERS
BUFFALO SINGLE STANDARD PIPES
aVPE M) WITH DIFFERENT AREA SUCTION
AND VARYING VELOCITIES
Speed and Power Requirements
in Feet per Minute

Exhauster Velocity Through Branch Suction Pipes
4000 4500
3000 3500
u i
^
6.6
1910 1675 3.3 2180 1925 4.72 2450 2140 10.8
79 10 1636 1410 3150 2170 7.60 3540 1485
,
2360 1620 2750 1890 5.1

30 12 113 113 12 3720 2240 9.2 4250 2552 13.50
154 14 I
3190 1940
6.00 3540 1775
1185 2.52 2750 1390 4.02 3150 1570 9.35 4790 1990
1350 3.90 3720 1570 6.25 4250 1780 14.60 6280 1470
154 4890 1775 10.00 5540 2020
1535 6.35
4790 1520 10.6
3720 1185 4250 1350 16.2
5540 1495 6280 1755
4890 1320 7980:1970 25.8
40 201 6200 1530 7080 1730
9.25 6280 1320 13.1
4890 1030 6.25 554011170 19.2
13.30 7980 1458
1201
6200 1140 9.10 7080 1295 19.20 9820: 1640 28.2
45 18 254 254 7650 1370 13.40 8730 1450
314
R9f^n OOî 7 45 7080 1028 11.0
7980 1160 15.85
254 18 5310 1130 15.7 9820 1270 22.30
6550 f6?0 9?fl0:70 8730 11890 1410 31.80
314 314 20 9250'lll0 15.20 10570 1260 22.2
380 22 7920
9820 1040118.7
314 20 6550 11890 1105125.4
380; 22 7920 ??Hididi2.t's''9S''?l 14150|1265 35.5
55 22
452 241
9450 |to,10:6o!nOOO| 9^17^
12550 1105,24.7
118901 940 21.5
9.75
22 7920 630 6.42I 9250 718 13.80 114150 1045 28.1
60 24 452 24 9450 6841 8.60 11000 808 18.90 16500 1100 38.0
26 11100 738 11.10 12900 862
14700 704'l9.4 16500 795I27.8
11100 530 8.3 12900 622 13.4 19200 858 36.6
14950 670 18.3 171001 744 25.8
70 28 616 12800 575 11.0 196001 820;33.2 22100 920 47.4
14700 620,14.3 17100 72022.6
22100' 690 35.6
707 I25000I 730 48.0
80 32 804 804
908 i3iPii!p,SlSiS ,28300 789 57.5
are computed on .he b^js

NOTE-Table,
that th-V'te ^>'
of one collector^ 1
^5
feet of piping incluchng equivalent the tan outlet.
rtirtl.fe'lTerS^'YrirSS^^^^ are ^Q^Pâ ent ^

the length of p.pe to which they
length, in order to determine the
total equivalent opera j
g^^^h
j from which
.^
^^^^ ^^^^
handled will be doublea.

double fans, power and ajr
343
BUFFALO SLOW SPEED REVERSIBLE PLANING-

MILL EXHAUSTERS (TYPE E)
Suction Side Bearing Side
Standard Slow Speed, High Efficiency Slow Speed Blast-Wheel for Stringy
Blast-Wheel Material
344
SLOW SPEED PLANING- MILL EXHAUSTERS
BUFFALO SLOW SPEED PLANING-MILL

EXHAUSTERS (TYPE E)
These fans are of the high efficiency type explained elsewhere,
and while intended for the same purposes as the standard mill
exhausters, require from 15 to 50 per cent, less power. The
speed is also reduced one-third, and the size of the fan is increased
nearly 50 per cent. For cases where refuse fuel is not available
for furnishing power, it is advisable to use this fan, on account
of the decrease in power cost.
CAPACITIES UNDER NORMAL WORKING CONDITIONS

Total Pressure in Ounces
4>
BUFFALO SINGLE SLOW SPEED HIGH EFFICIENCY MILL EXHAUSTERS

(TYPE E) WITH DIFFERENT AREA SUCTION PIPES
AND VARYING VELOCITIES
Speed and Power Requirements
Exhauster
SLOW SPEED PLANING-MILL EXHAUSTERS
1N30 yid
347
FAN ENGINE BRING BUFFALO FORGE COMPANY
BUFFALO DISK FANS (TYPE D)
Pulley Driven
Motor Driven
The ordinary disk or propeller fans should not be used in
con-
nection with a system of piping, but should discharge
directly into
a room, or exhaust from it without obstruction.
Although not
as efficient as a centrifugal cased fan, the
lower first cost, large
air capacity and simplicity of installation
account for the wide
use of fans of this type for ventilating engine
and boiler rooms,
kitchens, restaurants, small theatres, brass
foundries, etc. A
conservative table for the air capacities of actual disk
fan in-
stallations is given below for normal speeds, and
table on page
349 gives capacities and horsepowers for various speeds.
These disk fans are probably more often installed with direct
connected motors than for belt drive, and such outfits have
be-
come standardized for both direct and alternating current.
348
DISK FANS
tîOO ooo OOO ooo
COCCIM
(N l> <* t^ * -1 50 coco
CO 'I* l^ C5 (N CO
FAN ENGINEERING BUFFALO FORG E COMPANY
BUFFALO MULTIBLADE DISK FANS (TYPE CM)
Disk fans with many overlapping blades are better

suited
than standard disk fans for maintaining sufficient
pressure to
overcome a moderate piping resistance or other friction, and are
usually employed for boosters, in mine ventilation,
or for pro-
ducing air flow in cooling towers for condensing
plants. The
casing and bearings are self-contained to facilitate
installation.
The normal speeds, air capacities and horsepowers are given in
the following table:
Size
PROPELLER FANS
BUFFALO PROPELLER FANS (TYPE F)
Motor Driven
Pulley Driven
cent, greater
The propeller fan has a capacity 25 to 30 per
used for the same general
than a disk fan of the same size, is
may be furnished either pulley driven or with
purposes, and
direct connected motor. 48-inch and smaller disk and propeller
fans are made with overhung wheels, so that it is unnecessary
oiling the outer bearing.
to reach between the blades for
Table of capacities and horsepowers at normal speed is given
below.
Size Normal Speed
18 1050
24 800
30 650
36
42
48
'
bg
u
E
-J I
O c^
0. >
O J
li
PROPELLER FANS
PERFORMANCE OF 8-BLADE PROPELLER WHEELS 30 ANGLE
OPERATING AGAINST RESISTANCE
FAN ENGINEERING [B UFFALO FORGE COMPANY
354
ELECTRIC ^LOWERS
BUFFALO ELECTRIC BLOWERS (TYPE FB)
Constant Speed Variable Speed
This type of blower has been developed for furnishing small

volumes of air at pressures from one to three ounces for forge
fires and furnaces, for blowing church organs, and in fact any
purpose for which the various sizes may be applicable, handling

from 60 to 250 cu. ft. per minute.
These fans may be used for exhausting at the same pressures,
and are furnished with motors for 110 or 220 volts, either alter-
nating or direct current. The design of the blast -wheel and
casing is special, similar to the high efficiency steel plate fans,
so as to make the power consumption remarkably low.
No.
of
Blower
BABY CONOIDAL FANS (TYPE IC)
Small motor driven fans for exhaust ventilation are prefer-

able to the use of desk fans, since they provide means for intro-
ducing fresh air, or for positively removing fumes and odors.
They are also used for small drying outfits in connection with
steam or electrical heaters. Standard sizes are arranged for
direct or alternating current motors; fans are of Conoidal type
with housings which are reversible. The following table shows
capacities of some of the smaller stock sizes.
Size
Planoidal Type "L" Fan Wheel
Cowl Ventilator
357
SECTION IV
FAN DIMENSIONS
Included in this section will be found dimensions of various
fans and blowers. Dimensions of Planoidal, Niagara Conoidal
and Turbo-Conoidal fans are given for both full and three-
quarter housing, for top horizontal, bottom horizontal, up and
down discharge. Dimensions are given on pages 382 to 385 for
double width Niagara Conoidal and Turbo-Conoidal fans.
Dimensions of slow speed and standard planing-mill exhausters,

single and double width, will be found on pages 394 to 397,
and on page 398 are given dimensions of steel pressure blowers.
358
PLANOIDAL FAN DIMENSIONS
PLANOIDAL (TYPE L) FANS
OVERHUNG PULLEY
FULL HOUSING TOP HORIZONTAL DISCHARGE
Dimensions in Inches
Sire


OVERHUNG PULLEY
FULL HOUSING DOWN DISCHARGE
Size
PLANOIDAL (TYPE L) EXHAUSTERS
OVERHUNG PULLEY
THREE-QUARTER HOUSING TOP HORIZONTAL DISCHARGE
Size
I<AN EN CWNEERING BUFFALO FORGE COMPANY
OVERHUNG PULLEY
THREE-QUARTER HOUSINGBOTTOM HORIZONTAL DISCHARGE
Size H I M
60 21^ 281^22 5^ 29?Î26 18 22H26K 52% 25% 39% 20
70 25 35 26 J^ 34^^301^ 20 J^ 26^30% 61% 29 45% 21%
80 28^ 40 30 39^135 231^ 30^^;35i^
K 32% 49% 25
90 32^ 45 34^ 44 3^139 M 26 34^140 78 H 36% 53% 27
100 50 37^8 49^1435^ 28^ 37^44% 86% 40 58% 29%
110 39 >i 55 54 3^ 47 J^ 31 M 415,^48^ 95% 44% 63 31%
120 43 60 451^ 59 M 52^ 34 451^ 53 103^ 48% 66% 33%
130 65 49 >^ 63 >^ 561^ 37 49 57^ 112% 52% 73 % 37
140 50 70 53 61 39Jî53 61% 120% 56% 77% 39
ISO 53 J^ 75 56^73^ 65 Ji 421^ 56% 65^ 128% 59% 85% 43
160 57 80 60^178^ 69% 46 60 5^ 701^ 138% 64% 90% 45
170 60 M85 64^83 7^ 74 H 48 H 64^ 751^ 147% 47% 95
180 64^ 90 681^ 88^ 78^ 51 68 H 79% 155% 72% 103 51%
190 95 72 93^ 82 K 54 72 83% 163 % 76 107% 53%
200 100 7534' 98% 87% 56}^ 75% 172 |79% 115% 58

OVERHUNG WHEEL
This Style for 30 to 60-Inch Fans
366
>
a S
S
1 i

I
(A
O
OVERHUNG WHEEL
FULL HOUSING BOTTOM HORIZONTAL DISCHARGE
This Style for 30 to 60Inch Fans
368
>
<
o ^
O n
= Z
H
O E
- =
o
z
tin
;3
o
OVERHUNG WHEEL
FULL HOUSING UP DISCHARGE
" E
1 ^^ ^
r Jr-
J-L-e4^-^pz +
^
u ii I
^^
î
370
>
z I
5
3 o .2
O E "
o u.
OVERHUNG WHEEL
372
>
U
o
<
E
-J ^
uj Q
1^5
iS i
3 I
.2
Si i
>i
O o I
o
E
-J
374
CONOIDAL FAN DIMENSIONS
NIAGARA CONOIDAL (TYPE N) FANS

TURBO-CONOIDAL (TYPE T) FANS
M
Â
L lu
OVERHUNG PULLEY

OVERHUNG PULLEY

H AJ 1

OVERHUNG PULLEY
Size
I
I
3 17>i 157-^ 13% 201? 12 16% 13% 221i 14%

3H 20 18,H 15/gl 24% 14 18% 15 26% 15%
4 21 22 K 21 175^ 27% 16 20% 17 29% 16%
2ZH 25 M 23^ 197^ 31% 18 22%18% 33% 17%
5 2614 28 H 26 ,H 34H 20 24% 19% 36H 18%
28^ 313^ 29 H Hi 38A 21% 26% 21% 40A 18%
6 31^ 34 M 3lii 26^ 41% 23 % 28 % 23 43% 20%
7 36 M 39% 37 H 30^ 48t^ 27 32% 26% 50,% 24
42 45^ 42 3^ 35x'V 55 M 31 ,36% 28% 57% 27
47% 51% 47H 39% 62/b34%40% 31% 64x^5 61 31

52 H 56% 53 44% 69% 38% 44% 34% 71% 65 33
57 %j 62 58A 48% 76^42 49% 38% rSli 73% 37
63 68 63^ 5211 83% 46 153% 41% 85% 79 % 39 %
68% 73 H 68^8 57% 90t^49%58% 45% 93x% 83 41%
73}^ 79 74t% 61%' 97% 53 62% 47% 100% 92 46
78^ 84% 79 H 66^ 104x^6 57 66% 51% 107^ 49%
84 90% 84% 70% 111 60% 71% 54% 114% 107 53%
89^ 96 90iB 75 11711 64% 76% 58% 12111 112 56
72 94 H 101^ 95% 79t^ 1 24% 68 80% 61% 128% 120 60

76 199% 107 1001 8311 1311| 72
1 84% 64% 13551 125 62%
80 105 112% 106 88% 138:^4 75% 88% 67% 142% 128 64
378

OVERHUNG PULLEY
THREE-QUARTER HOUSING TOP HORIZONTAL DISCHARGE
Size

h- A/
^-=-^^-^
Li ^ // / // /
OVERHUNG PULLEY
THREE-QUARTER HOUSING BOTTOM HORIZONTAL DISCHARGE
Sizej A


DOUBLE WIDTH
Size
3 15 î 17>i 11^ 15^ 20 H 14 28 H 13M 38 191^

20 13
2!*
24 M 16^ 32 M 15 44
4 21 22 M
14K ^1 TS 27 M 18: 36 M
16 48
4H 23^ 25 5i 16^ 23 3^ siH 21 40^ 18 ^^4 55

5 28^ 18^ 26 3^ 34 H 23 441^ 19' 59
2S% 31J^ 20i''5 29 H 38 A 25 48 M 21 M 65
6 siy2 3414 22 31 H 41 A 5^ 27 H 52 M23 69 H

7 36 5i 39 M 26 37 H 48 r% 32 60 >i 26 H 77 J^
42 45 H 29^^ 42 3^ 55 y2 36 14 68 M28 M 87
9 47 M 51 ^
33 H 47 H 62 /b 41 76 Ji 31^ 100
10 52 J^ 56 Ji 37 1\ 53 69 H
45 84 Ji 34 3^ 110
11 57 M 62 H
40 H 58 ^ 76^491^ 93 M 38^ 124
12 63 68 44^ 63 5^ 831^ 54 101 K 41 J^ 137 H

13 6SH 73^148^ 68 7-^ 90 A 58 H 110^ 45^ 146
14 73 y2 79 152 tS 74 A 97 3^ 63 118M 47^ 158
15 78 M8434: 55 H 79 H 104 T^ 67 H 126 J^ 51^ 166 K

16 84 90^ 59 K 84^ 111 72 135 K Vs 54: 179
17 89 Ji 63 M 90t'b 11711 76 144 K 5SVh 187 M
18 94 H 101^ 6611 95^ 124 7/^ 80^ 152^ 61^ 194
19 107 70 H 100 H 13111 85 160 M 64^ 202
20 106 112M 74 H 106 138 M 89 H 168 M 67 H 212

DOUBLE WIDTH
ll-
NUGARA CONOIDAL (TYPE N) FANS

M
^j|a.
.-^..^^/f^
I
DOUBLE WIDTH
w3


OVERHUNG WHEEL
/ H k-t/
This Style for No. 3 to No. 6 Fans
CO
^ C/3

OVERHUNG WHEEL

FAN ENGINEERING BUFFALO FORGE COMPANY ]

OVERHUNG WHEEL
390
>
u
a
H <
"" H EQ ^
J "^
^
3 '
i
2
z .2
E g
U3 g
u o ggo
I
S o
U.
o

OVERHUNG WHEEL
392
>
fc CO
<
nils
^^ c
oo -
E I
a S2
C^ :> -
o U^
> t/3
2
E
o o S
I
r
o
O
STANDARD REVERSIBLE PLANING-MILL

EXHAUST FANS (TYPE M)
BOTTOM HORIZONTAL DISCHARGE
Size A B C F
PLANING- MILL EXHAUST FAN DIMENSIONS
STANDARD REVERSIBLE DOUBLE PLANING-MILL

EXHAUST FANS (TYPE M)
Size
SLOW SPEED REVERSIBLE PLANING-MILL

EXHAUST FANS (TYPE E)

Size A B
PLANING-MILL EXHAUST FAN DIMENSIONS
DOUBLE SLOW SPEED REVERSIBLE PLANING-MILL

EXHAUST FANS (TYPE E)
Size
FAX ENGINE ERING B UFFAL O FORGE COMPANY
STEEL PRESSURE BLOWERS (TYPE P)

No. B H K X \
1 3^ SH\ 7H 14 K 7^ 53^1 5A 3H 6K
5H 57"8
411 2 7îlF8:2H
2.7/8
2 I
4 10 i
6H
7H 19 M! 5^! 5A 6 35i 9i^ 3^' 3^2'A2-,
3 I 4 H15 23 8M 9H 8^ SH 4^ 10^4 4 a: 3,% 254 13
1 8i2 3^4
17A10 HlO^ 11 9!| 9H HlSVi 3H 41|3H'3?4

O^-H 18?il3 H 24 .411 111% 12AlUS10i| VoUH 4J8, 43431.44
6fy21?^15 M,27 3-2l2 M13M 13l|l2?s:ili2 5^ 5H 6 334 4 H
7 >i25 16 34 15^14 13 5^ en 19 5H 7 ^4 5''8|5
29?il7 40 H\7% 18H,16^ 15 H 8H 217^ 7H

10 35^18 41Jîl7 "-^20 20?iil8K 17^ 9H 24 H 9 9 5.^'7 34
10 12H43H20 45 !o.J ''2\H\25H 26^251^ 22 >^ 10 ^ 130 A 11 I10A5348

11 |14^48H23 VJOO '23J^:30" 32 h;29h 27 H 11H;36 liH 11 eîSH
ii'îe 57î26}4\5SH28 32 H 12H\41H 13 3^ 12Jij7 95^
12 18 '57H2e}4 53H2SH3oTi\37y8 34 32 H 12hUiH 13^ 12>^9H10
SECTION V
HEATERS
The heaters used in connection with fan systems of heating
are usually some form of pipe coil heater like the Buffalo heater
or of cast iron like the Vento heater. Either style is made up in
sections or units in the direction of the air flow, which makes it
possible to assemble a heater of any desired depth. The general
arrangement would be the same for either kind of heater, each
being enclosed in a sheet-iron case or jacket.
Buffalo Heaters
The Buffalo Standard Pipe Coil Heater is usually one of two
styles, the regular open area pattern (usually written R. O. A.
pattern) or the return bend pattern (R. B. pattern). The cuts
on page 400 show clearly the difference. Both are made of one-
inch full weight steel pipe screwed into a cast-iron base, the pipes
being spaced on 2^-inch centers. These sections are ordinarily
made four rows deep, and are called four-row sections. Detailed
dimensions of these heaters will be found on page 451, of the
piping connections on pages 452 to 455, and of the sheet-iron
casing on page 456. Other special forms of pipe coil heater are
also made, such as the mitre coil shown on page 400, or the in-
direct heater shown on page 459.
These heaters are usually connected up to steam and drip

headers, with separate connections running to each section. In
case it may be required to shut off part of the sections during
mild weather, both the steam and drip connections to each sec-
tion are fitted with a valve. Each section should always be
fitted with an air vent which should always be thoroughly blown
out on turning steam into a section.
Buffalo Regular Open Buffalo Return Bend Heater

Area Pattern Heater
Buffalo Four-Row Open Buffalo Miter Coil Heater Without

Area Pattern Section Casing or Connections
400
HEATERS
o o oaoiPD o o o o o oê^T-
o o o o o o o o o#o oo o o @-eFTt:
oooooooo o o o o o o o o o-e^T
o o o o o o o o dio o o o o o o ^-o--^
Detail of Buffalo Heater Construction
Vento Heaters
The Vento cast iron heater, an illustration of which is shown
on page 402, is designed specially for use in fan and blower work.
The tables and data herein given concerning this heater are
taken from the catalog of the makers.
These heaters are made in two standard sizes, called the

Regular and the Narrow section, and ordinarily either 40, 50
or 60 inches high. The sections may be so mounted as to make
a heater of any desired size.
4Q1
- I'm m ]m m m 3m J ^ - i^fj- J* -
Vento Heaters
402
HEAT TRANSMISSION
Heat Transfer through Metal Surfaces with Forced Circulation
The transfer of heat through metallic tubes, such as a pipe
coil heater, from gases and liquids to gases and liquids may be
considered of the same nature as already explained for building
material in Part II, Section I. That is, there exist three separate

operations the transfer from the warmer fluid to the initial
surface of the tube; the heat passage through the tube wall, and
the transfer from the secondary tube wall to the cooler fluid.
The amount of heat transmitted will depend on the existing
conditions such as the nature of the gas or liquid, the arrange-
ment of the surface, the velocity over the heating surface, or to
some special conditions.
The total amount of heat transfer per square foot of surface

in agiven time will depend on the rate of transmission, upon the
temperature difference between the two sides of the surface, and
to a certain degree upon the absolute temperatures considered.
That is, the total heat transmitted per square foot of surface
per hour will be
H = K(ts-t)m (95)
where H = total heat transfer in B. t. u. per hour.
K = B. t. u. transmitted per sq. ft. per hour per deg.
temp. diff.
(ts t)m = mean temp. diff. between the two sides
of the surface.
As may be shown the rate of heat conduction between steam

and water is approximately, as indicated by the diagram on
page 404, based on data obtained from condenser tests. This
gives the rate of conductivity from steam to water as
v_ 1
0.000394+î^ m
where K = B. t. u. per hour per sq. ft. per deg. temp. diff.
V = velocity in ft. per second.

Condensing coils give a much more rapid rate of conductivity
per degree difference in temperature than steam coils, owing to
the additional effect of condensation.
The steam to air under conditions

rate of transmission from
based on a series of tests made by the engi-
of longitudinal flow,
neering department of the Buffalo Forge Company, is indicated
403
HEAT TRANSMISSION
-
HEAT TRANSMISSION
z:
t-7
^
%
g?
%
SOLID LINE = K FROM
TESTS AND FORMULA

I
K =
0.0447+ -^^^
DOTTED LINE =K FROM
I
APPROX. FORMULA
2
3
I K= K. [^^
AOO dOO 1100 \bOO 1000

Velocity Through Heater Feet per Minute
Rate of Heat Transmission for Indirect or Pipe Coil Heater
Transverse Flow of Air
407
.

The theory of heat transfer with forced circulation under con-
ditions of transverse flow has been fully discussed in the paper
'Air Conditioning Apparatus"* already referred to, in which the
coefficient of heat transmission from steam to air with indirect
or fan system pipe coil heaters is shown to be as indicated by the
solid line of the diagram on page 407. This curve is based on
tests made with a Buffalo Forge Company standard pipe coil
heater composed of one-inch pipes placed on 2^-inch centers.
The coefficient K for transverse flow as determined from the data
obtained may be expressed as
1
K=
0.0447 -h^^ (99)
where K = B. t. u. per hour per sq. ft. per deg. temp, diff
V = the velocity of the air through the clear area of
the heater in ft. per minute.
This formula was derived by plotting -^to^ in a similar

manner to that already explained in the case of longitudinal
flow.
As already shown, the total heat transmission is dependent

on the rate of transmission K, and upon the difference in temper-
ature between the two sides of the conducting wall or surface.
That is
H = K(t8-t)m (100)
It frequently stated that the rate of transmission under

is
the above conditions varies as the square root of the velocity,

but as a matter of fact, over the range of velocities ordinarily
used in fan system heaters, the transmission varies approxi-
mately as the two-thirds power of the velocity. This is shown
by the dotted curve in the diagram on page 407, which is plotted
from the formula as given. Thus we see that up to a velocity
of 1200 feet per minute there is but slight error in assuming the
above relations.
The mean temperature difference between the two sides of
the surface of a heater as expressed in this formula may be deter-
*"Air Conditioning Apparatus," by Willis H. Carrier and Frank L. Busey,
Am. Soc. Mech. Engrs., 1911.
408
HEAT TRANSMISSION
n n (ts-ti)-(t.-t2
^^'-^^-= /t.-ti\
)
(101)
where ts = temperature of the steam,

ti= temperature of the entering air.
t2 = temperature of the leaving air.
ts ti = entering temp. difT. between the steam and
the air.
t8 t2 = leaving temp. diff. between the steam and

the air.
Coefficient of transmission may be found approximately

from the formula
where K = B. t. u. per sq. ft. per hour per deg. temp, difif.
Cp = specifie heat of air.
G = weight of air in lbs. per hour passed through the

heater.
S = total sq. ft. of heating surface.
The amount of heating surface in square feet required may be

calculated approximately from the formula
S = (0.1119 Q + 127 A) logio(^^^) (103)
and the temperature rise from

/t,-ti\ f
^î*> (10^)
ilT^T^j =0.1119V + 127
where S=sq. ft. of heating surface.
A = clear area of heater.
f = ratio total surface to clear area.
Q = cu. ft. of air at 70 F.
V = velocity of the air through the clear area (at 70
F.) in ft. per min.
ta = steam temperature.
ti= entering air temperature.
t2 = final air temperature.
The derivation of the above formulae may be found in the

papers on "Air Conditioning Apparatus" referred to on page 408,
and the results obtained, while only approximate, will be found
sufficiently accurate for calculations based on the temperatures
409
obtained in heating work with exhaust or low pressure steam.
Since the transmission varies slightly with different steam
temperatures, for accurate work or for higher steam tempera-
tures will be found necessary to use the following formula.
it
This the formula used in the calculation of the heater tables

is
and curves included on pages 418 to 438.
f [(0.0001791
, VTs + 126.8)logio(^^3^)
t2
0.000003474 V^ (t- ti)"| (105)
0.0447 V + 50.66 i"
where T8 = the absolute temperature of the steam.
This same theory of heat transmission has been applied to the

Vento Cast Iron Heaters * and the following formula derived as
an expression of the coefficient of transmission.
1
K i-QQ
0.47 + (106)
While this investigation shows that at the same velocity

the heat transmission from pipe coil heaters is greater than from
Vento heaters, the frictional resistance is correspondingly greater.
But it was also shown that with the same effective velocity, or
with the same frictional the rate of transmission was prac-
loss,
tically the same for the two types of heater.
A further study has been made between pipe coil heaters

having one inch pipes on 2%-inch centers and the Vento Cast
Iron Heaters by L. C. Soule.f In discussing the results of
his tests the author states: "These results show that former
temperature charts published for pipe coils having 2^ -inch
centers of pipes read much too high and are, therefore, unsafe
to use. These results agree with both the Vento tests and the
Buffalo Forge Company tests and by their consistency show
their entire reliability." These tests further show that for the
same friction the Vento requires 35 per cent, greater velocity
than the pipe coil heater on 2^-inch center, but with the same
friction loss the heat transmission was practically the same.
* "Heat Transmission with Indirect Radiation," by Frank L, Busey, Am.

Soc. H. and V. Engrs., 1912.
t "Heat Transmission with Pipe Coils and Cast Iron Heaters under Fan
Blast Conditions," Am. Soc. H. and V. Engrs., July, 1913.
410
PERFORMANCE OF HEATERS
Temperatures Attained with Indirect Heaters
While it is true that the total rise in temperature will be
greater with a greater depth of heater it is also evident that
after air has passed over the firstfew rows of coils it approaches
more nearly the temperature of the steam in the coils, hence the
rate of transmission is very much less, and added surface is not
of proportionate value. For this reason it is seldom advisable,
in heating work, to attempt to raise the temperature of the air
above 135 or 140. For special work such as drying, where
higher temperatui-es are required, it is customary to use high
pressure steam in the coils.
In case the system is used for ventilation only, and the heat
loss iscared for by direct radiation, the temperature of the air

leaving the heater should be from 10 to 15 above that of the
room, depending on the drop of temperature in the ducts between
the heater and outlets. The temperature of the air leaving the
outlet should be within a few degrees of that of the room.
The ratio of the temperature difference between the steam

and leaving air to the temperature difference between the steam
and entering air is approximately constant for a given depth of
heater and a given air velocity through the clear area. That is
= approximately a constant.
ts-ti
Condensation in Coils
The weight of steam condensed in the heating coils may be
determined either from the B. t. u. as given in the heater tables,
or from the cubic feet of air handled and the temperature rise.
The heater tables give the B. t.û. per hour per lineal foot of pipe
for any given conditions. This, multiplied by the total number
of lineal feet in the heater and divided by the latent heat of the
steam at the pressure used will give the condensation in pounds
per hour.
The weight of steam condensed per hour may also be found

by means of the formula
Cu. ft. air per min. Xtemp. rise X 60
C=
Cu. ft. air per deg. per B. t. u. X latent heat of steam
411
For ordinary conditions with dry air at 70 F., it has been
shown on page 56 that one B. t. u. will raise the temperature of
55.2 cu. ft. of air one degree, hence we have
r. Q X (t2 - ti) X 60 , ^^Q (t2 - tl)
(107)
55.2 L
where C=lbs. of steam per hour.
Q = cu. ft. air per min.
tl = temperature air entering heater.
t2= temperature air leaving heater.
L = latent heat of steam (=960.6 at 5 lbs. press.)
Velocity of Air Through Heaters

The proper velocity for the air through the clear area of the
heater will vary with the different conditions such as pressure
carried and character of the installation. The following table
of velocities is based on the assumption that the pressure loss
through the heater should not exceed 50 per cent, of the total
pressure on the fan.
MAXIMUM ALLOWABLE VELOCITIES OF AIR THROUGH CLEAR AREA

OF HEATER FOR VARIOUS FAN PRESSURES AND FOR VARIOUS
DEPTHS OF HEATER
Total Fan Pressure in Inches
VELOCITY THROUGH HEATERS
MAXIMUM VELOCITY ADVISABLE THROUGH HEATER FOR DIFFERENT
INSTALLATIONS
Depth of Heater in Sections

some of the more important deductions being here reproduced.
These diagrams are drawn on the assumption that an allowance
of 25 per cent, on the cost of the heater be made to cover interest,
depreciation, space, rental, etc. The curve shown on page 413
indicates the maximum velocity allowable under the above
assumption for various costs of power.
Q 15
APPLICATION OP HEATER TABLES AND CURVES
yearly cost allowed for interest and depreciation to the yearly
cost of power, to the one-third power. The following deduction
was then made regarding the velocity through the heater.
"As regards the heater, the most economical point will be
reached when the installation is so proportioned that the yearly
cost of power due to the frictional resistance of the heater amounts
to 28.6 per cent, of the annual interest and depreciation allow-
ance on the first cost of the heater. This is true regardless of
variations in the depth of heater, temperature rise or steam
pressure,"
Application of the Heater Tables and Curves
Two sets of heater tables will be found on the following
pages, one to be used in connection with Buffalo Standard Heat-
ers and the other with Vento Cast Iron Heaters. The methods
here described for the use of the Buffalo heater tables are equally
applicable to the Vento tables. The values given for the Buffalo
heaters are based on data obtained by W. H. Carrier from
an extensive series of tests made for the Buffalo Forge Company.
The method of making these tests and of working up the data
therefrom has been fully described in the paper* "Air Condition-
ing Apparatus" already referred to. Large diagrams similar
to the charts on pages 432 to 438 were drawn and the values for
the heater tables determined.
The heater tables on pages 418 to 431 are computed for
various steam pressures and give the final temperature of the
air and the B. t. u. transmitted per lineal foot of pipe per hour
for different entering temperatures and velocities of the entering
air. These results are given for different depths of heater,
varying from one to eight four-row sections.
The curves showing the relation between the heater surface
and air temperature are useful for obtaining the final tempera-
tures when the entering temperature or velocity is different
from that giverf in the tables. As an example, we will assume
a steam pressure of five pounds, an entering temperature of plus
20, a velocity of 1000 feet per minute through the clear area
of the heater which is five sections deep. Starting from the left
side of the diagram at 20, follow to the right to the intersection
of the 1000 velocity curve, and then downward to the base line
at 2.05 sections: adding to this the five sections which we have
* "Air Conditioning Apparatus," by W. H. Carrier and F. L. Busey,
415
assumed the depth of the heater gives 2.05+5.00 = 7.05
for
sections. Passing upward from the point 7,05 to the 1000
velocity curve and then to the left side of the chart again gives a
final or leaving temperature of 113.
In case the entering and leaving temperatures are assumed

to be 0 and 140 respectively, the steam pressure five pounds,
and the velocity 800 minute determine how many sec-
feet per
tions of heater will be required. Passing from the left side of
the diagram at 0 to the intersection of the 800 velocity curve
and then downward to the base line we find a point of 1.2 sec-
tions. In the same way from the 140 point we intersect the
base line at 8.6 sections. Then the difference, or 8.6 1.2 = 7.4
sections, will be the number of 4 row sections or seven 4 row and
one 2 row sections, making 30 rows of pipe deep.
The lower graduations on the base line are for use with other
than the Buffalo Standard Heaters, w^here the value of the ratio
f is known. In the case of the Buffalo Standard Heater, the
value of f for a single four-row section is 12.335. That is, there
are 12.335 sq. ft. each section to each square
of heating surface in
foot of clear area. But in the case of other than the standard
heater where the pipes are on different centers the value of f will
be different. As an example we will assume that on measuring
up the surface in a pipe coil heater it is found that the total
square feet of surface is 1000 and the clear area for the passage
of air is 25 sq. ft. This gives a ratio of surface to clear area of
f = 40. Assuming the air enters the heater at 15 above zero
with a velocity of 900 ft. per minute find from the diagram on
page 433 for five pound pressure what will be the final tempera-
ture of this air. Passing from 15 on the left edge of the chart
to half way between the 800 and 1000 velocity curves and then
to the bottom line, we find a value of f = 20. Adding to this the
value of f as found from the heater measurements, gives a total
of f = 20 + 40 = 60. Passing vertically from f = 60 on the bottom
scale to a point corresponding to 900 velocity and thence to the
left edge of the diagram gives a final air temperature of 83.
These curves may

be used in connection with any heater
when the is known.
ratio of surface to clear area In the case of
pipe coil heaters composed of one inch pipes the value of f per
row of pipes deep is a fixed quantity for each distance between
centers of the pipes in the heater. The values of f for different
416
APPLICATION OF HEATER TABLES AND CURVES
spacings are given in the following table for heaters one row deep.
From this we may see that with the standard Buffalo heater
having centers of 2^
inches the value of f per single row of pipes
is 3.084 and for a four-row section f is 12,335.
VALUES OF f PER SINGLE ROW OF PIPE

ON DIFFERENT CENTERS
C=Pipe Centers in Inches

Spacing of 1
Steel Pipe
"
GO OS IM <r> 1-H (N t^ lO * CO
lO OX
0 -H t^ CO Cs XI
O O 00 001^
'-' WO5 1OC0
:M^ O 05 00 CO 00 t> t^ o
1
05 05 00 t>- 1^ CC I
05p0C0O'*_i0M ! so 0 lO *_ 05 Xo* lOiocoostôcoqcq

II CO o CO o kd 00 o cm" IN CO oco' iC o o
O^
, 01^ O^ CO (X ci 05
(N << <D 00 O 1
t>-
'
I
I
<-i CO O
CO 00 05 cMTfict^xciO'-i
O O C5 ^ ' t^ LO T^CCCOCOXr-ct^cO ic X CM o r^ t^ X --I
(^^t^^t^co05lC^^
C^l
t^CMl^COCSCOiMO CO X * O CO CO C: CO
Xt-t^t^cCCOiCO
O X t^ t> CO CO CO XXl>t^cOCCcCiO
lO 00 o X * -: * * 1
^p CO Oi p o CC_ p. CO CM ppppp
Qi cE lO t^ t>; Tt O CO '*' t>^ 1 ic Tf lo ci d (M
dX -^ CO CO -- 1~^ ^' ^' *' :*'
C r-" CM
CM-*COXO>0^ CO lO t^ OS CM -^ CO t^ C5
U rT
I
H
<
X* M O *X
Xc
^Xt^COiOOt^cO lO CM CO 1"
r .
CO t^ CM X * O t- -<}<
r^ CI i> 01
Ci CO 05 lO 1-1 lO CM
--I
iCOCOCMX>OCMC>
*
Cue xr^t^cocotoioio t^t^COCOCOOiO"5 t^ t> CO CCiO o iC
J =
. o-T-
ca
lO CO 05 ^ CO iq -H ^_ : CO l> t>; p 1^ P 00 p CI '^ p

.
t>: t^ p iC
t>;,-H^'dcddco-*' '
do6o6co'^'ocot-^ ^'ccocMcco^f-
COiOt^XO^CM I
rHCOiOt>00' iCM
!
I
CM-*COX050CMCO
ClcO-^t-'^'^cOO
CM t^ CM X * X CO '
XCOOi'OCMClCC'^ lO O CO CO C t^ * CM
COCO'CiO'O'*'*'*
t^COCOOiCiO-<f* COCOOiOiOTfTt<-1<
11 CO J
t^CMCOCOCÔcOX I
-i^ C 05 O r^ * O O O --I CO CM kC CM
eE
%i Q^^^^COXiO OSCDCMO'-iiO'-iO lO .-I O

05 CO t^ CO *

QU CiÔt^->*<-^XCO lO I
mio-^-^-^cococo
X lO CM 05 t^ * CO 05 CO CM
lO <* ^
"^
r^ O
CO
*
CO CO CO
iOOiO'<*'1"*COCO
Z3
X p CO CO r-H Tj< t^ I> lO OS t^ r-l X -^ * p 00 CO p CO CO CM
1^ SE
t
cm'
.-H
-^_
03 CO CO
CO CO X O
OS d
CM
< CO'
r-l Tt<
d *' lo -^ d
CJ iC b- 05
CO d
CI CO ^ --^
*"
o a
CO OS X CM X 0 0 X X CM CM CO CO CM CO t^ O * CO CM O 05 CM CO
X 0 CM O X CO -^ OCOCOÔSCOiOCO
P
u
"^ Cue einS
C0C5COTt<Ô>tîO
rf CO CO CO CO CM CM CM <1*C0C0C0C0CMCMCM
T-H
* CO CO CO CM CM CM CM
icppppppoq
p 00 rH CO p p p CM >* pppoq-^cMCMic
a.
II d lO 'H CM d d 00 d CM CM t^ t^ lO d ci ci CO d CO d CO ic id
cm"
i-ctitÔS'-<CMC0O O OS CM * lO t>- --I COCOXO^CO'^'O
CO CM t^ CM OS X b-
T(< CO CM OS CJ* O * CO r-i i-H -^CCcDi-HXCOiOcO
O X t^ in "^ CO CM
i S J s
Tj< CM O X ^- 0 * CO
ClCMCMrH.-^rHt-(^
CO r-i OS t* CO lO
ClCM^^.-HH,-i^
CM
CO CO OS ^ O t^ t^ CO ôO'-^ppcocMco I
poqoqoqpppin
SE odddodddcdco
CMiOXOCM-îOCO
I
2 g :;j gj S! J2 J5
<^'
COCOOS^CM-*incO
suoipss .-I CM CO * lO CO t^ X --( CM CO >* O CO t^ X -< CM CO <** iC CO t^ X

J3JB3H JO -ON
jjV 8uuau3
*jo 'diuax
418
BUFFALO HEATER TABLES
CO
0<
t/5
u
<:
= u:
09 .
Ess
O a-
= 1
a
^

CO CO 00 I C^ lO 00 ^ CO CO CO CO COW'* (NW w X lo Tf ^ r^ CO IN
1-H <0 M
00 Tf
WOOOOt^t^t^lOCO
t>- O CO IN 00 * O t^ << -t OJiocô>coeo>-io
oooot^t^t^cococo
ca j^= CO CO CO 0 0 0 lO <*<
0 OS O
l^ l^ (N -I ^ TjjoqccoTjjMos-*. coqiNp-ôscot^
II -5 d
00 co' t^' ^' -I
CO 0 CD 00 Oi --I IN CO
d d QO c d CO lo c lo
IC t^ 05 O (N
'4
.-H
OS
^'
M CO M
O rH CO
*
co'
0
tri
*
00 -^ ^ O O 00 00
CO OS lo 1^
Tj< 0 W5 CO Tji ,-( O OS O
OOt^t^t^COCOCOiO
o i^ 1 Tt< OSOi-it^'Ji^tîO
t^t>-t^cococooo C0C0O00'
CO u
'-^'oosocqcqoo OS lO O rH ,-H lO -^Jt O CO O O >* -(J* r^ O IN
co'CO ^' 00 in" ui Ni ^-ôi'tood^'^'
u 2 CO lO t^ 00 O --I <N CO
t>.*
^ CO t^ OS O (N CO Tf<
<
O O
O* OO
00 OS CO Tt< T^ t^ COCOOOCOCOIN CO 00 OS -H -^ 00 t^
lO rt CO CO OS CO CO
t^t^C0C0U3000
o :-<t^ CO o
CO CO 00
t^ CO CO CO lO lO 0 Tt<
o t^ CO 00 lO CO 00
m "^ dm lO 0 lO * Tl ijt CO
Q <^ O 00 (N ^ CO CO IN 00 CO >* 00 ^ Tt< O CO <

O iC <*< <t X IN (N <N
oO O ^
In. CO CO --< OS <N 00 CO CO i lO t^
^ 00 tosOTj*o5cot*r*
O CO * N
y 0 t^ Tjt 00 0 IN
CO CO lO lO 0 Tt< ^ rj(
IN 00 0 I 00 C CO
CO0OiO'^'*Tj<Tj<
OS CO <*
^rJ^^ÔOCOCOCO
I
ii q 00 CO q q q q >* lo
dq *
--^ 00 <N (N 00 c ot>;Ti<iN,-Hqqq
09 ^ cE t^ d
<-< Os' lO 00 rH lo 00 00 d CO 0 d 00 00 d d d
Tjl Tji Tji <j
CO 00 O IN CO
oo'
CO CO 00 OS -i(N Tf o
E;2 00 OS (N lO CD CO 1* .-I *(
t^ CO OS 00
O
* o ô >o OS OS r^ .-H CD Tj< Tj( tîooooeo'*t^'-i
CO CO CO r^ 00 CO T)<
lO lO Tl* Tjt Tf CO CO CO
^ t>. Tji rH OS CO rtt M r-i 00 0 CO ^
OS t^ CD
"^ "C
lOT^^Tjteocococo Tft CO CO CO CO CM IN <N
^
Z 3
<^ 0 CO *
CO t>. t>; CO q cot>.qqTj<oqTtec qcoqtN.T-iqot--
II Os' iC l>^ co' ci Co' OO' Os'
CO CO 00 O
(N CO 1< 0
00 M *' ^' d
Tt<N.OS^(MTftlOCO
r>; in' in W (N OS
00 o CO
co' co' t>^ co' -^
*< >o CO r^
oa I
^ <N ?< CO (N O <N lO OS 00* IC

o 00 CC CD 1-1
4^ W V A l^ >* OS r^ lO rt< OOICINOOOCD'ÎN
-"t
OS
11
<o
)
CO CO CO IN IN (N (N CO CO CO CO CM (N IN (N
CO
CO CM CM CM CM CM
*< CM
^ 00 .-.--CO'J<CMCMC>)
wor|-coTf<cocM ^ OS 00 CO W3 CO CM *
CO J^=
St
CO CO t^ 00 1^ CO CM .
q h- CO 00
q ^ OS T*< cî CO CM CM coo.
d CM CO
lo o
O
lO 00
d t^
-H -H co' o6( '
"
'
Ec2 00 <*< 00.
suoipas
J3IB3H JO 'ON ieMC0T<OCOt*M I CM CO ^ 0 CD t 00 I CM CO Tj< lO CD t* 00
jfV 8u|jaju3
JO 'duiax
421
FAN ENGINEERING-BUFFALO FORGE COMPANY
t/}
4>
U
% .
=
06 IC
1
< N
o .
z 5:
< ^
3?
aa r
t/3
EJCO 1
H cS
?^
Q UJ
UJ 0^ d
Z 3
<;
CO Oi
o a.
uu
oô
H <
S
UJ
&
u
CO
O
u
1
1
U. u
CO . j
r
c
^ 2 i
u u
2 3 if
<
si
u 1
a
r>cooo<Nosoc-H < * IN ^ b- lO CD 05 *^
T}< t^ M t>- c^ t^ ro o >eOMOO-*00 .-1 1^ (N 00 t^ -"l"
i^dS^ ôoosooooooo ) O OS O) 00 00 00 1^ OJ 00 00 t^ t^ t^ to !0
O O t^ _ cc M <N OJ
d 00 o ^
t>I
M CO
00 00 d
ro *
c>j
t>.
*
t- t^ Tt< t^ Ifl Tf OS O *o
q p c
00 ec t^ d c^
.-I *
OOO'-^CJ^COtÔO
t-; t>: ^^
--H
= uu5 -H X 05 <M t* 0 CO O
IN 00 M 0> iC !M
C O CO "5 X CO (Nt^<NC000iO>OtO
OS lO I t- Tt ^ X
rji <}(
OW
Tf "tOSTfOtOClOS t
OS 05 00 00 t^ t* OS OS X 00 00 1^ t^ to X 1^ to to to lO
t>. t>- t>-
iOMtCOOOiO-* M q q q >o q q O 00 O * (N CO C^ OS *_
SE IN CO CO d d (N
eo * to t*
t>^ to'
t^ OS
Tj* t-i
UJ
H
< N
UJ 0-*-4tOO<Mt^CO
"t 00 CO 00 1* to CO o
lO 05 OS l> CO o
OS *
OS 0 t^ rH ^ *o t^<N'<tt^C0(N'*tO
to <N t^ CO r>- ^ "-I o
OS OO 00 t^ t^ t^ to to X 00 t- 1^ t^ to to to t^ t^ to to to lO >o o
cfl 2
qM c^_ t-; q -j
q ^' iq q 00 o to
_ ,_ CO t>; q -^ (N q <N
II -< 00
<j<;0OS'-<C0'*tOt^
w cc M oo' r- OS OS t-
OS* to'
"* r^ 05 ri CO
C
1
d * OS CO o
lO t>;
XOM-Ôt-XOS
to'
u"" u 3 CCOS0C'CCTj00iOO
^ lO -H t^ M OS to CO X CO OS Tf t^ ,H
O W CO 00 t* to
i^c-ix-*'--ootoeo
,-. rji
1>. .-I
00tt-tO'.OOOO t^ r^ to to to o lO o CO to o m lo ^ ^ *
u. u
U. H OtOOtOOSOSCO CO 00 to OS OS lO OS X
OQ . cE
^-OSiOt^-ÔC^JiO Tf lO CO to (N IN CO O X OS CO to N OS 00Tjt
t^ c^
;OtOiO>OiO'1<T}<1i
X * 00 in <N <l<0--0(NOStOCO^ O t^ -f I OS to *
Tjt
lO 0 }< Tf Tt CO CO CO
I Tj C>1 Tt Tj< OO rttClCCt^'-nn^-rf q (N CO in

1-1 CO q 00
c E dd 00 r^ ^' d c^ <'
^ CO in b- 00 o ^
S E;2
(COOtOXOS OS
^^^^^(N(N
- a
OSMOSWOtOCON
^ OS lO CO CO t* CO ^
.-I
^ OS
OS o o
in CO '^ Tf X
OS to M OS t^ (N O t- *X 0 CO < t^ in (N 00 to *
M
^^^cococococo ^<l"*COCOCOCOPI ' CO CO CO CO (N <N
il Cfl _3
I
W O i-nO OS M Tf< ic ^qq
<*; "-^ q in q in q q co ^-^^ c^_
UJ
II d CO IN d t^ d ci CO ci d t^ in d ci co
<*<
inosNTfcoaos-- OS M to X OS I w
^^^^^ ^ ^ ^ ^ ^ IN cî
-"t
C>>
UJ
H C^ O OS O CO 00 -^ N N OMtO-HXtO
XtOCOCÔOOt^tO r^ in CO -^ OS 00 to lO
î^ ?1 N N M C^ ^ -H M C^ IN N -H ^
>- .-I .-H g5?325
OQ
ooq(NOJiqqqo5 X q q in CO q 00 ci CO ',0
1^ o c^ q CO
_
se to d 00 *< r>^ d ^' CO ^'N CO d d CO d --'

r; ddd
cscomrôs c^co
d * in
h-i
iOOSWiCtÔS'-C>l toc ^m
^ CO ^X ^
^ OS c^j
C^j cî
suofpas 1 N CO I* 0 CO t^ 00 -< M CO * lO to t* 00 I M CO * in to 1^ 00
J3)e3|1 JO -ON
"
|o 'd'uiax
42.5
>
CO 3 u"" k 3U. *-
lO OS CO CCOSCOb-MlÎNX
lO t^ O CO 05 C
<*' Tj< t>. <1<
.
J be
u u o . t>. t^ lO OS ^ 05 0
>-<
M !M c O O Oi OS COIN^ÔOCSCS C^l^ÔOCSOSX
aa
. o-r
2^ CO
O CC <N
M O lO 00 OS
t>; 00 ^_ -"t ^ O OOOOt^COXO X * OS <t U5
r>^ rjH OS
>(Neoio
1>-
lO * -O O -H CO -^ !N t^ Tt r>. c OS cî cc o I ^ IQ CO OS Ni t^
Tj<
TjitÔ'^CS'ÔStO >o^:cc><
^ C^j ^ ^ O OS OS X 00 osco^c
^'OOC5 0SXX
tooscot^cit^coos
I O O OS OS X X t^
CO lO (N O ^ >0 <*' CO ccocoxco-^co I

O C^J C^J O O (N
c^xtôx-<xx OS CO -H ^ OS ^ CO * co>OMOs-xrô
ai c-^^xcot^cox-* t>-
>0 C C (N<*< N. M
-.C(N r^ CO o
r}< -x;
UJ
Ô0S0SXXtl> OOOSOSXXt^t- COSCsXXt^t^t^
H
.xoO'j' CCX'-iXO-H, lOcoc^^cCTfoiNC;
E It
< lO OS t^ O X OS (N TjfcOOOiOiOXC^ U3 1^ CO * ^ CO 0 (N
X *
> . u o t^ O OS 0 O CC CO
<?'
ososxt^tî>cc-o
C^ l^ C^ X ^ O * OS C CO N OS
c:xxtî><c;c;o OSXt^t^t^tCCCW
t>- >-< <*<
o '- CO o * o q q *
'X) OS C^ 'I* O lO * OS X CO 00 q N; lo q (N
^' o
t>.
c E o o X
(N lo X ^ CO c r^ X
'
^'
co' N^ OS X ci O lO
*' TjJ
<l COCCOSC-lTfcCXOS
'^ M Ol CO lO OS CO lO OS OS lO CO O t^ X
Tj( CO 0 1^ LO X W < C^l
X t^ CD CO CO lO lO lO t^^CCC^XO^X
CQ o-*os-*ococoo <*<os'*occcoor~-
CO CO c o o ^
a-
t s
tfl
CQ 2^ t> t>. tCCCOCO0'CC'^
(N q q lo ^'
-H c>) ifl cî qqcoioqcoq'j; q^_t>.qqicoqo
^ d ^
^d d CO d i6 <N
*' Tf CO i^ ic c^' <N OS r^ -^ ci '-i *"
iv^
* CO X o O (N lO t^ X O t^ o CO ic t^ OS o
E^ C^ CO OS C^l CO CO
<*< CO "5
OS Tt*
W O X X t^coosioMosr>.'*
OO X
t^ CO
C^X O I CO o o o t^ O^C0r-OC0X0
>o lO ^ t^ Tf ^ X 0 CO
0 0 -^ Tt
0 lO 0 ^ <'<* CO CO CO CO CO lOlO-^-^-'tCOCOCO Tt*
f QQ 2
qqqq * q M q i^'^-^c^qqicq 't CO q CO q c^_
q
io
c E d lo *' c" c^ d o COt^'i-^COXC^
OS
C^tÔCOCCXO^
I^ N^ -I d lO iC CO 00 id <}< t^ lo OS os' d^
r- ^ ^ ,- O) C^
O OS 'H O <N CO CO OS O * C^ (N 0 Q t^ ^-0>0*ÔXTJ^^

a. O CO
<J< X CO (N O OS
N <N N (N ^
(MOtîOCOÔX
<** T-( OS CO T}< C^ OS O X
~? S"
CO CO (N COCO<NN(NM(M^ CO C^ M
(N (N <N t-H F-(
u (^^
H
qqqoq 00 00 oiq c^qio-^qqqcq coocjo-^c^co
d ci
cE Xd <t x' lo d -H CO id d
^' OS 00 id ^'
^' --< *'
co' c^' t>^ OS
C^ lO t^ OS CO * X lO X C CO lO 05 CO CO X O C^ CO
'
CO C^l C-)
^ ^ ^ Cl C^l C^l
suoipss
jajB3H JO -ON
(NCCîOCOt^X "(NCO'îOCOt^X < (N CO <*< lO CO t X
jiV Suuaûg
426
CO o o o o t^ t^^ oi ^ o "S S
Ti
5
i Tf C3 CO 00 9 S;
'J'lO
^ O C5 OJ Oi X X CiCSXXXOt^*^
rî
q ^ q q t^ qo* -; ^ q
q q o q Oi Ci w
,-H 1-H
O lo
0rTl<t>^t^5OC^t^r'
ic
t>:
CO
d X cc lO
Tji Tf cj
* t- C5 >- CO r^ -"t
t^.
c>i CO rr CO
00 O
c
ic ci ;- CO
CJ TT u-; -w X2
ôc^cîooxp X CO X gj IN Oi o CO t^CJiOC02O5OO
O'O'-'têoocoeo
^ a . o iH00 05XXXl> 0003 0>XXt-t*
1' CQ J "^
2 o <N M q q <N. c^. o^
(/] cE
c^
U4
S
H oot-0'*>'5xeo OSCOXOiCCOOJ
J.U. SSSi::;coMWo
-^ S^
o
fs OOXXC'^t^<>' o o q q q ^^ "? <3?
--<*<oqx)0''5 ^ <N X O t; .-' Cî
a cid^'--i-^'coccc^
X CO m <C X OS O
t>;
c 11 CCDJX'OOX'^X lOXOCI-ÔtÔ
es
ul
o OSXl~'*'OXCOC"l Mt-eoc;5C(NOtr
t> t>: i-H ( (N t^
* o
t;- WXWOJO^XJ-0
) 00 r^ t> o
o "3 00t>.t^;OODOO
< <
"S
9 en j^
u.u u
o ^ X t^ X) ^^ q o * q C5 CO c^
CO --' t^coqqoqoqiO'*
U.H <
lo
TjJ IN X 2i
r--' 'S'
'*' Ci iC 00 X lo c ^ o
i lOX^COOt^XO
cocceoiooit-xo ^ ^ t^ ^ Tt CO 00 -!
^ o c c^ OJ o ^ OiiC XCNOt^
lO iC >0 -* t * * CO
OS
O m
U ^ ic X <N -* X loeo *
lO i q
o o ''s fî '^
d ko CO ;* o * CO
CJ CO m' x'

O
C5
^ 5o c-i
Tt 00 CO t- oi rt ^ ^ ^ Cl IN
CO C: CO IN
.-it'^^0CCO--<
X X * * iM c 1* c- X o; -
W
lO -^ ^ rf <t CO CO CO rr Tf CO CO CO CO IN
is CO J
iotôt^qpt>;>o
O o
CO 6 CO q
-I CO q q ccont^qqqQq
d CJ o x' cdt-^-^diftC-ijOQCi
CO d ;*
CO c^' t~^ eo'
rp
u e E
a. E(2i
u ^ X eg
* O o
oX Ci -; CO c>> CO CO c
ojor^5oxe;jxio lO r>.
05 X CO '-' iCCO'-C-.XCO'O'
^ -"-"-*
oxoco--oxt^
- -*
a. t^ 1^'
C^ (N <N (N W --"^ -i (N C^ (N '
CO <N IN !N IN <N
OQ J
oeoqq"*.f-:'9 t q 00 q X CO t^ rîqeoq-^co0'-<
-<.
II M t^ ic d d d o o6 d ci d CO d 21 ^doddcqc3c-j
r-'
2222g?l?3 2S22g?^SS
E|2
SUO{P3S ^Mco'.roict-x ^Nco'f fl5r-c

-iNco'*o<or*c
j3)e3H JO "ON
jiV 8uu3u3
jo 'diuai
427
o
00
u.
u. u
U. H
09 .
IS
03
O
u
Jo
Q.
UJ
^
i^ ^CCeC(N^ÔO -VCCCÎO-hOOO COIMCI-hÔOOS
!
lO-^XOiCiOiMCO CCiMXiCMtX'i ,
OOt^CCXi-OX
CC C^ rH ^ O C 05 O ; <N lO 05
;iMr-OOC505X
-"t X * 05 (N
iOX(NCCC0^l-
iÔOOSOX
I
CO "j^=
0-.C'*'*'<totêc O rf C-J CC OXO -"t 0 M iC X n X CO r}i
si <T c :c X r^ ic ^
cî
CS t- 0 t^ <N O* '-I (N
X X lO <N t^ CO t^ ^ lO M CO * CC
o eci^ CI CC
t^ (M X
1
ecccc-<"o;i<occ
Tj*
C^l O O 05 OS X X ^COC5C5XXt^ ÔOOSXXXt^
(NOrf^XOOtNO q q cc c X q lo o w q q (n lo m
>*;
^'
<*;
05 X m' ^'
C^j C-i t^ C ci Tf t^ t^ IC
im" t>^ I
(Ncco;^-* '-CX05 cotôci'^ccxo
O O * W CO o c 05
C^l
lO X C^ CD (O (N X ^ <N O 1^ X * CO lO OC
kO 05 CC X -^ C- CO
O OJ
C^ ^ CO (N W lO
CO^CCfNX'H
OO 03 X X t^ t^ CO oc5xxt^t>.r^co O O X X l> t^ CO CO
I
CO j^
MOXiO'4<iOMO OiO OiOOCiO TfX;
cE CO rooted 05 CO (NXCXiM-*CCO: CiCCOCCt^Xt^<N
C-J :C 05 <M ^ CO X O CO CO O C^4 lO t^ Ci C , T t^ O CCiC t^ 05 I
M-ÔSOCOrÔiO OSOiXOlI^XW ^ to X X ^ CO to
'I*
tÔIÔiO^XTf CO t^ (N l^ CO OS CO CO ^ to o to ^ X * --I
XX t-- 1- CO CO 0 lO X t^ t^ CD CD 0 0 0 Xt-t-COCOtOlOtO
CO J
q-*toqcoTt<-^t>; ^CCCCCCCOt^CO COCOtÔltOC^-*
II 00 X CO
(NCOOCOCOXC^
'l* d CO CO 05 cotoocMor^cceo
CON. CCCOXOOl
t CO CO -t C3 C CD
*X'-"*'CCC. C^l
^ ^ ^ rt ^J (>J r- CIC^ ^ ^ ^ ^ (N ?!
tOCOC0-*M cox
oi t^ CO oi CO CO r^ o
'
CD XOXX(Nt^
ocD X-* osco
, an exj- ;
cDtoiotO'^'*Tjeo CD to lO * -"t -* * CO I
CO to to **-* CO 00
CO qqqq t>;io toqoqqcoqr>;M qtoqcoxqqq

OMXX ^'cdcd"m
j
cE CO oo cc' 00 cd' oi ci co' to CO ceo mo; '
wr^- ^1*1^05- CO rfXMtOXCMCO tOC5MtOXOM->H

MM MMC^J I
MMM
X r^ X CO M M <1' lOCOOStO-^tOXCO COI>Osr-CDXMX

*: s cocooxco'*Mi_ tOMOSfîOCO- O * XCOt1<M OS
. <a
COCOCOMMMMM COeOMMMMMM COCOMMMMM
I
&e
'5^
^
CUiS I
00
t>;q q -^ q q !>; tOtOXCOtOCOXM CD * OC to to t^

d m' to 05 cJ osm X
xt^r>. o
CD -^ CO CO CO to ^
MMM
co'
Tf 0> CO CO OS CO to
co'
M * to
MMM
Tf OS CO t^ OS
I
tOO-'ttÔM'^CD
MMMM
suoipss I M CO * to CO t^ X M CO -"t to CD t^ X M CO Tj< to CO r^ X
J9VB9\\ JO 'ON
jiy 8uiJ3)ug
JO 'diuax
430
i
1u
G
CO
IS
r
c
i
<
o
>
10 IP
-40 '"r '"r'"i""i'" i""i" "I"' 'I I""!' 'l""l

20 40 60 80 100 120 14-0 160
r _ SQ.FT. HEATING SURFACE
' " SQ. FT CLEAR AREA
Relation Between Heater Surface and Temperature of Air

at Various Velocities Measured at 70 F.
STEAM PRESSURE Lbs.

Temperature, 212 F.
Application of Heater Curves. For example, the air enters

the heating coils at 5 deg. above zero, with a velocity through
the clear area of 1000 ft. per min. What will be the final air
temperature with a Buffalo standard pipe coil heater, seven
sections deep? Follow dotted line from + 5 deg. to 1.6 sections
on bottom edge. Adding 7 sections gives 8.6 sections. Following
432
HEATER CURVES
240
200
l?0
r SQ FT HEATING 5URFACE
''
5Q FT CLEAR AREA

STEAM PRESSURE 5 LbS.

Temperature, 227 F.
dotted line upward to 1000 velocity curve and to left edge gives
final temperature of 124 deg. Reverse this process where the
depth of heater is required for a given temperature rise.
For more complete directions see "Application of Heater
Tables and Curves" on page 415.
433
FAN ENGINEER ING BUFFALO FORGE COMPANY
240
200
160
120
M '''''
' ' '
'm ''
' ^
m ''' 'M'
'
''
U ''M ''M '^'Vi^ '''|''
'
' ' ' ' * |'

i
F' ' '
M'
'''I
' ' ^
20 40 60 80 100 120 140 l&O

r _ SQ FT HEATING SURFACE
* 50 FT CLEAR AREA

STEAM PRESSURE 20 Lbs.

Temperature, 258.8 F.
For Explanation of Curves, see pages 415 and 432
434
HEATER CURVES
240
H i|rl>i |n lr 'iiir p n
i|-ii.i|
|
'
r| I'l
60 SO 100 120
5Q FT HEATER SURFACE
SQ FT CLEAR AREA


For Explanation of Curves, see pages 415 and 432
435
280
T
60 60 100 \20
SQ FT HEATING SURFACE
SQ FT. CLEAR AREA

STEAM PRESSURE 60 Lbs.

For Explanation of Curves, see pages 4 1 5 and 432
436
HEATER CURVES
260
240
80 100 120 140

r SQ FT HEATING SURFACE
' SQ FT. CLEAR AREA


For Explanation of Curves, see pages 4 1 5 and 432
437
riBni^KBmiKO BUFFALO tOUQM. COMFAJil
740
too
10
I?
AC \-
VENTO HEATER TABLES
FINAL TEMPERATURES WITH VENTO CAST IRON HEATERS
REGULAR SECTION 4 ;^^-INCH CENTERS OF LOOPS
STEAM 227, 5 LBS. QAUQE

REGULAR SECTION S-INCH CENTERS OF LOOPS
STEAM 227^ S LBS. GAUGE
1
VENTO HEATER TABLES
REGULAR SECTION S%-INCH CENTERS OF LOOPS
STEAM 227, 5 LBS. GAUGE
s

NARROW SECTION 4 ^=INCH CENTERS OF LOOPS
en
VENTO HEATER TABLES
FINAL TEMPERATURE WITH VENTO CAST IRON HEATERS
NARROW SECTION 5-INCH CENTERS OF LOOPS
STEAM 227*. 5 LBS. GAUGE
w
FAN ENGINEERING-BUFFALO FORGE COM
PANY
FINAL TEMPERATURES WITH VENTO CAST
IRON HEATERS
NARROW SECTI0N-5K-INCH CENTERS OF LOOPS
JS
Left-Hand Bottom Horizontal Discharge Fan
Drawing Through Heater
Full Housing Top Horizontal Discharge Fan Blowing Air

Through and Underneath Heater
445.
FAK ENGIXEERINO BUFFALO FORGE COMPANY
00
H
< JS
= i s
g ^
ffl
= -s
^ a^ c
o < 2
"J a
o ^
?i
< Z.
Jt
FRICTION THROUGH HEATERS
FRICTION OF AIR THROUGH VENTO HEATERS

Loss in Pressure in Inches of Water
>u
Friction Through Heaters
On pages 446 and 447 will be found tables giving the friction
loss or drop in pressure through both the Buffalo Standard
Heaters and the Vento Cast Iron Heaters, with different veloc-
itiesand depths of heaters. The values given for Buffalo heaters
are based on tests made by the Buffalo Forge Company and will
be found accurate for pipe coils. The table of loss through
Vento heaters is based on tests made by the American Radiator
Company.
Sizes and Dimensions of Buffalo Standard Heaters

The table on page 449 of the Sizes and Dimensions of Buffalo
Standard Heaters gives the information required for the selection
of a heater for any specific case. The third column gives the
length of the section (or of the cast-iron base) as also the number
of rows of pipe in the section. The fifth row gives the various
heights that are made on each base. Thus a heater 4'X6'10"
is 4 feet across the face by 6 feet 10 inches high.
Three columns are given showing measurements of the surface
in the section. One gives the actual measured lineal feet of one
inch pipe in each section. The next gives the actual effective
square feet of heating surface in the section, counting in the
exposed portions of the base as well as the surface of the pipe
fittings. The third column gives the equivalent of this surface
expressed in lineal feet of one inch pipe. Thus in the 4'X6'10''
section there are 428 feet of one inch pipe, but the total exposed
heating surface is equivalent to 455 lineal feet of one inch pipe.
The column of clear areas gives the actual clear area,
expressed in square feet, for the passage of air through the heater.
Having the quantity of air and the velocity through the heater
given, the values in this column decide the size of heater to be
used. The number of sections in depth of the heater will depend
on the desired temperature rise to be obtained.
448
SIZES AND DIMENSIONS OF BUFFALO HEATERS
SIZES AND DIMENSIONS OF BUFFALO STANDARD HEATERS
BUFFALO STANDARD PIPE COIL HEATERS
TOP or CASING
MAXIMUM DMP
PIPZ C
Regular O. A. P. Heater
rap OF Casino
H NO. or PIPE IN &ECTION.

J CJ.EAR AJiEA IN SQ. rr..
K LINEAL FEET OF I 'PtPE.
Return Bend Heater
450
DIMENSIONS OF BUFFALO HEATERS
i^
STEAM, DRIP AND AIR CONNECTIONS FOR

REGULAR 0. A. P. HEATERS
HECEIVSH
s u
LIST OF FITTINGS FOR ONE SECTION
Steam Connections 1 Elbow "G"

1 Globe Valve "A" 1 Check Valve "H"
1 Elbow "B" 1 Nipple "J"
2 Nipples "C" 2 Nipples "K"
1 Nipple "D" 2 Short Nipples "M"
1 Bushing"N"
1
Drip and Air Connections
Tee "E"
\W Pipe "P"
134" Pet Cock "Q"
long
12''
1 Box Union "F" Female Thread
Main Drip
1 Pipe Plug "R"
Tees "S," Same number as Number of Sections
Nipples "U." One less than Number of Sections
452
CONNECTIONS FOR REGULAR O. A. P. HEATERS
1
STEAM, DRIP AND AIR CONNECTIONS FOR

RETURN BEND HEATERS
STEAtt
RECEIVER
S U
LIST OF FITTINGS FOR ONE SECTION
Steam Connections
CONNECTIONS FOR RETURN BEND HEATERS
s
1
z
DIMENSIONS OF HEATER CASE FOR BUFFALO

STANDARD HEATERS
HBTUHM J3ENO
H'UiA. or HOLE ran steam pipe.
J -JJLA. OF HOLE FOP URIP PIPE.
Size of Section
HEATER CASE, CONNECTION AND BY-PASS
Heater Case for Buffalo Heaters
Detailed dimensions of the casing used for the Buffalo fan
system heaters will be found on page 456. Care should be
taken to have the connection between the fan and heater case of
such a character that it will not restrict the flow of air or offer
unnecessary resistance. This precaution is frequently over-
looked, either throwing excessive pressure on the fan, or curtail-
ing the quantity of air handled.
The following table gives the approximate lengths of con-
nection advised for a draw through installation.
LENGTH OF HEATER CONNECTION

FOR DRAW THROUGH EQUIPMENT
Size of Fan
Indirect Heaters
This a special form of pipe coil heater, details and dimen-
is
sions of whichwill be found on page 459.

As the table shows, a variety of sizes are built, the smallest
being six pipes wide and eight pipes long. Under the heading
of "Size," the first row of figures gives the number of pipes
across the steam supply and drip ends, and the second column
the number of pipes in the length of the coil. Cast iron
manifolds are used for the bases into which the pipes are screwed,
as in the regular fan system heaters. The indirect heaters may
be used in an upright or horizontal position, according to the
requirements. These heaters are shown as the solid base type
and a diaphragm in same compels the steam to flow evenly
through all pipes. The steam supply enters the heater base at
one end and the water of condensation is removed directly
opposite. These coils are designed for the use of either live or
exhaust steam.
Blast Area of Buffalo Heaters
The blast area of any heater may be determined by the
formula
A. P. M.
Ab = Ypress, (109)
4005y drop in in.
Thus, if we assume a case where 10000 cu. ft. of air per minute
is to be passed through five sections of Buffalo heater at a veloc-
ity of 1000 feet per minute, we find from the table on page 446
that the pressure loss will be 0.479 inch. Then from the
above formula
Blast area = , =3.61 sq. ft.

40051/ 0.479
A general discussion on the subject of blast area, together
with an example illustrating the application of the general form-
ula to an entire heating system, will be found on page 126.
458
INDIRECT HEATERS
INDIRECT HEATERS
'B^ STEAM STEAM 'B

SIDE ELEVATION END ELEVATION
ooooo'ooooo
ooooo ooooo

o o o o q";o o o o
o
o o o o o 'o o o o o
8TiAit[ ooo
o o ooo-'oo
ooooo'-ooooo
ooooo' ooooo
ooooo v'ooooo
PLAN
ACTUAL LINEAL FEET 1-INCH PIPE IN EACH SECTION
No. Pipes
VENTO CAST IRON HOT-BLAST HEATER
REGULAR SECTION RATINGS AND FREE AREAS
Regular 40 Inch Section, 10.75 Sq. Ft. Height 41^64 Inch. Width 9H Inch
VENTO HEATER RATINGS
REGULAR SECTION RATINGS AND FREE AREAS
Regular 60 Inch Section, 16 Square Feet. Height 6OIV16 Inch. Width 9 Inch
NARROW SECTION RATINGS AND FREE AREAS
Narrow 50 Inch Section, 9.5 Square Feet. Height 5029/33 Inch. Width 6M Inch
DETERMINATION OF GUARANTEES
Determination of Guarantees
The case often arises that a guarantee to heat a building to
a certain specified temperature must be demonstrated at a time
when the outside temperature is much higher than called for in
the guarantee. It then becomes important to know the exact
relation between the increasein outside and inside temperature
when apparatus operated to its full capacity. This relation
is
has been published for heating with direct radiation, but it

varies considerably from the results obtained with the fan sys-
tem. Naturally the rise in indoor temperature will be less
than the rise in outdoor temperature owing to the fact that the
condensing capacity has been shown to be directly proportional
to the difference in temperature between steam and air, while
with direct radiation it is not directly proportional owing to the
variation in convection currents. The same relation between
indoor and outdoor temperature may be shown to hold true
whether the system was designed to take the air from outdoors
entirely or to recirculate air within the building. The formula
expressing the relation between indoor and outdoor tempera-
ture in either case is:
_ t/(t.-ti)-Hts(ti-tiO
When the guarantee is based on an outside temperature of

0 the formula becomes
, t/(ts-tl)-htsXtl ^^^^^
tr (111)
tr = temperature of building obtained with outside

temperature ti.
ti = any outside temperature at which test is made.
tr' = temperature of building guaranteed,
t/ = specified outside temperature,

ts = temperature of steam at pressure specified.
The table on page 464 gives corresponding indoor temper-

atures as derived from equation above for various outdoor tem-
peratures with guarantees at 60 to 95 in zero weather.
The table on page 465 giving mean monthly temperatures

in different localities will be found useful in many instances in
laying out heating systems.
463
TABLE OF AVERAGE INDOOR TEMPERATURES

MAINTAINED AT VARIOUS OUTDOOR TEMPERATURES WITH 5 LBS.
STEAM PRESSURE
Outdoor
Temp.
MEAN MONTHLY TEMPERATURES
A
BUFFALO DUPLEX AUTOMATIC FEED

PUMPS AND RECEIVERS
RURPÂt^rt
""-I^^Si^^'-"- m
The unit consists of a suitably constructed cast iron receiving
tank,mounted in combination with a Boiler Feed Pump on a
common bed plate. The tank is mounted slightly above the
pump, giving a sufficient head of water above the suction valves
to insure the pump always receiving a full supply of water.
Within the tank is provided a float connected to a chronom-
eter valve controlling the steam supply to the pump. Inflow-
ing water causes float to rise, thereby opening the steam supply
and starting the pump. When the water level has been lowered,
the float automatically cuts off the steam. In this way the
condensation water is returned to the boiler as fast as it
accumulates.
c
u O Û. CB 2 c
- I
Q.C3 5 '3
C
.St n 8 O
Oê.
2-5 iS=
^
3
eg ^ E
a> tA=
E a-
^ 1/5
With Standard Boiler Feed Pumps

3
PUMPS AND RECEIVERS
BUFFALO CENTRIFUGAL AUTOMATIC FEED

PUMPS AND RECEIVERS
Centrifugal Pumps and Receivers are designed primarily

for returning condensation from low-pressure systems into boil-
ers, especially where steam pressure is so low as to prevent using
reciprocating steam pumps. Centrifugal pumps should be
especially designed for handling hot water and equipped with
enclosed type polished brass runners or impellers. Receivers
should be cast iron or similar material, strong enough to stand
50 pounds pressure. Centrifugal pumps are ordinarily equipped
with 40 gallon receivers but smaller size receivers can be used if
desired. Larger size receivers are not desirable, as the accumu-
lated water should be returned to boiler as promptly as possible
before it loses temperature.
The general method of operation of all electric driven pumps
and receivers is the same. The condensation collects in receiver
tank, raising large seamless copper float, until at a maximum
point the float, by its connection, closes the float-switch, and
an automatic starter starts the motor driving the pump. As the
pump drains the receiver the float falls, until at a minimum
point the float-switch is opened and the motor stops.
In determining proper size outfit to use it is necessary to
know amount of radiation, boiler pressure, lift and pipe friction
to boiler and details of electric current. On low steam pressure
outfits about 30 per cent, margin should be allowed in figuring
power, as water at 10 pounds pressure cannot be forced into a
boiler carrying 10 pounds steam pressure.
467
GAS HEATERS
Various forms of heaters have been devised for use in con-
nection with fans, utilizing the heat of the gases direct rather
than through the medium of a steam boiler.
The efficiency of a gas-fired steam boiler, according to tests by-
Jay M. Whitman, is seldom in excess of 65 to 70 per cent.
Some forms of gas heaters have been short-lived, no provision
having been made to prevent temperatures in the heating
surfaces so high as to destroy them in one or two heating sea-
sons. To be reasonably long-lived, the heat must be transmitted
from gas to air through surfaces which are not exposed to tem-
peratures above 1200 F., and the construction must permit
the renewal of this heating surface at least as conveniently
as in the case of a boiler. These requirements can best be met
by a design in which the heating surface proper consists of
boiler tubes expanded into heads, the gases passing through
the tubes, and the air drawn across them by the fan, while
for best economy the range of temperature for the gases is
from 1200 to 400 F. With natural gas or producer gas
fuel, this design of heater has been combined with a com-
bustion chamber provided with fire brick checker work, which,
becoming incandescent, provides for the complete combustion
of the mixture of gas and air before it leaves the chamber; a
mixing chamber in which the high temperature products of
combustion are mixed with low- temperature gases which have
already passed through the heater, and which in any desired
proportion may be recirculated by an induced draft fan; an
exhaust chamber from which the induced draft fan draws the
cooled gases, part of which are discharged and part recirculated,
and a suitable setting with boiler fronts and inspection doors,
so as to make the various chambers accessible.
With natural gas having a calorific value of 1000 B. t. u.,
the loss in the waste gases discharged at 400 is approximately
GO B. i. u., corresponding to an efficiency of 94 per cent. When
running at part load, and allowing for possible poor regulation
of the burners, such a heater still has an efficiency better than a
good steam boiler. Where producer gas is available instead of
natural gas, similar economj^ will be shown. The high tempera-
ture exhaust from gas engines, if of sufficient volume and regular
in quantity, waste gases from furnaces, or even under some con-
ditions from boiler plants, may be utilized to good advantage.
468
SECTION VI
AIR CONDITIONING APPARATUS
Air Washers
Air washers are generally used in connection with ventilating
systems for public buildings, offices and residences. Their effi-
ciency in purifying the air varies greatly wdth their construction
and also depends in a large measure upon the nature and quan-
tity of the impurities in the air to be washed. In general, the
heavier particles in the air, such as street dust for instance, are
comparatively easy to remove even with a washer of simple
construction. On the other hand, the very fine particles often
existing in city air, especially where it is taken some distance
from the ground, where the impurities consist chiefly of fine ash
and smoke particles, are exceedingly difficult to remove, and the
most efficient air washer construction is required to get satis-
factory results, or in fact, any results which wîll be worth the
cost of installing an air washing device.
Principles of Air Washer Construction
It has been found by experiment, and is now generally
acknowledged by engineers, that in washing air or gas the first
essential is to fill the chamber through which it is passed with a
finely divided spray or mist in order to get as great a contact
surface as possible between the water and air and to secure a
thorough mixture. Probably the most satisfactory way of
accomplishing this is by the use of a large number of uniformly
spaced centrifugal nozzles with large orifices to prevent clogging
with foreign material. It has been found practicable to use
orifices %o-inch in diameter in centrifugal nozzles which will
give a satisfactory division and distribution of spray and at the
same time will not clog. The nozzles should spray in the direc-
tion of the air flow.
An adequate filtering system should be provided, where the
water is recirculated, in order to remove any large obstructions
that might otherwise enter the spray system.

Note For a general discussion on the subject of Humidity, see page
28; and on Air Washing, Cooling, Humidifying and Drying, see page 67.
409
The velocity through the washer for best results should be
between 400 and 500 feet per minute. It is equally important
that the air be distributed uniformly over the entire area of the
washer. This is often difficult to accomplish and can only be
secured by means of a difTuser or distributing plate at the washer
inlet.
While some work may be done with a finely divided spray, it

cannot be depended upon alone to give satisfactory cleaning
effect. The air after having been moistened must be brought
into repeated contact with wetted surfaces and subjected to
the combined action of impact and centrifugal force. For the
best results, the air should also be divided and broken up into
as narrow layers or strata as may be possible mechanically, in
order, 1st, that as great a contact surface may be secured as pos-
sible, and 2nd, that the solid particles contained in the air shall
have as small a distance to travel as possible before coming into
contact with a wetted surface where they will be entrained.
This is best accomplished by placing in the eliminators large,
independently flooded vertical surfaces. The plates forming
this flooded surface should be placed as closely as possible, pref-
erably about 1 inch apart, arranged vertically and flooded from
the top. An extension of these surfaces should be provided
with lips for the removal of all traces of free moisture. A satis-
factory ratio has been found to be 64 sq. ft. of combined washing
and eliminating surface per 1000 cu. ft. of air per minute.
In public building work provision should be made for heating^
the spray water and controlling the moisture content of the air
in cold weather. The simplest method of accomplishing this is
to regulate the temperature of the air leaving the washer by
means of a thermostat, at the same time saturating the air by
means of a heated spray at a variable temperature.
Air conditioning apparatus for controlling the humidity of
air for manufacturing processes may be broadly classified,
according to use, into humidifiers proper, which add moisture
to the air in required amounts; and dehumidifiers, which remove
a variable quantity of moisture from the air to reduce it to the
required standard. The relative humidity of the air may also
be altered, and in a measure regulated, simply by changing its
temperature without affecting its moisture contents.

470
HUMIDIFIERS AND DEHITMIDIFIERS
Types of Humidifiers
Humidifiers may be classified into the spray and evaporative
types, the latter being divided again into direct and indirect.
The humidity of the air may also be increased by the direct intro-
duction of steam into the air supply or into the room. Since
the total heat of the vapor at atmospheric temperature is some-
what less than the total heat at steam temperature, this raises
the temperature of the air perceptibly and is therefore intolerable
in the majority of cases. Added objections to the direct use of
steam are that it frequently gives a noticeable odor and that it
is difficult to regulate! The spray and evaporative types of
humidifiers have a distinct value aside from humidifying in
their possession of a cooling effect which is in direct proportion
to their moistening effect. The direct spray type of humidifier
is distinguished from the evaporative type in that it introduces
a finely divided or atomized spray directly into the room in

constant volume, while the evaporative type introduces only
the water vapor. There is also a mixed type which discharges
both moist air and free moisture into the room.
In what may
be termed the indirect evaporative humidifier
the air is partly or entirely taken from the outside and
is humidi-
fied and conditioned before it is introduced into the room. In

the direct evaporative type the water vapor passes directly into
the air of the room. The indirect system of air conditioning is
also termed the central system, and is known commercially as
the Carrier System.
The Dehumidifier
In the dehumidifier, relatively cold spray water is used to
condense the moisture out of the air. The water is either re-
frigerated or taken from an artesian well. When the water is
artificially cooled the refrigerating coils are usually placed in a
chamber underneath the spray chamber, and the water is so
distributed as to flow uniformly over the cold surface, dropping
to the tank underneath. The dehumidifier has its sprays opposed
to the direction of air flow as in the humidifier, but differs
from the latter in having usually two sets of sprays in series in-
stead of one. Two or more dehumidifiers are frequently placed
in series when the range of air temperature is great or when an
economy of cooling water is essential.
471
Elements of Design of Humidifiers
The degree of saturation of the air leaving any type of air
washer or humidifier depends upon the intimacy of the contact
of the air and water, and upon the relation of the water tempera-
ture to the wet-bulb temperature of the entering air. It also
depends to some degree upon the length of the spray chamber
as well as upon the velocity of the air passing through it.
The size of the nozzle orifice is also a very important factor

in determining the degree of saturation obtained. In general,
the smaller the nozzle orifice the more perfect will be the humidi-
fying effect with a given quantity of water. For humidifiers it
isstandard practice to use centrifugal nozzles having a 3-32 inch
and where rotary strainers are employed for filtering the
orifice,
spray water the nozzle orifice may be reduced to 1-16 inch in
diameter to advantage.
With the centrifugal type of spray nozzles the water pressure
is a most important element affecting the degree of saturation.
The accompanying diagram shows the humidifying effect secured
with various velocities and at different pressures on the spray
nozzle, in a standard humidifier having four 3-32 inch orifice cen-
trifugal spray nozzles per square foot. This data was obtained
from a test in which the wet-bulb depression of the entering air
was maintained constant at 16. It will be noted that an in-
crease of 23^ pounds in the spray pressure permitted a greatly
increased velocity with perfect saturation, an effect which was
undoubtedly due to the increased fineness of the spray rather
than to the increase in the amount of water discharged. In
this test, as in all standard humidifiers, the water was discharged
in the direction opposite the air flow, increasing the efficiency of
saturation.
When the spray water is recirculated without heating, as in

warm weather, it remains all times substantially at the wet-
at
bulb temperature of the entering air, while the wet-bulb temper-
ature of the air leaving the washer or dehumidifier is unchanged;
therefore it follows in conformance with the theory, that when
the air is completely saturated as in the humidifier the air is
cooled to the wet-bulb temperature of the incoming air. This
cooling effect is due to evaporation and is therefore in direct
proportion to the moisture added to the air. The wet-bulb
depression in atmospheric air averages from 12 to 15 in summer,
472
DESIGN OF HUMIDIFIERS
1V6
Elements of Design of Spray Type of Dehumidifiers
Dehumidifiers may
be of the spray type previously described,
or of the surface type. A knowledge of the relation of water
temperature to the leaving air temperature in either type is
essential. In the spray type of one stage having two banks of
opposed nozzles, the air temperature leaving is practically
identical with the temperature of the leaving water, the differ-
ence never exceeding one degree in a properly designed appara-
tus. The air will always be saturated when leaving and under
some conditions there is a slight tendency to entrainment even
after thorough elimination.
The degree of entrainment is dependent upon the range of

temperature of both the air and the water. In general, the
smaller the temperature range, the less the tendency is to mois-
ture entrainment or supersaturation. This may be reduced
where a considerable lowering of air temperature is required by
passing it successively through two or more dehumidifiers in
series. When the system is properly designed, the entrainment
should not be sufficient to raise the true dew-point temperature
more than one degree.
Refrigeration Required for Dehumidifying
The heat to be removed in cooling a known weight of air

from a given temperature and moisture content to a given dew-
point temperature, is evidently the difference of the total heat
quantities contained in the air under these respective conditions.
These values of total heat are given on the charts on pages 36
and 37. It is there shown that the total of latent and specific
heat in one pound of pure air is dependent upon the wet-bulb
temperature only. The upper table on page 475 shows the
amount of refrigeration required to cool and dehumidify 1000
cu. ft. of air between various given wet-bulb temperatures and
final dew-points.
Power Required for Operating Humidifiers
The lower table on page 475 exhibits the power required to

saturate 1000 cu. ft. of air per minute at various velocities.
This is based on overcoming the resistance of the humidifier,
using a fan with a static efficiency of 45 per cent., a fair value.
474
REFRIGERATION AND POWER TO HUMIDIFY
B. T. U. REFRIGERATION REQUIRED TO COOL 1000 CU. FT. OF AIR

(MEASURED AT 70 DEO. F.) FROM A GIVEN WET-BULB
TEMPERATURE TO A GIVEN DEW-POINT
^
ELEVATION
Carrier Type "A" Washer
476
AIR WASHER DIMENSIONS
DIMENSIONS FOR CARRIER TYPE "A" AIR WASHERS
'J "-.J
ELEVATION
Carrier Type "B" Washer
478
u B-LENGTM-4'IO- H
TO SEWER
ELEVATION
Carrier Type "C" Washer
480
SECTION vn
STEAM ENGmES
In the following section will be found capacity and specifi-
cation tables of Buffalo Steam Engines.
Two diagrams are given, showing the water rate of high
speed engines and the ratio of the water rate at any cut-off as
compared to the rated water rate, when the engines are rated at
% cut-off. As an example of the use of these diagrams, we will
take a case of a 100 H. P. high speed engine with steam pressure
at 100 pounds and cutting off at ^stroke. From the curve
on page 486 we find the steam consumption will be 28 pounds
per indicated H. P., or a total of 2800 pounds per hour. In case
this was a heating job this would then be a measure of the amount
of exhaust steam available for use in the heating coils. If, for
any reason, this engine should be set to cut off at ^stroke, we
may determine the resulting steam consumption from the dia-
gram on page 487. From the point marked ^ cut-off on the
right edge of the chart, pass horizontally to the left until the
cut-off line is intersected, thence downward to the curve, and
horizontally to the left edge, where we find the water rate will be
104.2 per cent, of the rated. That is, 1.042x28 = 29.2 pounds
per I. H. P. per hour.
The per cent, of rated load that will be developed may be

determined from the scale at the bottom of the diagram. Drop
vertically from the intersection of the ^cut-off with the cut-off
line to the bottom of the chart, where it will be found that the
engine when cutting off at ^ stroke will develop 137.5 per cent,
of the rated load, or 1.375 X 100 = 137.5 I. H. P.
The table of mean effective pressures has been calculated

from actual indicator cards taken from automatic high speed
engines. These values are applicable to engines of this class
when exhausting at atmospheric pressure. In case back pres-
sure is carried on the engine a corresponding correction should
be made.
482
STEAM ENGINES
The horsepower tables for the various classes of engines give
the brake horsepower per R, P. M., the maximum speed allow-
able and the corresponding horsepower developed. There are
two factors limiting the speed of these engines; first, a maximum
allowable piston speed; and second, the maximum load allow-
able, as indicated in the dimension table of each particular
engine. In the case of the automatic engines the governors do
not operate at a speed less than two-thirds of the maximum.
There are a number of cases in the tables on pages 490 and 491
where, at certain steam pressures and cut-offs it was found neces-
sary to limit the speed to less than two-thirds of the maximum
in order to avoid overloading the corresponding engine frame.
As indicated on the tables, these engines will have to be operated
throttling, since the automatic governors will not be operative
at these low speeds.
The two tables on pages 496 and 497 are applicable to Buffalo
Class "A" automatic high speed engines. The maximum speed
for each engine is shown, together with the minimum speed when
operating with the automatic governor and also the brake
horsepower that will be developed at the different steam pres-
sures. Except as indicated these values are based on the engines
operating at one-half cut-off. Those cases marked will have
to cut off at less than one-half stroke in order to avoid overloading
the engine.
483
Spiro Driven Niagara Conoidal (Type N) Fan for Forced Draft
Spiro Driven Gas Blower Unit
These illustrations show the adaptability of direct connecting

various types of fans to Spiro Steam Turbines, which are made
in sizes from 1 to 100 horsepower and operate very economically
at a comparatively low speed.
484
M. E. p. OF HIGH SPEED ENGINES
M. E. P. OF HIGH SPEED ENGINES*

Allowance Made for 10 Per Cent. Clearance and without Back Pressure
ill
jnoH J3d ({ -u 'I J3d iUB3)S jo 'sqq
486
WATER RATES OF HIGH SPEED ENGINES
Cut-Off in Per Cent, of Stroke
O ^ O O O
uondiunsuoo J3|bav P^Jb^ JO jua^ jaj
487
O M
HORSEPOWER OF ENGINES
<oocc"5 wMop i^MO tqiop piop wpc^
>ootô6
^_ otc-H
C^CC^
oioo MOOOO tOOiO CCiNcd-^
n-SrO CO00 05(NC^ <*'0C
O
00
ooo
OOOJ
ooo ooo 000 OiCOO
CO COM CO<MiN (N(N<N (N(N(N(N
u C0-H(MCO OS-*'* ^tîO OCOCO OiiOiOO

^<N0CO 00-- COtO-* (NiOOO
o = 2.0:
iiNCOCO
OOCO lOCi-*
OO-H
-itîO
^-H(N (MCOCO
OOOiO
C0-<J<O
CO IN CO r*
< I
o
OS c^
I
CO < .0. (NOOiOCO '*0 CO O
OO^
M
00
Otvlco*
(NCOt^
1^
OiOOO ooo ooo OOiO

oooo OOO
t^ t^ t^
iCiOOO
(N CJ O O
^ = lO't'fco -^-^co cocoes cocoes (N (N C^ N C^q(N (N
:^
o _. :S 0. S iO
OSOOiM
OOÔJ (NOOIN ooo
^ lij r- Qj r* 1 ' uj
z S O fCO (NCOOO
5t^ *
O00t|<
00 1^
t*oo OSOCO coojgt^
ui < a
.
oooo OO O (N ^(Meo
^ w
O
.
H Of
u E .a COCOCit^ NMCO ;001D C00500 hoco loqcooq
^ J 2 CO -< * cJ CO CO CC
< E 1= :CO>0 O -^t^

= ^ 1 OQ
oooo OOO
OOO 000 OOO
OOO OOO oooo
o > g iOI>0<N
lO * ^ ^ ^^*
lOiOC^
COCOCO COCOCO
(N<NOO
(NIN(N(M
^S
OSCOiO<N *C5C0 cot:-o 00CO-*(N
oa < Q^ t^ 0 CO (N iC C) 00 tN.tCcO (NCOCC-*
r ô: O
oooo 1>) *
CO
OOO
1
05 CO *eooo
<NC0C0
to
TfiOt-OO
I i OQ ^
qr>;Oso o^^ cCiOOO >opoco
*' to' d CO
CO CO CO
(NCO doood (N
Tt<
OSOCOCO
f g
000i0 OOO
? I c *** OOO
lOtîOM
COCOCO COCOCO (N(NM MWMM
CI *
cor>. ot>.oi IM 00 lO 1" 00 OO OOcO C0OCOt
1^ * 00 O
C^J t^ c^j MM n Tf< 00 ts.as'f coiNt^
O -I'M CO
oooo OOO O
CO lO CJ r^ co eoc5<N
Mo lo
(M CO CO
o CO
* o
* CC 00
OQ q2'
'9iA>er>> 000000 OOO fw >o>ooooe

X X M X X M X XXX XXX
vin>etN <oaoo
489
1
I3
V3
II
S d
O ua.
U
2 s
U.
3 2
oa &

o
r
o
r
25 ^
UO g
=H g
o Oi .
< > ^
^ u
t
o
&

o
E
a.
BUFFALO LOW PRESSURE ENGINES AT THREE-QUARTER CUT-OFF
AND 3 LB. BACK PRESSURE
Brake Horsepower per R. P. M. and /Vlaximum R. P. M. and Horsepower
Allowable at Different Steam Pressures
I.
1
""
o
Z Q
UZ
. <
UO
Û
Q
O3
<t
%<
3U
OQ
o la
O W3
=i
<
>
ui-
2 Q
o z
z <
c/3 oo
<
.J
^ U.
Q O
Ss
< <
U. U
U. H
O <
> "J
> u
O Q.
vi
0.
o 2
r (^
BUFFALO HORIZONTAL ENGINES
CENTER CRANK, CLASS "A"
MAXIMUM HORSEPOWER ALLOWABLE FOR CORRESPONDING FRAME

High Pressure

SIDE CRANK, CLASS "A"

High Pressure

SIDE CRANK, CLASS "N"
3:
SIDE CRANK, CLASS "S"

BUFFALO SINGLE VERTICAL ENGINES

CLASS "O"

CLASS "A"

High Pressure
PÂN ENGINEERING BUFFALO FORGE COMPANY

CYLINDER BELOW SHAFT, CLASS "I''
Max.
Horsepower
BUFFALO DOUBLE VERTICAL ENGINES

Single Acting
&
SECTION VIII
PRACTICAL APPLICATIONS AND THE SELECTION

OF APPARATUS FOR HEATING AND VENTILATING
As has been previously shown, one of the most important
applications of fans is in connection with the heating and ven-
and public buildings.
tilation of industrial This work will be
considered under two general divisions or classes: First, to sup-
ply heat; and second, to supply ventilation, where the heating
may be done by direct radiation or by means of the fan as cir-
cumstances may determine. The first class embraces such
buildings as shops and factories. The second class is more
likely to be confined to such buildings as theatres, churches,
hotels, etc. These buildings may be supplied with ventilation
only, or with a combination of heating and ventilation, but the
fan system is seldom used in such cases for heating alone.
The following examples will serve to illustrate these various
systems, and the use of preceding rules, tables and data required
in connection therewith will be explained. There are three
factors entering into such calculations, the air required, the heat
loss due to transmission and infiltration to be cared for, and the
rise intemperature of the air above room temperature. In the
various examples two of these conditions are given and the third
is to be determined.
Class I. Heat the building, using either all return air, all
outdoor air or a mixture of the two.
In this case, the total heat
loss is assumed to be known. For heating work where part or
all of the air is returned from the room to the apparatus, it is
customary to use a heater six sections deep with low pressure

steam or five sections deep with high pressure steam. Where
all outdoor air is used, either six or seven sections may be used
with low pressure steam and five or six sections with high pres-
sure steam. Knowing the temperature of the air entering the
heater and assuming a suitable velocity of the air through the
clear area, we may find from the heater tables on pages 418 to 431
the final or leaving temperature and so determine the tempera-
ture rise. Knowing the heat loss in B. t. u. per hour and the
606
SELECTION OF APPARATUS
temperature rise, we may determine the air requirement from
the formula
Q 60^ft!-") (for heating only)

H = B. t. u. loss per hour due to transmission and
infiltration.
t2 = temperature of air leaving heater
tr = room temperature.
In ease all return air is used, the above H is a measure of the
heat required to be delivered to the air by the heating coils.
Where all or part of the air entering the heater is outdoor air at a
temperature t^, a greater amount of heat, H', will be required, due
to raising the temperature of this air from t^ to room tempera-
ture t,. This total heat, H', is a measure of the condensation
and steam requirement of the heating coils. It may be deter-
mined by the formula
, ^ QX60(t2-ti) /r u * N
heatmg)
552 ^
where H' = total heat required at the coils in B, t. u. per

hour.
Q = cu. ft. of air per min.
ti = temperature of air entering heater.
The three following examples illustrate in detail the various
steps in the calculations of heating propositions:
Example 1. Supply heat only, using all return air.
Example 2. Supply heat, using all outdoor air. This condi-

tion frequently happens when the apparatus is so located that
it is impracticable to run return ducts from the building to the
heater.
Example 3. Heat the building, using part outdoor air and
part return air.
Class II. Supply ventilation, either with or without heat.

As already stated this is the usual requirement to be met in the
case of public buildings. Here the known factors are the re-
quired amount of air for ventilation, the temperatures of the
outdoor and of the room air, and the consequent temperature
rise. The air to be delivered to the room at practically room
temperature. The amount of heat and consequent size of heater
507

isto be determined, the B. t. u. per hour to be supplied at the
heater being indicated by the formula
, Qx60(tr-ti) ,, ^., ^.
H'=-^^ (for ventilation only),,
^^
where H' = B. t. u. per hour required at heater.
Q = cu. ft, air per min. for ventilation.
tr = room temperature,
ti = outdoor, or temperature entering heater.
The amount of steam required by the heater may be deter-
mined by

T- --;
H'
latent heat of
;;
:
steam
=lbs. condensatlou per hour,
For ventilation work the final temperature of the air leaving

the heater, ts, is usually taken at the room temperature, tr,
when the room is heated by means of direct radiation, although
it iscustomary to have the air leave the heater say from two to
fivedegrees warmer to provide for radiation from the piping. In
case both ventilation and heating are required, we have a differ-
ent set of requirements. The known quantities then are the
heat loss, H, to provide for the transmission and infiltration
losses; the air quantity, Q, required for ventilation; and the
temperature of the air entering the heater. The temperature
of the air leaving the heater will have to be enough above the
room temperature to care for the heat losses, and may be deter-
re O TT
t2 =^ '
+ tr (for heating and ventilating)

v^ X oU
As already explained, the heat required at the heater will be
, Q X6 (t2-ti)
^ = 5572
Knowing the temperature rise (t2 ti) and the air quantity,
and assuming a suitable velocity of the air through the heater,
the depth of the heater may be found from the heater tables
on pages 418 to 431,
The three following examples illustrate the calculations
necessary under this class of installations.
Example 4. Supply ventilation only, with a specified air
change.
Example 5. Supply ventilation only, with a specified amount
of air per minute supplied a given number of people.
508
ELECTION OF APPARATUS
Example 6. Supply a specified amount of air for ventilation
and heat the building.
As an illustration we
assume a brick building 110X200
will
ft. and an average wall height
in size, with a 13-inch brick wall
of 20 ft. Building to be open to the roof, that is, no ceiling under
the roof, which is to be of 2-inch boards, paper, tar and gravel.
Loss from the floor to be neglected. Building to be warmed to
70 in zero weather.
From the tables of radiation coefficients we find that the
factor for a 13-inch brick wall is 0.29, for glass surface is 1.1,
and for a roof of this construction is 0.26. The total wall sur-
face will be 12400 sq. ft. of which we
assume 3000 is glass
will
surface and 9400 is brick wall. We
then have as the heat
will
loss per hour per degree difference between the room and outdoor
temperature.
Brick wall 9400 sq. ft. X 0.29 =2740
Glass 3000 sq. ft. X 1.1 =3300
Roof 23000 sq. ft. XO.26 =5960
12000 B. t. u.
or a total loss due to transmission of

Ht = 12000X70 = 840000 B. t. u. per hour.
The cubic contents of this building will be 440000 cu. ft.
Example 1. Supply heat only, using all return air.
As already shown, the heat loss from the building due to

transmission will be 840000 B. t. u. per hour, to which we will
add 10 per cent, giving the corrected loss as 924000 B. t. u. per
hour.
In a system using return air at the heater, it is customary
to allow for a certain amount of
leakage or infiltration, vary-
air
ing from once in one-half hour to once in two hours. This
depends on several factors, such as the size and construction of
the building, purposes for which it is used, etc. Assuming a
loss due to infiltration of one change in two hours, we will have
220000 cu. ft. of air per hour to be warmed from 0 to 70.
The heat required to care for the infiltration loss will be
H. = ??0^ÂZ0 = 379000B.t.u. per hour
The total heat required will then be

H = 924000 -1-379000 = 1202000 B. t. u. per hour.
509

stated, when using exhaust or low pressure steam
As already
it isconsidered good practice with return air to use a heater six
sections deep. As the air is to be returned from the room, it
will enter the heater at from 60 to 70. From the table on
page 421 we find that when using steam at five pound pressure
with an entering temperature, t,, of 60, with six sections of
heater we will have a leaving or final temperature, tg, of 145 if
the velocity through the heater is 1000 feet per minute.
If it is desired to figure closely, and the heater is so located
that there will be no loss in temperature in returning the air
that is, the air enters the heater at room temperature, or tr =

70 the above value of tg = 145 will not be correct. As the
table does not give the final temperature when the entering
temperature is 70 it will be necessary to refer to the curve on
page 433. We find from this diagram that with an entering
temperature of 70 and a velocity of 1000 feet per minute, the
final temperature, tg, will be 149. The method of using this
diagram is as follows: Selecting a temperature of 70 on the
left-hand margin of the chart pass to the right to the intersection
of the 1000 velocity curve, and dropping from here to the base
line we have a reading of 4.35 sections. Selecting a new point
of 6 + 4.35 = 10.35 sections on this scale, pass upwards to the
intersection of the vertical with the 1000 velocity curve and
then to the left, where a final temperature of 149 is indicated.
The quantity of air at 70 required as a heat carrier will then
be
^- 55.2X1303000
60(149-70)
-15^"UA.F.M.
With a velocity of 1000 feet per minute through the heater

we a clear area of 15200-7-1000 = 15.2 sq. ft.
will require From
the table on page 449 we find that a heater 4'6'' wide by 7'4'' high
will have a clear area of 15.3 sq. ft. We will then require a
heater containing six sections of this size.
As we would use a draw-through apparatus in a case like
the fan will handle the air at 149 instead of at 70, and the
this,
volume of the air will be correspondingly greater. This volume
will vary inversely as the weight per cubic foot of the air, and
the ratio of the volume at different temperatures as compared
to the volume at 70 may be found from the table on page 13.
Thus we find the same amount or weight of air at 150 will have
510
1.1512 times the volume at 70. Then the fan must be selected
on a basis of
15200X1.1512 = 17500 cu. ft. per min.

Assuming that in an installation of this nature, the total
resistance against which the fan is to operate will be equal to
one inch of water pressure, we will find from the tables of rated
fan capacity the size of fan required to deliver 17500 A. P. M.
against one inch total pressure. From the table on page 229
we find that a No. 8.Niagara Conoidal fan will deliver 17340
cu. ft. minute against one inch total pressure when
of air per
operating at 253 R. P. M. and will require 3.87 H. P.
This horsepower based on the values given in the capacity
table is for air at 70 while the fan is to handle air at 149. At
constant capacity and speed the horsepower will vary approxi-
mately inversely as the absolute temperature, hence it will
require less than the rated horsepower to handle this air which
is at 149. The actual brake horsepower required by the fan
will then be
3-X^|^ = 3.36H.P.
If the fan is to be motor driven, it will be necessary to select
a motor of the next larger standard size, or 5 H. P. On account
of the slow speed this fan should be belt driven.
Example 2. Heat the building, using all outside air at 0.

As already explained under Example 1, when using return
air it is customary to add 10 per cent, to the computed heat loss
and also provide heat necessary to care for the infiltration loss.
When using all outdoor air no provision is made for infiltration,
but the calculated heat loss is generally increased by a greater

margin say 25 per cent. Sufficient heater must be provided
to raise the temperature of the air from zero to room tempera-
ture, and enough higher to care for the heat loss from the build-
ing.
Adding 25 per cent, to the radiation loss gives 1050000

B. u. per hour required for heating.
t. As shown by the heater
table on page 420, with a velocity of 1000 feet per minute six
sections of heater will raise the temperature of the air from 0 to
117. Allowing a 2 drop due to the radiation loss from the
piping gives a warm air temperature of 115 delivered to the
room.
511
Then the quantity of air at 70 required will be
^- 55.2X1050000
60(115-70)
-^l^SOA.l-.M.
This means approximately a 20 minute change. If possible

the apparatus should be so arranged that return air may be used
in the morning in order to heat up rapidly.
As the velocity through the heater is to be 1000 feet per
minute, this calls for a clear area of 21.45 sq. ft. From the
table on page 449 we find that a heater 6'0"X7'10'' will be the
nearest standard size.
Example 3. Heat the building, using one-half outside air at

0 and one-half return air at 70.
As already shown the heat loss from the building will be
840000 B. t. u. to which we will add 10 per cent., making the
total loss 924000 B. t. u. per hour. This is the same as the loss
figured for in Example 1, but since we are to use only one-half
return air, we will allow but half the infiltration loss. This will
require 190000 B. t. u. per hour. The total heat to be provided
will then be
H = 924000-^190000 = 1114000 B. t. u. per hour.

Since half the air entering the heater is to be at 0 and half
at 70, we will assume an average of 35. As none of the heater
tables show the temperature rise for an entering temperature of
35, we will make use of the diagram on page 433. Assuming a
velocity through the heater of 1000 feet per minute, we pass
from a temperature of 35 on the left edge of the chart to the
intersection of the horizontal with the curve marked 1000 feet
per minute. Dropping from here to the base line we have a
reading of 2.65 sections. Assuming that we will use a heater
six sections deep we will point off a new location on the base
line at 2.65-1-6.0 = 8.65 sections. Passing from here vertically
to the intersection of the 1000 velocity curve and thence to the
left edge of the diagram shows a temperature of 132 for the air
leaving the heater. Allowing for 2 drop due to loss of heat
from fan housing and piping gives an effective warm air tempera-
lure of 130.
The quantity of air at 70 required will then be

^ 55.2 X 1114000 _,__ . p -,
Q= 60(130-70)
=17100 A. P.M.
512
SELECTION OF APPARATU
As we are to use a velocity of 1000 feet per minute through
the clear area, a heater with a clear area of 17.10 sq. ft. will be
required. From the table of heater dimensions on page 449,
we find that a section of 5'0"X7'10'' will have a olear area of
17.7 sq. ft. We will then use a heater 5'0''X7'10" six sections
deep.
As shown above, the air required will be 17100 cu. ft. per
minute at 70, but if the fan is arranged to draw through the
heater it will handle this air at 132. The volume of the air
will be greater than for the corresponding weight at 70, the-
ratio as given in the table on page 13 being 1.114. This means
that the fan will be required to handle
17100X1.114 = 19100 A. P. M. at 132.
It isprobable that the static resistance of an installation of
this nature will not be over one inch, due to the resistance of
0.574 inch through the heater, the friction loss in the piping
and any entrance or discharge losses that may exist. By refer-
ring to the static pressure tables of the Niagara Conoidal Fans
on pages 247, 249 and 251, we find that we may use either a No.
7, 8 or 9 fan. Either of these fans may be used to give approxi-
mately the required capacity, but we see from the following
summary that the outlet velocity, speed, and horsepower will
be different in each case.
Size
In any case, the horsepower as given on page 513 would be
based on air at 70, while the fan is to handle the air at 132. That
is, while the fan is to handle the 19100 cu. ft. per minute, due to
the higher temperature, the density and therefore the horse-

power required will be less than for the same volume at 70. At
constant capacity and speed, the horsepower will vary directly
as the density of the air, and approximately inversely as the
absolute temperature. We will then have the above horse-
power decreased by the_ratio
460 + 132^
460 -H 70
That is, the above horsepower should be divided by 1.114.

Example 4. Supply ventilation only, with a ten minute air
change.
In is to be taken care of
this case the heating of the building
by and the air required for ventilation will be
direct radiation,
taken from outside at 0 and introduced into the room at room
temperature, or 70. We are not concerned with the heat loss,
but merely an amount of air equal to the cubic con-
in raising
tents of the room through a 70 temperature rise once in ten
minutes or six changes per hour. Then we will have as the air
at 70 required
440000 ..^^^
44000 cu. ,^
ft. per mm.
10
Assuming that this building is to be used for a factory, and

inasmuch as the heater will be comparatively shallow with a
corresponding low pressure drop, we may use a velocity of 1200
feet per minute through the heater. The total pressure against
which the fan in such an installation is likely to operate will be
about one inch. The temperature of the air leaving the heater
should be about five degrees above room temperature to allow
for radiation loss from the piping, etc.
From the heater table on page 420 we find that with five
pound steam pressure and with 1200 velocity a temperature of
81.8 will be obtained with four sections of standard heater and
will be the nearest temperature obtainable at this velocity.
514
Since we are to handle 44000 cu. ft. per minute at 1200 velocity
we will require a heater having a clear area of
44000 _ _ _
= 36.7 _ .,
sq.ft.
-j2oo-
From the table of heater dimensions on page 449 we find that

a section 9'6'' X 8'4'' has a clear area of 36.7 sq. ft. so this will be
the size to use. We will then require a heater Q'G" X 8^4'' and four
sections deep.
In a large room of this construction it is customary to use a
draw-through system, attaching the piping directly to the fan
outlet. As already stated, the total pressure required for an
installation of this character using but four sections of heater
would probably run about one inch.
Referring to the table of fan capacities we may select a fan
that will deliver 44000 A. P. M. against a pressure of one inch.
From page 208 we find that a 160-inch Planoidal Exhauster will
deliver 41220 A. P. M. against one inch at a speed of 164 R. P.
M. and require 14.2 H. P. The 44000 A. P. M. required is 106.5
per cent, of the above rated capacity so it will be necessary to
operate at a speed of 164X1.065 equals 175 R. P. M., and the
power required will be 14.2 X (1.065)3 equals 17.2 H. P.
This method of arriving at the speed and power required is
only approximate, since when operating at other than the rated
point the pressure will not be constant. For small increments
of over or under load the speed may be changed slightly to bring
the pressure to the desired amount, but for accurate work the
method of using the diagram on page 215 as explained under
Example 5 should be followed.
Example 5. Supply ventilation only, with 30 cu. ft. of air
per minute for each of 500 occupants. This will require
30 X 500 = 15000 A. P. M. at 70
which will be equivalent to 440000-^15000 = 29 or approxi-

mately two changes per hour.
With a velocity of 1200 feet per minute through the heater,
this will require a clear area of 12.5 sq.ft. From the table on
page 449 we find that a section of 4'0''X6'10" may be used. As
the same temperature rise is required as in Example 4, the heater
515
must be four sections deep and the final temperature 81.8. If
the heater and fan are located in the building so that there will
be little or no radiation loss from the piping, the air may leave
the heater at approximately 70. We see from the heater table
that with a velocity of 1000 feet per minute three sections will
give a final temperature of 69.5. By selecting a heater with a
clear area slightly greater than indicated by 1000 velocity, we
will obtain a final temperature somewhat above 70. From
the table of heaters we find that a 4'6" x I'A" section will have a
clear area of 15.3 sq. ft. and give a velocity of 980 feet per min-
ute through the clear area. Thus we see that when using a final
temperature of about 70 we may select three sections deep of
4'6" X 7'4" heater.
As we are using a low velocity and a heater only three sec-

tions deep, the loss in pressure through the heater will be only
0.287 inch, and is probable that the static resistance of the
it
entire system not be over 0.6 inch. As under rated condi-

will
tions the static pressure of a Planoidal Exhauster is 79 per cent,
of the total pressure, if we use this type of fan operating at rated
capacity the total pressure developed would be 0.6 -r- 0.79 =0.76
inch or approximately ^
inch.
From the table of fan capacities on page 207, we find that a

100-inch Planoidal Exhauster will deliver 13940 A. P. M. at
228 R. P. M. against %
inch total pressure and require 3.6 H. P.
As 15000 A. P. M. is required, it will be necessary to operate
this fan at greater than its rated capacity. We note from the
diagram on page 214 that if this style of fan is operated at con-
stant speed beyond its rated point the pressure will be less than
the rated, so it will be necessary to operate at a speed corres-
ponding to a certain higher pressure, in order to still have the
required pressure when working over the rated capacity. The
speed and horsepower to meet the required overload condition
may be found approximately by means of the diagram on page
214, but as explained in the example on "Fan Selection" on
page 183, the more accurate method is to use the diagram on
page 215.
The outlet area of a 100-inch Planoidal Exhauster is 8.75
15000 A. P. M. the outlet velocity will be 1715 feet
sq. ft., so at
per minute and the corresponding velocity pressure equals 0.183
inch. Since the static resistance of the system is 0.60 inch, the
516
rated total pressure will be 0.60 + 0.183 = 0.783 inch. The ratio
of static to velocity pressure = 0.6 0.183 = 3.28.
-;- From the
diagram on page 215 we find that with the above ratio we will
be operating at 105 per cent, of the fans rated capacity, with
102.5 per cent, of the rated H. P., and the speed will be the rated
speed for this fan when developing a total pressure of 0.783 inch.
This fan will give the following rated performance at the two
different total pressures.
0.75 in. -13940 A. P. M.-228 R. P. M. -3.6 H. P.
0.783 in. -14250 A. P. M. -233 R. P. M.-3.85 H. P.
But 15000 H- 1.05 = 14300 A. P. M. as the rated capacity required,
which is practically as given for 233 R. P. M. Then the power
required under the overload condition will be
3.85X1.025 = 3.95 H. P.
From the above we see that the 100-inch fan will deliver
15000 A. P. M. against a static pressure of 0.6 inch at 233
R. P. M. and require 3.95 H. P.
Example 6. Heat the building and supply a 12}^ minute
air change for ventilation.

The outdoor air to be handled by the fan will be
140000^33200 A. P.M. at 70
This air must be raised to room temperature for ventilation,

and enough higher to supply the heat lost by radiation and
leakage.
As already shown the heat lost by radiation from this build-
ing will be 840000 B. t. u. per hour. It is customary to allow
an extra 10 to 50 per cent., depending on the construction of the
building and the purposes for which it is used, to care for the
heat lost by leakage, opening of doors, and similar causes.
Allowing an extra 30 per cent, we will have as the total heat loss
H = 840000 + 252000 = 1092000 B. t. u. per hour.
To determine the final temperature required we will have
55.2.x 1092000
35200>^60~ + ^^ = ^^-^
^'
517
From the heater table on page 420 we find that with a veloc-
ity of 1000 feet per minute five sections of heater will raise the
temperature of the air from 0 to 103. As we are to handle
35200 A. P. M. at 1000 velocity, a heater having a clear area
of 35.2 sq. ft. will be required. From the table on page 449 it
may be seen that a heater section S'O^xS'lO'' has a clear area
of 35.3 sq. ft. so this will be the size to use, the heater being five
sections deep. In case this heater is too tall for the particular
space it is to occupy, we may use two sections placed back to
back, each having a clear area of 17.6 sq. ft. This would call
for ten sections of 5'0''x7'10" placed five sections deep.
COMBINATION FAN, ENGINE AND HEATER TABLES

The four tables on pages 520 to 523 indicate what are con-
sidered as being the proper combinations of heaters and engines
for the different sizes of Planoidal and Niagara Conoidal fans.
The cubic feet of air at one inch for public buildings and two
inches for industrial installations may be considered as the prob-
able maximum conditions to be encountered.The engine sizes
given are for direct connection, and in most cases could be
made to answer for these extreme conditions. Several sizes
and styles of engines are given in order that a choice may be
made to meet different requirements. For instance, where a
7x7 and an 8 x 8 inch cylinder are given for the same size fan,
a higher steam pressure would be required for the smaller cylin-
der. A steam pressure of from 20 to 25 pounds will be required
for the low pressure, and from 80 to 100 pounds for the high
pressure engines.
The heater given in the table indicate the proper heaters
sizes
for use with the different sizes of fans. A heater should be
selected with a clear area that will give the desired velocity of
the air through the heater. This will range anywhere from 800
to 1200 feet per minute, depending on the conditions. Lower
velocities should be used for public buildings than for industrial
installations.
518
510
WITH PROPER COMBINATIONS OF HEATERS AND ENGINES FOR PUBLIC
BUILDINGS AND INDUSTRIAL INSTALLATIONS
Size
NIAGARA CONOIDAL (TYPE N) FA.NS
WITH PROPER CO.nBINATIONS OF HEATERS A.ND ENGINES FOR PUBLIC
Fan
No.
PART V
APPENDIX
Complete directions for conducting Fan Installation Tests are
included in this part.
This section also includes complete specifications and guar-
antees for various types of fans, heaters, piping, engines,
motors, air washersand humidifiers; and detailed dimensions
of Planoidal Steel Plate, Niagara Conoidal Multiblade and
Turbo-Conoidal High Speed Multiblade Fans.
Data for the design of chimneys, with table giving size of
chimneys with appropriate horsepower of boilers is included.
Miscellaneous engineering data is also given, including size
of steam pipes, area of circles, temperature and pressure conver-
sion tables, steam tables, logarithms, dry kiln capacities, many
useful factors, etc.
An extract of the report of the committee of the Am. Soc. of
H. & V. Engrs. on "Standards for Ventilation Legislation for
Motion Picture Show Places" as presented in Jan., 1913, is re-
produced.
A very complete and thorough index and cross-index, in
addition to "Outline of Contents" in Part I, is given in this part.
Note
All temperature? given in this book are in degrees Fahrenheit un-
less otherwise specified.
524
FAN INSTALLATION TESTS
DIRECTIONS FOR MAKING FAN
INSTALLATION TESTS
The general subject of fan testing has been discussed on
pages 190 to 204, with complete directions for using the pitot
tube and for laying out a traverse of a pipe or duct. The de-
tailed methods to be used in making a test on a fan installation,
together with detailed instructions for working up the results of
the test, will be given in the following:
Measuring the Air Quantity
1. Traverse over outlet or inlet pipe.
2. Traverse over fan outlet.
The velocity of the air and the quantity delivered should be
determined by means of the pitot tube as explained on page 190.
Whenever the nature of the installation makes it possible the
most accurate results are obtained by making a traverse in the
discharge or inlet duct. These velocity pressure readings should
be taken at least 10 diameters, in the direction of the air jflow,
from the fan outlet, or from any bend, elbow, change in section,
or other detail of construction that will cause a disturbance in
the flow of air preceding the point at which the readings are
taken. Readings at the fan inlet should be taken one diameter
or more from the inlet.
While a traverse of the pipe is preferable, it is usually im-
practicable in actual installations to find a point where the air
flow is undisturbed. In such cases a traverse with the pitot
tube over the outlet area of the fan should be made.
The outlet area should be divided into 25 or more equal
rectangles (5 or more each way) and velocity pressure readings
taken in the center of each small area. The velocity correspond-
ing to each of these velocity pressure readings should then be
calculated as explained on page 18, and an average taken of
these velocities. This gives the average velocity over the entire
fan outlet, which in turn when multiplied by the area of the
outlet in square feet gives the quantity of air delivered in cubic
feet per minute. It will be noted that the average velocity over
the outlet should be obtained by taking an average of the various
velocities and not of the various velocity pressure readings, since
the velocity varies as the square root of the pressure.
In taking the velocity pressure by means of a pitot tube, the
connection between the two legs of the tube and the two sides of
525
the gauge should be the same, whether the readings are taken
in a duct either on the inlet or on the outlet side of the fan.
Measurement of Pressure Produced by the Fan
The between total, static and velocity pressure as
relation
produced by a fan has been explained on page 176. The meas-
urement of the total or dynamic pressure is of especial import-
ance since this is the pressure upon which the efficiency of the
fan is based. This pressure is a measure of the total energy
imparted to the air by the fan and is the difference between
the average absolute total or dynamic pressure of the air
on the two sides of the fan. The absolute pressure expressed
in inches of water for air at 29.92 inches barometer will be the
measured pressure plus 407 inches. The absolute total pressure
at any point consists of the absolute static pressure plus the
velocity pressure.
Since the total dynamic energy at any point in a stream of air
is measured by the absolute total or dynamic pressure at that
point, the gain or loss in energy between two points in a stream
of air is measured by the difference between the absolute total
pressures at these points. It follows that the total energy im-
parted by a fan to a given quantity of air is measured by the
difference between the absolute total pressures measured at the
inlet and outlet connections respectively and that the total or
dynamic pressure produced by a fan is equal to the difference
between the absolute total or dynamic pressures at these two
points. Thus if an exhaust fan gives static and velocity pressure
readings on the inlet side of 3" and 0.5" respectively, the abso-
lute total pressure at the inlet will be (407-3)4-0.5=404.5".
If the static and velocity pressures on the outlet side are 2" and
V respectively, the absolute total pressure at this point will be
(407 + 2) + 1= 410. Then the total pressure against which the
fan is operating will be 410 404.5= 5.5". This may be expressed
in gauge pressure as follows:
The static pressure at fan outlet

+ the velocity pressure at fan outlet
Total pressure of
+ the static vacuum (draft or suction)
fan equals
(or static pressure) at the fan inlet
the velocity pressure at fan inlet.
526
The static pressure produced by a fan equals the total pressure
minus the velocity pressure at the fan outlet. It follows that
The static pressure at the fan outlet
Static pressure + static vacuum at fan inlet (or static
of fan equals pressure)
velocity pressure at fan inlet.
The Difference
in static pressure
and
The static pressure produced by the fan
at the inlet
outlet of the fan
+the velocity pressure at the fan inlet.
equals
It should be noted that the static vacuum at the fan inlet

minus the velocity pressure at the fan inlet is the total pressure
at this point.
The static pressure as produced by the fan in accordance
with the accepted definition does not necessarily correspond to
the difference in static pressures as measured at the fan inlet
and outlet respectively.
(a) Where the velocity in the inlet connection is negligible,
the difference in static pressures at the inlet and outlet of the
fan corresponds to the static pressure produced by the fan.
(b) If the inlet and outlet connections to the fan are equal,
the difference in the static pressures will not correspond to the
static pressure produced by the fan, but will be equal to the
total pressure produced by the fan when measured as previously
described.
(c) the fan inlet connection is smaller than the outlet
If
of the fan, the difference between the static pressure readings
taken at the inlet and outlet will be greater than the total pres-
sure produced by the fan.
This apparent discrepancy in the static pressures is due to
the fact that a certain amount of static pressure may be pro-
duced by conversion from velocity pressure and is not produced
by the fan itself. The static pressures specified for the various
fans assume the fan to be exhausting from a large chamber
having a negligible air velocity. Itevident that the pressures
is
are greatly modified when exhausting through a duct.
627
The proper arrangement of the gauges and the points at
which the readings should be taken are shown on the drawing,
page 528, which represents a fan drawing through one duct and
blowing into another. A pitot tube is shown at A with the
central or impact opening connected by a rubber hose to one
side of a water column or draft gauge B. (For description of
pitot tubes see page 190.) A second pitot tube is shown at C,
having its impact or total pressure leg connected to the other
side of the same gauge B. A cross connection is also made from
this impact connection from pitot tube C to one leg of another
gauge D and a connection made between the static leg of pitot
tube C to the other side of this gauge. The readings obtained
at D are then the difference between the total and static pres-
sures, or the velocity pressure at the fan outlet. The gauge B
gives the difference in total pressures at the two points A and
C, or the total pressure developed by the fan. The static pres-
sure will be the difference between the total and velocity pres-
sure readings.
As already explained, a traverse should be made over the fan
outlet, and the velocity pressure read at each point. The cor-
responding velocities should then be determined for each read-
ing (see page 18) and an average taken of these various velocity
readings. The velocity pressure corresponding to this final
average velocity will be the average velocity pressure reading
at the fan outlet, which on being added to the static pressure
gives the total pressure on the fan.
Method of Calculating Fan Tests

The following are the various factors entering into the calcu-
lation of the performance of a fan as based on the results of a
test. The object of such a test is to determine the pressure
developed, the air quantity delivered, the power required and
the efficiency of the fan. The various steps in the calculation
may be stated as follows:
1. The velocity pressure is the pressure corresponding to
the average velocity over the area of either the inlet or outlet of
the fan.
2. The total static pressure produced by the fan is the arith-
metical sum of the static readings on the two sides of the fan
minus the velocity pressure at the fan inlet.

3. The total or dynamic pressure is the sum of the velocity
and static pressures.
4. The air quantity is the product of the average velocity
pressure times the area of the duct in which the readings were
taken,
5. In case the fan is motor driven the total power input will
be the product of the volts times the amperes as shown by the
meters connected at the motor.
6. Careful readings of the speed of the fan and motor should
be made at time of taking other readings.
7. The PR loss in watts for the current taken by the motor
may be determined by means of a voltage drop test made with
the wheel blocked. The I^R loss is frequently assumed as being
from 2 to 3 per cent, of the full load current of the motor.
8. The belt and bearing loss may be determined by removing
the wheel from the shaft and taking power readings. Where
this is impractical this loss may be assumed at from 3 to 5 per
cent, of the power input, depending on the bearing conditions
and on whether the fan is direct connected or belt driven. Some
allowance should also be made for belt slip.
9. No load power readings should be obtained from the
motor by disconnecting it from the fan or throwing oif the belt.
This no load current may frequently be determined from the
characteristics of the motor.
10. The actual power consumed by the fan will then be the
total watts input minus the no load watts and the various losses
enumerated above, equals (5) [(7) + (8) + (9)].
11. The brake horsepower of the fan will be the net watts
from item 10 divided by 746.
12. In case the fan is driven by means of a steam engine,
indicator cards should be taken with the fan in operation and
with the wheel removed or disconnected. The difference be-
tween the two sets of cards will give the brake horsepower con-
sumed.
13. The air horsepower is the product of the air quantity
handled in cubic feet per minute times the pressure in inches of
water times a constant 0.000157 (see page 175).
14. The total or dynamic efficiency is the ratio of the product
of the air quantity times the total or dynamic pressure in inches
times 0.000157 divided by the brake horsepower.
,
Total

^ , en. =
A. P. M.
^^
tt in
X total press,
-,
Brake H. r>
P.
in. X 0.000157
530
SPECIFICATIONS
15.The static efficiency is the ratio of the product of the
air quantity times the static pressure in inches times 0.000157
divided by the brake horsepower and may be expressed as above
by inserting static for total pressure in the formula.
It frequently occurs that a fan is guaranteed to give a certain
performance under other than actual test conditions as to speed
and temperature. In this case the test results should be cor-
rected to the guaranteed speed and air density. As shown on
page 179, the pressure developed will vary as the square and the
power consumed as the cube of the speed. Both the pressure
and power consumed will vary directly as the density of the air
and should be multiplied by the ratio of the air densities under
the two conditions. The density of the air may be determined
from the table on page 17. These corrections should be made
to the average pressure and the net power readings, before the
air quantity and efficiency of the fan are calculated.
SPECIFICATIONS
STEEL PLATE FAN
Furnish and erect ( ) inch steel plate three-quarter
(or full) housing, discharge exhauster (or blower) having
a capacity of cubic feet of air per minute delivered against a
static (or total) pressure of inches at a speed of
R. P. M., and requiring not over H. P.
Housing to be constructed of the best commercial steel plate
No.
gauge, with riveted lap seams and braced by vertical

and horizontal angle irons, x and with angle iron base
,
frame,
x drilled for holding-down bolts.
,
Fan to have one two)

(or inches in diameter and an
inlet
outlet X inches. fan has a single
If (exhauster)
inlet
there shall be a cone extending inward to the inlet of the blast-
wheel, so as to gradually increase the velocity of the air entering
the wheel and so reduce the loss at entrance.
Blast-wheel to be inches in diameter, constructed with a
heavy cast iron hub into which T-iron arms are cast, firmly
mounted by means of a key and set screw on a steel shaft
inches in diameter. The blades to be made of No. gauge steel
631
plate riveted to x T-iron spider arms
cast into the hub,
and to be tapering wider at the inlet than at the peri-
in shape,
phery. Side sheets of wheel to be flanged outward at the inlet
and riveted to sides of blades.
Blast-wheel to be carefully balanced to prevent vibration.
Bearings to be spherical self-aligning ring-oiled type, lined
with best quality babbitt, and so designed as to allow easy ad-
justment for wear. Bearings to be provided with large oil
reservoir, and is mounted in the fan inlet it is
in case a bearing
to be provided with suitable arrangement for preventing oil from
being drawn along shaft and into the fan by the entering air.
MULTIBLADE FAN
Furnish and erect ( ) No. multiblade three-quarter
(or full) housing, discharge single (or double) width fan
having a capacity of cubic feet of air per minute delivered
against a static (or total) pressure of inches with a velocity
through the fan outlet of feet per minute, at a speed of
R. P. M., and requiring not over H. P.
Housing to be constructedof the best commercial steel plate
No. gauge, with riveted lap seams and braced by vertical
,
frame, x drilled for holding-down bolts.

,
Fan have one (or two) inlet

to inches in diameter and
an outlet x inches. If a double width fan is used, the
wheel is to be composed of two separate single width wheels
mounted back to back. Each inlet to be fitted with an inlet
cone in the space between housing and wheel, having a minimum
clearance with the flared inlet of the blast-wheel. In order to
obtain the greatest possible conversion of velocity head at tip
of blades into static pressure at fan outlet, the inner edge of the
outlet is to be approximately tangent to periphery of wheel, and
the height of the outlet approximately equal to wheel diameter.
Blast -wheel to bê of the forward curved multiblade type,
having thirty-two blades of No.
gauge steel plate riveted at
the back to a boiler plate disk which in turn is to be hot-riveted
to a conical cast iron hub. These blades to be connected by a
flange at the inlet edge of the wheel. Hub is to be attached to
the shaft by key and set-screws and to the inlet flange by four
heavy tierods. The mean diameter of the blast-wheel to be
inches.
532
SPECIFICATIONS
Blast-wheel to be carefully balanced to prevent vibration.
The heel or inner edge of the blades to be so arranged as to
give a decreasing inlet diameter from front to back in order to
give a uniform radial velocity through the wheel. The angle
of the blades at entrance shall vary across the width in order to
insure the entrance of air with the least possible loss by shock.
The curvature of the blades to be such that at normal or rated
capacity the air will leave the tips with a velocity pressure
approximately twice the pressure corresponding to the peri-
pheral velocity of the wheel, in order to reduce the required
speed of rotation.
with best quality babbitt, and so designed as to allow easy ad-
justment for wear. Bearings to be provided with large oil reser-
voir, and in case a bearing is mounted in the fan inlet, it is to be
provided with suitable arrangement for preventing oil from
being drawn along shaft and into the fan by the entering air.
HIGH SPEED MULTIBLADE FAN

Furnish and erect ( ) three-quarter (or full) housing,
discharge single (or double) width high speed multiblade fan
having a capacity of cubic feet of air per minute delivered
against a static (or total) pressure of inches at a speed
suitable for direct connection to motor specified, and requiring
not over H. P.
Housing to be constructed of the best commercial steel plate
No. gauge, with riveted lap seams and braced by vertical
,
frame, x drilled for holding-down bolts.

,
Fan have one (or two) inlet

to
inches in diameter and an
outlet X inches. If a double width fan is used, the
wheel is to be composed of two separate single width wheels
mounted back to back. Each inlet to be fitted with an inlet
cone in the space between housing and wheel, having a minimum
clearance with the flared inlet of the blast-wheel. In order to
obtain the greatest possible conversion from the high velocity
pressure at tip of blades into a correspondingly high static
pressure at the fan outlet, the inner edge of the outlet is to be
approximately tangent to periphery of wheel, and the height of
the outlet approximately equal to wheel diameter.
533
Blast-wheel to have thirty-two curved blades of No.
gauge steel plate riveted at the back to a boiler plate disk which
in turn is to be hot-riveted to a conical cast iron hub. These
blades to be connected by a flange at the inlet edge of the wheel.
The hub to be attached to the shaft by key and set-screws and
to the inlet flange by four heavy tierods. Blast-wheel to be
carefully balanced to prevent vibration.
The heel or inner edge of the blades to be so arranged as to
give a decreasing inlet diameter from front to back in order to
give a uniform radial velocity through the wheel. The angle
of the blades at entrance shall vary across the width in order to
insure the entrance of the air with the least possible loss by shock
at this point. The angle
of the blades at the tip, or periphery
of the wheel, to be such that a uniform delivery and pressure
will be obtained across the width of the wheel.
with best quality babbitt, and so designed as to allow easy
adjustment for wear. Bearings to be provided with large oil
reservoir, and in case a bearing is mounted in the fan inlet, it is
to be provided with suitable arrangement for preventing oil
from being drawn along shaft and into the fan by the entering
air.
FAN SYSTEM HEATER

Furnish and erect four-row sections pipe of coil fan system
heater, each section to be long x
ft. in. ft. in.
high. Each section to have heating surface equivalent to
lineal feet of 1-inch pipe, and a clear area of sq. ft. for
the passage of air.
The heater bases are to be of cast iron of uniform thickness,

with heavy box section and extra heavy tops drilled and tapped
for 1-inch pipe on 2^-inch centers, adjacent rows to be staggered
so as to bring the air in intimate contact with the heating sur-
faces. For the purposes of accelerating the circulation, the
base is to be provided with a partition separating the inlet from
the return space. Steam and drip connections to be tapped as
may be directed. Pipes to be threaded at each end with stand-
ard dies and screwed into base. Sections to be tested and made
tight at 100 lbs. cold water pressure.
Heater casings to be of No. 18 gauge steel plate, stififened
at all edges and seams with 1 1^" x 1 ]/^'' angle iron and extended
534
SPECIFICATIONS
to connect with the fan. The heater casing is to extend to the
foundation so as to entirely enclose the cast iron bases, prevent-
ing radiation losses. Casing to cover both sides, top, and bottom
of the heater.
A cast iron steam receiver is to be furnished with companion
flanges tapped for individual connections to the several sections,
and with a flanged opening for main steam supply.
FAN ENGINE
Furnish and erect ( ) x
horizontal (or vertical) center (or side) crank engine for belt
drive (or direct connection). Engine to have cylinder
inches in diameter by inches stroke, and to operate at
R. P. M. with initial steam pressure of
lbs. Steam pipe to
be inches with exhaust pipe inches in diameter.
Engine to have a balanced piston valve so constructed as to
take up wear in the surface of valve and valve seats. The
eccentric rod is to connect to the valve stem crosshead by phos-
phor bronze bearing. Crankshaft to be a steel forging to which
cast iron counter balance disks are solidly fitted.
Connecting rod to be of steel, with locomotive type end for
crank pin, and solid end for carrying the crosshead pin boxes.
The crosshead pin boxes are to be of phosphor bronze. Crank
pin boxes are to be of cast iron lined with the best babbitt metal.
Both ends of the connecting rod are to be provided with adjust-
ment for taking up wear on the pin.
Crosshead guides are to be bored and have ample bearing
surface.
Crossheads to be of cast steel, fitted with wedge-adjustable
shoes for taking up wear and keeping the wrist pin in alignment
with the cylinder.
All running surfaces are to be true to form and well polished.
The engine is to be filled and given a well finished painted
surface before leaving the factory.
The following fittings to be furnished: Throttle valve, sight
feed lubricator, necessary oil and grease cups, and full set of
wrenches.
DUCT WORK
To be constructed and installed in accordance with Drawing
No. . Make all sheet metal ducts of best quality galvanized
steel sheets, with slip jjpints in the direction of the air flow, rec-
^35
tangular ducts to have standing seams, and wide ducts to be
stiffened by angle irons where necessary.
All round pipes of less than 5}4 to 8 sq. ft. of No. 24 gauge;
8 to 101^ sq. ft. of No. 22 gauge; IQi^ to 133^ sq. ft. of No. 20
gauge; 13^ to 223^ sq. ft. of No. 18 gauge; above 22}/^ sq. ft.
of No. 16 gauge.
All rectangular pipes less than 18 inches wide are to be made
of No. 26 gauge; from 19 to 30 inches of No, 24 gauge; from 31
to 60 inches of No. 22 gauge; from 61 to 118 inches of No. 20
gauge; above 118 inches wide of No, 18 gauge.
No square turns are to be used at any point where it is pos-
sible to use curves, so as to offer the least possible resistance to
the passage of air. All joints are to be smooth and tight, and
all pipes are to be firmly hung and rigidly fastened in place.
The work is to be left in first-class condition throughout.

Each branch rectangular duct is to be provided with a
damper and quadrant which may be set and locked in position.
Round branch outlets to have adjustable butterfly dampers.
After erection test and set dampers for proper air distribu-
tion.
DIRECT CURRENT MOTOR
Furnish ( ) H, P, Volt direct current motor
for direct (or belt) connection to fan at the speed specified.
Motor to be of standard construction and equipped with starting
rheostat, and furnished with
a rheostat for starting duty only.

b combined starting and speed regulating rheostat, capable
of reducing the speed 50 per cent, below normal by armature
resistance.
c combined starting and speed regulating rheostat capable
of increasing the speed per cent, above normal by weakening
the field circuit.
If direct connected to fan, motor is to be furnished without
base but with flanged coupling, key-seated, faced
rails or pulley,
and polished.
If belted to fan, furnish pulleys of the proper size and single
leather endless belt.
ALTERNATING CURRENT MOTOR

Furnish ( ) H. P. Volt cycle phase
motor of standard construction with starter complete.
536
SPECIFICATIONS
If direct connected to fan, motor is to be furnished without
base rails or pulleys, but with flanged coupling, key-seated, faced
and polished.
If belted to fan, furnish pulleys of the proper size and single
leather endless belt.
SUBBASES
For fans direct connected to motors, furnish heavy steel
plate subbase to be made tapering and with rounded corners and
fitted with continuous angle iron extending around the base.
Subbases must be braced inside and provided with hand hole
for bolting down' the motors. After erection the contractor is
to fill the subbases with concrete to prevent transmission of
sound.
CARRIER TYPE "A" AIR WASHER AND HUMIDIFIER

Furnish and erect where shown on plans one (1) air washer
of a design as specified herein. Washer is to have a capacity of
cu. ft. A. P. M.
The velocity of the air through the washer shall not be
greater than 500 ft. per minute and the total guaranteed resist-
ance of the washer shall not exceed .25 inch water. Washer
is to be 7' 2%6 " long, wide and high.
Casing. The washer shall be constructed of galvanized iron
of No. 18 gauge. Settling tank at least 16 inches high to extend
under the entire washer and to be made of No. 16 gauge galvan-
ized iron. The casing and tank shall be braced on the outside
with IJ^" x 1 14" galvanized angles. These angles shall not be
spaced further apart than 3 feet. All joints inside of casing
shall be either soldered or made tight with rubber gaskets and
bolts. All rivets and rivet holes shall be soldered over on inside
of casing.
A perforated galvanized distributing plate on the inlet of
the washer having 50 per cent, free area is to be provided.
Inspection Door. In the side of the washer casing is to be

provided a door not less than 15" x 2V in size to allow easy and
convenient access to the machine for inspection and cleaning.
The door shall be of cast iron with two glass panels, each glass
being not less than 9" x 12^^. The door frame is also to carry a
l^" X l^" pure rubber gasket against which the door is to close.
The frame is to be cast iron and riveted to the washer casing.
637
The door is to be held closed by at least three cams on each side;
and be sufficiently rigid to prevent cracking of glass when clamp-
ing tight against gasket.
Sprays. The brass spray nozzles shall be evenly spaced

over the cross section of the washer and shall be placed at least 4
feet from the eliminator plates and in a plane parallel thereto.
The spacing of these nozzles shall be such that the entire inter-
ior of the washer between the nozzles and eliminator plates shall
be uniformly filled with a finely divided spray. There shall be
at least five nozzles for every 2000 cu. ft. of air handled per
minute. The nozzles shall give a finely divided "mist like"
spray. No water passage or way to be smaller than 27/1000
sq. in. area, nor have less than %e-inch minimum dimensions in
any water passage. Stand-pipes shall be of IJ^-inch galvanized
extra heavy wrought iron pipe screwed into a cast iron header.
The flooding nozzles over the eliminators are to be spaced on 3-
inch centers and handle 1 G. P. M. each.
Eliminators. The eliminators shall set vertical in position

and be made No. 24 gauge galvanized iron. The angles of
of
the eliminators shall not be greater than 35. The eliminators
shall be so set that the air in passing through is deflected at least
six times. Eliminator plates to be bolted or riveted directly to
galvanized iron supports. The angles of deflection in no case
being greater than 35. No separate metal clips will be allowed.
The space of air passage between any two adjacent eliminator
plates shall not exceed 1 inch.
The washer be so arranged that the first four bends of

shall
the eliminator plates shall become a washing surface, A sepa-
rate set of sprays (independent of the main sprays) is to be pro-
vided for maintaining a constant sheet of water flowing down
these four surfaces continually. The amount of washing surface
thus provided shall not be less than 40 sq. ft. per 1000 cu. ft.
of air per minute.
The last two bends of the eliminator plates are to remove

effectively all free and entrained moisture. The total washing
and eliminating surface shall not be less than 60 sq. ft. per 1000
cu. ft. of air per minute.
Piping. 2-inch galvanized overflow and 2-inch drain to

sewer, the latter provided with gate valve.
538
SPECIFICATIONS
The washer is to be provided with galvanized iron flanges
for piping connections.
Note. All piping between the washer, settling tank and pump
shall be galvanized and be installed by contractor in accordance
with details furnished by the air washer manufacturer.
Pump. The air washer is to be furnished with a double
suction centrifugal horizontally divided shell pump having a
capacity of gallons per minute when discharging against
sufficient head to obtain perfect spray effect of all nozzles.
The pump is to be of the horizontal type having enclosed
runner and is to be provided with cast iron base plate for direct
connection to a H. P. motor of suitable current.
The casing of this pump is to be of grey cast iron, horizon-
tally divided for convenient inspection, suitable to withstand
an excess over the working pressure and designed with ample
water ways for proper velocity.
All surfaces not machined shall be rubbed down, filled and
painted a suitable dark color as directed.
The usual piping drains, fittings and grease cups are to be
included.
Accessories. The washer is to be provided with an automatic

float valve for maintaining a constant water level.
One strainer of 20-mesh copper screen is to be provided with

the washer for straining all water recirculated by the pump.
This strainer is to extend the entire width of the washer and
shall have not less than 1 sq. ft. of surface for each 4000 cu. ft.
of air handled per minute.
CARRIER TYPE "B" AIR WASHER AND HUMIDIFIER

Specifications for the Type "B" Air Washer and Humidifier
are the same as for the Type "A" as given on pages 537 to 539
with the following exceptions:
Washer is 9' 0%" long instead of 7' 2Vi(i'' long.
There shall be at least five nozzles for every 1000 cu. ft. of
air handled per minute instead of five nozzles for every 2000
cu. ft. of air handled per minute.
539
CARRIER TYPE "C" AIR WASHER AND HUMIDIFIER
To furnish and erect where shown on plans one (1) air
washer of a design as specified herein. Washer is to have a
capacity of cu. ft. A. P. M.
The velocity of the air through the washer shall not be

greater than 500 ft. per minute and the total guaranteed resist-
ance of the washer shall not exceed .375 inch water. Washer
is to be 4' 10'' long, wide and high.
Casing. The washer shall be constructed of galvanized iron
of No. 18 gauge. Settling tank at least 16 inches high to extend
under the entire washer and to be made of No. 16 gauge galvan-
ized iron. The casing and tank shall be braced on the outside
with \]/2" X \]/2" galvanized angles. These angles shall not be
spaced further apart than 3 feet. All joints inside of casing
shall be either soldered or made tight with rubber gaskets and
bolts. All rivets and rivet holes shall be soldered over on inside
of casing.
Inspection Door. In the side of the washer casing is to be

provided a door not less than 15" x 2V in size to allow easy and
convenient access to the machine for inspection and cleaning.
The door shall be of cast iron with two glass panels, each glass
being not less than 9" x 12". The door frame is also to carry a
3<" X 34" pure rubber gasket against which the door is to close.
The frame is to be cast iron and riveted to the washer casing.

The door is to be held closed by at least three cams on each side,
and be sufficiently rigid to prevent cracking of glass when clamp-
ing tight against gasket.
Sprays. The brass spray nozzles shall be evenly spaced over
the cross section of the washer and to be placed at least 4 feet
from the eliminator plates and in a plane parallel thereto. The
spacing of these nozzles shall be such that the entire interior of
the washer between the nozzles and eliminator plates shall be
uniformly filled with a finely divided spray. There shall be at
least five nozzles for every 2000 cu. ft. of air handled per min-
ute. The nozzles shall give a finely divided "mist like" spray.
No water passage or way to be smaller than 27/1000 sq. in.
area, nor have less than îe-inch minimum dimensions in any
water passage. Stand pipes shall be of 1 34-inch galvanized
extra heavy wrought iron pipe screwed into a cast iron header.
540
SPECIFICATIONS
Eliminators. are to be made of corrugated
The eliminators
sheets of No. 24 galvanized iron. These eliminators are to be set
vertically in rows so that the air has a tortuous passage through
them and they are to be evenly spaced not further apart than
3^*. They shall be braced and stiffened with galvanized angles.
At least two of the corrugations of the sheet are to be pro-
vided with a lip for catching any entrained moisture which
otherwise may pass through eliminator.
Piping. The washer is to be provided with galvanized iron

flanges for piping connections.
One (1) overflow connection and drain is also to be provided.
Note. piping between the washer, settling tank and

All
pump be galvanized and be installed by contractor in
shall
accordance with details furnished by the air washer manu-
facturer.
Pump. The air washer is to be furnished with a

double suction centrifugal horizontally divided shell pump hav-
ing a capacity of gallons per minute when discharging
against sufficient head to obtain perfect spray effect of all
nozzles.
The pump is to be of the horizontal type having enclosed
runner and is to be provided with cast iron base plate for direct
connection to a H. P. motor of suitable current.
The casing pump is to be of grey cast iron, horizon-
of this
tally divided for convenient inspection, suitable to withstand
an excess over the working pressure and designed with ample
water ways for proper velocity.
All surfaces not machined shall be rubbed down, filled and
painted a suitable dark color as directed.
The usual piping drains, fittings and grease cups are to be
included.
Accessories. The washer is to be provided with an automatic

float valve for maintaining a constant water level.
One strainer of 20-mesh copper screen is to be provided with
the washer for straining all water recirculated by the pump.
This strainer is to extend the entire width of the washer and
shall have not less than 1 sq. ft. of surface for each 4000 cu. ft.
of air handled per minute.

541
FAN ENGINEERIN G B UFFALO FORGE COMPANY
HUMIDITY CONTROL
The washer is to be provided with a system of humidity con-
trol arranged for maintaining a constant dew-point or saturated

temperature of the air leaving the washer throughout the winter.
This constant dew-point being maintained by varying the water
temperature through which the air passes.
A thermostat is to be placed in the chamber between the
washer and reheater coil. This thermostat shall be of a gradu-
ated action type to the approval of the engineer. In the suction
line to the washer pump is to be provided an automatic com-
bined ejector water heater and diaphragm valve to which a
steam line is to be connected. In this steam line shall be placed
a pressure reducing valve of the Mason Regulator Company's
manufacture or equal by the steam contractor designed to main-
tain a pressure of 5 pounds per square inch on the ejector if high
pressure steam is used. The operation of the automatic ejector
water heater shall be gradual. This water heater is to be oper-
ated by a diaphragm operated by air pressure and a constant
temperature thermostat placed between the eliminators and the
reheating coil. A globe valve is to be provided in the steam line
within a few feet of the water heater by the heating contractor.
Provide a reverse acting diaphragm steam valve which is to
be placed in the steam line to the water heater and operated
through a safety relay. The valve and relay are to be so con-
nected that should either the water pressure or the air pressure
fail, the steam supply to the ejector water heater will be
automatically shut off.

One (1) 3^-inch opening will be left in the air supply line of
the thermostatic control system for supplying compressed air
to the thermostat mentioned above. The humidity control
shall be set and left in proper adjustment by the manufacturer.
One (1) Pot Strainer. For insertion in the line running
from the centrifugal pump to spray header is to be provided a
cast iron galvanized pot strainer with two baskets, one to be
used as a spare. The baskets of this strainer are to be made of
14-mesh copper wire cloth screen, giving an area of at least six-
teen times area of pipe connection. The top of this strainer is
to be held by clamp and screw so it can be instantly and easily
removed and replaced with clean strainer.
A steam line is to be run to reverse acting diaphragm
valve.
542

GUARANTEES
GUARANTEES
Apparatus
The apparatus is guaranteed to be first class with reference
to both workmanship and design. Parts which may prove
defective within one year after shipment are to be replaced with-
out charge.
Fan
The fan shall be capable of handling cu. ft. of air per min-
ute at degrees F. and 29.92 inches barometer, at approxi-
mately inches static (or total) pressure, when running at a
speed not to exceed R. P. M.
The fan will require under above conditions not over H. P.
Heater
The free area through the coils shall be of such size that the
velocity of the air in passing through them shall not exceed
feet perminute and of ample capacity to heat cu. ft. of air
per minute from to
degrees F. using steam at pounds
pressure per square inch gauge.
Engine
The engine shall be belted (or direct connected) to fan,
and shall be capable of running at R. P. M. with steam pres-
sure of pounds per square inch gauge, when fan is deliver-
ing cu. ft. A, P. M. against in. static (or total)
pressure.
Heating
The apparatus is to be of sufficient capacity to heat the build-
ing to degrees F. when outside temperature degrees
is
F., using per cent, return air and per cent, fresh air,
and to give a minute air change.

Ventilating
The apparatus is to be of sufficient capacity to supply
cu. ft. minute; (or in case of school,

air per cu. ft. air per
minute per pupil,) at degrees F. and 29.92 inches barometer.
Ducts and Outlets
The velocity shall not exceed: in the mains feet per
minute; in the risers feet per minute.
The air shall enter the room at not less than feet
from the floor, at a velocity not to exceed feet per minute.
The vent shall be of the size of the inlet and shall be taken
off at the floor line where practicable.
643
Air Washing
The washer is to remove 98 per cent, of the solid material
carried by the entering air.
Humidity Control
The apparatus is to automatically control the dew-point
temperature within one degree of that desired, when the out-
side wet-bulb temperature is less than the dew-point for which
the control is set.
Cooling With Air Washer

The apparatus when recirculating water in summer and han-
dling rated capacity of air, is to reduce the temperature of the air
70 per cent, of the entering wet-bulb depression (i. e., 70 per
cent, of the difference between the dry and wet-bulb tempera-
tures) of the incoming air.
The apparatus when using cold water in summer is to reduce
the outgoing difference of air and water temperatures to less
than 25 per cent, of the difference in the incoming temperatures.
Cooling With Humidifier
Cool to the wet-bulb temperature of the entering air.
Mechanical Draft
The fan shall be used for (induced or forced) draft for boilers
burning pounds coal per hour, having a heat value of
B. t. u. The fan shall be of sufficient capacity to handle
cubic feet air per minute at degrees F. and 29.92 inches
barometer and maintain a pressure of inches of
water at the grate, when revolving at a speed not to exceed
R. P. M.
The fan will require under above conditions not over
H. P.
544
."2 M
= M CO M CO
<M (N (N IN Cl ^M iM
-t-* U5 lO
33 xxxxxxxxxxxxxxxxx i< <}<
(N W WM <M C^l iM M CC eC CO ^ * ^ * Tf It
xxxxxxxxxxxxxxxxxxxxx
s^.
lN(NCJ(N<MiNiMMCOCOeOCOCOTj<Tl<TjH0000C
XXXXXXXXXXXXXXXXXXXXX
MiM(N(N(N(N(NlMiMCvi<N(M<NCOeCCOCO-*rf<Tf<-ttt
e S93ue|j <C00<r><D<OCO<OCOCOTt<T<Tt<'*<Tf<^Tt<(MC^]C^jiM
i_
*- > :-2
sapeig <*<'J<'*Ti<Tj<T}<rJ<-<tTj<Tj4iM(N<N(NiM(NiMOOOO
3(2 5-
I
S33qSI|OJ3S *'^^^t1<tJ<t1t)4M(M(NiNC>1NOOOOOOO
SIMMS 3P!S C^C^<NM(N(N<NW(N(NiM(N(MN<N(NOOOOO
ssSueij 00Q00000000000000000cDOCOO<riCDi'^Tj<TttTjt
sspeig COCOCOCO><OcOC050COTt<TjiTt<TjiTt<Tt<Tt<(N(M<N(N
sjaaqs IIOJOS DcOcD;OCO<OCCTt<rJ4Tj<-^-<fTj<(NiM(N(NiN(M(M
S}33I|S 3P!S *'*Tj4-^TjtTj<Tj<-^Tj(li('^ltTj<fi<TfT}<(NlN(MIN(M
I
IJBJQ
I psDnpuf (N IN (M CO CO CO
SunqjaAQ < !N <N IN IN CO CO
8unqj3A0
J3MO|g 0>>-HCCiOCCXCCOtÔ-^t^'lOXr-HiOOJiMiOO
i-i^ .-I .-H -H 01 01 01 00 CC CO ^ Tf 't lO lO C CO ?o
:;::?: ;^::?^.; \t:jt ::?t::?r :^:^:sj
J3)snBqxa| !
SjV.
J3|3IUefQ 05e'jcx<Nioc)0c^r^Tt<or>.coococooso<Noo
i-HC)W(NC0C0C0'^lCOCOt^t-000SO5OO-HMM
::i^:sx';i^:^::j*::^ ;:?^;?r::^::?^^^ r-x;?t^::S!:5j:st

J9MO|g
j9)snei|X3 coorâ5co'Coo>oooooooocoiooioo
_^^^0îMOlC0'*<<JtiOCCO50t^t^00000i01O
r-l
OiM'tr^05'-<i000iNO05C0CCOC0tÔ-<ft^.-l
' ^ÔlOIOlCOCOeO'l'-îOiOiOCDCOCOt*
xxxxxxxxxxxxxxxxxxxxx
OiN't'Ct^CS"OGOiMiOCiCCCCOCOtÔ'*t>.-l
r-i.-ir-H.-i^,-iOIOI01COCOCO'^-f>OiOiOCOCOCOt
uvjazis >eoo<
645
X y, y, '/i X y. X y, X y, X ',< X i<>i X X

CI (M OJ Ol iM M C> CI CO CO * * * T< 0 0
X X X X X X X X X X X X X X X X X
XXX X X XX X X X X X X X X X X X X X X
(NC^DICKMCICKNCOeOCCCOW^-t-îOiOiOiOiO
X X XX XX XX XXX X XX XX X XXXX
C4<N(N(MCIC^C<I(NC^C1C^(N(NCOCOCOCO-<1<'*'*-*
93UVIJ (N<N(Ne^(N(N(N(NOOO000000000000C00000W
Msia OOWOOOOGOOOOOOOOOOOOOCOCOCOCOCOCOCOOOO
u.
C a
sapBig "Tl<(M(M<MOOOO00000000
sjaaqs UOJ^S *'*<''l->*-*-l<TfN(N(N(M(Ne<(N<NOOOOOO
Suisnofi J3| C<N(NiNiNOOOOOOOOOOOOOOOQ0

13 -jenb-33Jiix
B mc^cî<n<noooooooooooooqooooooo
SuisnoLiiinj
aSuBij * * * rf Tfl< * <* N C (N O O O O 00 00 00 00 00 00
isia oooooooooooooooooooeoeoeoooo
S !
4
sapuia I
00000000C0i'^'^'*(N(N(N(NOOOO0000
li
SldailQ IIOJ3S !
'*OOcOcDOt'*Ttt-^-*Tf-f<CICJ(NW(M(NM
8UISnO(| J3| *<< Tf Tt .J4 (N (N (N(N(M(N CI (N(N(N CI <N<N(NC0
8u|sno||||nj i-*N(N(N(NCMWMClCIOOOOOO
UBJMIS ^ iS ^
647

CHIMNEYS
The following rules for the design of chimneys are given in
"Steam," published by the Babcock & Wilcox Company.
(a) To find the draft in inches of water produced by a given
chimney. Divide 7.6 by the absolute temperature of the exter-
nal air (ta + 460); divide 7.9 by the absolute temperature of the
gases in the chimney (tg + 460); subtract the latter from the
former, and multiply the remainder by the height of the chim-
ney in feet. This may be expressed as
7.6 7.9
,
d =h /
,
\ta + 460 tg+ 460,
(b) To find the height of a chimney to give a specified draft

expressed in inches of water: Proceed as above through the
first two steps, then divide the required draft by the remainder,
and the result will be the height of the chimney in feet. Ex-
pressed as a formula,
\ta+460/ \tg+460/
The draft attainable with any chimney when the tempera-
ture of the external air is 70 F, and the temperature of the
flue gases 550* F, multiply the height above the grate in feet
by 0.0065 and the product is the draft pressure in inches of
water.
548
CHIMNEYS
Square
Actual Area Feet
FAN ENGINEERING BUFFALO FORGE CaMPANY
TABLE OF AREA AND CIRCUMFERENCE OF CIRCLES
AREA AND CIRCUMFERENCE OF CIRCLES
TABLE OF AREA AND CIRCUMFERENCE OF CIRCLES
o
CO
7
PRESSURE CONVERSION TABLE
PRESSURE IN INCHES OF MERCURY EXPRESSED IN EQUIVALENT

POUNDS PER SQUARE INCH
In.
SIZE OF STEAM PIPES
100=Foot Length
u
3
SIZE OF S T P: AM PIPES
SIZE OF STEAM PIPES
100=Foot Length
20 Lbs. 40 Lbs.
i
SIZE OF STEAM PIPES
100=Foot Length
s
IZE OF STEAM PIPES
PROPERTIES OF SATURATED STEAM
Temp.
DRY KILN
o
IS
t/3r
1

LOGARITHMS
Proportional Parts
1 ! 2
I 1234 5 67 8 9
10 0000 0043 0086 0128 0170 0212 0253 0294 0334 0374 4 8 12 17 21 25 29 33 37
11 0414 0453 0492 05310569 0607 0645 0682 0719 0755 4 8 11 15 19 23 26 30 34
12 0792 0828 0864 0899 0934 0969 1004 1038 1072 110613 7 10 14 17 21124 28 31
13 1139 1173 1206 12391271 1303 1335.1367 13991430 3 6 10 13 16 19 23 26 29

14 1461 1492 1523 1553 1584 1614 1644 1673 1703 1732 3 6 9 12 15 18 21 24 27
15 1761 1790 1818 1847 1875 190311931 1959 1987 2014 3 6 8 11 14 17 20 22 25
16 ;2041 2068 2095 2122 2148 2175 2201 2227 2253 2279 3 5 11 13 16 18 21 24
17 2304 2330 2355 2380 2405 2430 2455 2480 2504 2529 2 5 1 10 12 15 17 20 22
18 2553 2577 2601 2625 2648 2672 2695 2718 2742 2765 2 5 1 9 12 14 16 19 21
19 2788 2810 2833 2856 2878 2900 2923 2945 2967 12989! 2 4 9 11 13 16 18 20
20 3010 3032 3054 3075 3096 3118 3139 3160 3181 320l!2 4 8 11 1315 17 19
21 3222 3243 3263 3284 3304 3324 '3345 3365 3385 3404 2 4 8 10 1214 16 18
!
22 3424 3444 3464 3483 3502 3522 3541 3560 3579 3598 2 8 10 12 14 15 17
23 3617 3636 3655 3674 3692 3711 3729 3747 3766 3784 2 7 9 11 13 15 17
24 3802 3820 3838,3856 3874 3892 3909 3927 3945 396212 7 9 11 12 14 16
i
25 3979 3997 4014 4031 4048 4065 4082 4099 4116 41332 9 10 12 14 15
26 4150 4166 4183 4200 4216 4232 4249 4265 4281 4298 2 8 iO 11 13 15
27 4314 4330 4346 4362 4378 4393! 4409 4425 4440 445612 8 9 11 13 14
28 4472 44871 4502,4518 4533 454814564 4579 4594 4609 2 j
8 9 11 12 14
29 4624 4639 4654 4669 4683 4698 4713 4728 4742 47571 7 9 10 12 13
30 ,4771, 4786 4800 '48 14 4829 4843 4857 4871 4886 4900 1
i 7 9 10 11 13
31 14914 4928 4942 4955 4969 4983 4997 5011 5024 5038 1 3 7 8 10 11 12
32 5051 5065 5079 5092 5105 5119 5132 5145 5159 5172 1 3 7 8 9 11 12
33 j5185 5198 5211 5224 5237 5250 5263 5276 5289 5302 1 3 6 8 9 10 12
34 5315 5328'5340;5353 5366 5378 5391 5403 5416 54281 6 8 9 10 11
35 5441 5453 5465 5478 5490 5502 5514 5527 5539 5551 1 6 7 9 10 11
36 5563 5575 5587 5599 5611^ 5623 5635 5647 5658 5670 1 6 7 8 10 11
37 5682 5694 570515717 5729' 5740 5752 5763 5775 57861 6 7 8 9 10

38 5798 5809 5821 5832 5843 5855 5866 5877 5888 5899 1 6 7 8 9 10
39 5911 5922 5933 5944 5955 5966 5977 5988 5999 60101
1
5 7 8 9 10
I
1
40 6021 6031 6042 6053 6064' 6075 6085 6096-6107 !61171 5 6 8 9 10

41 6128 6138 6149 6160 6170 6180 6191 6201 6212 6222 1 5 6 7 8 9
42 6232 6243 6253 6263 6274, 6284 6294 6304 6314 6325 1 5 6 7 8 9
I
43 6335 6345 6355 6365 6375' 6385 6395 6405 6415 6425 1 5 6 7 8 9
44 6435 6444 645416464 6474 6484 6493 6503 6513 6522 1 5 6 7 8 9
45 6532 6542 6551 6561 6571 6580 6590 6599 6609 6618 1 5 6 7 8 9
I
46 6628 6637 6646 6656 66651 6675 6684 6693 6702 167121 I
5 6 7 7 8
47 ,6721 6730 6739 6749 6758 6767 6776 6785 6794 6803 1 6 5 6 7 8
48 !6812 6821 6830 6839 6848 6857 6866 6875 6884 ,68931 4 5 6 7 8
'
49 6902 6911 6920 6928 69371 6946 6955 6964 6972'69811 4 5 6 7 8

50 6990 6998 7007 7016 7024 7033 7042 ,7050 7059 17067 4 5 6 7 8
51 70767084 7093 7101 7110 7118 7126 17135 7143 ,7152.1 4 5 6 7 8
52 7160 7168 7177 7185 7193 7202 7210 7218 7226 7235 1 4 5 6 7 7
53 ,7243 7251 7259! 7267 7275 7284 7292 7300 7308 73161 4 51 6 6 7
54 7324! 7332 734017348 7356, 736417372
j
739611 4 5 6 6 7
560
USEFUL FACTORS
1 gal. (U. S.) = 231 cu. in.
= 0.13368 cu. ft.
1 gal. (British) = 277.274 cu. in.
1 cu. ft. = 7.480.5 gals.
1 cu. water at G0
ft. F. = 62.37 lbs.
1 gal. water at 60 F. = 8.34 lbs.
1 cu. ft. water at 212 F. = 59.76 lbs.
1 gal. water at 212 F. = 7.99 lbs.
1 barrel water at 00 F- = 31K gals. =262.7 lbs.
1 in. mercury = ]}4 ft. or 13.6 in. water
= 0.491 lb. per sq. in.
1 lb. per sq. in. press. = 2.304 ft. water at 60
Height of a column of water
in feet X 0.434 = lb. press, per sq. in.
A column of water 1 sq. in.
and 2i/<5ft. high = approximately lb. 1
1 calorie = 3.97 B. t. u.
1 kilogram = 2.2046 lbs.
Calories per kilo X l.S = B, t. u. per lb.
1 kilowatt (1000 watts) = 1.340.5 H. P.
1 horsepower = 0.746 K. W.
1 kilowatt = 56.9 B. t. u, per min.
1 mech. horsepower = 42.4 B. t. u. per min.
= 2545 B. t. u. per hour
='33000 ft. lbs. per min.
1 boiler horsepower = 33479 B. t. u. per hour
1 B. t. u. = 778 ft. lbs.
1 ft. lb. per sec. = 1.356 watts
562
USEFUL FACTORS
WATER CONVERSION FACTORS*
r. S. gallons X8.33
V. S. gallons
U. S. gallons
r. S. gallons
Cu. in. water at 39.1*
Cu. in. water at 39.1*
Cu. in. water at 39.1'
Cu. ft. water at 39.1'
Cu. ft. water at 39.1'
Cu. water at 39.1'
ft.
Pounds of water
Pounds of water
Pounds of water
REPORT OF COMMITTEE OF AMERICAN SOCIETY OF HEAT-
ING AND VENTILATING ENGINEERS ON STAND-
ARDS FOR VENTILATION LEGISLATION FOR
MOTION PICTURE SHOW PLACES
January, 1913
Ventilation and sanitation requirements cannot be too

strongly emphasized when dealing with the question of legisla-
tion relating to motion picture show places. The widespread
neglect, in a very large number of communities throughout the
country, of proper ventilation and sanitation in such motion
picture show places, has many times been correctly characterized
as a "menace to public health"
materially affecting the moral
tone as well.
Allowing the great importance of fire protection and struc-
tural requirements for the protection of life ; elimination of low-
class vaudeville, enforced lighting during performances, super-
vision of pictures exhibited, and other essential matters for the
protection of morals; ventilation and sanitation requirements
loom up large for the protection of health.
The Committee has been appointed to deal with the subject

of ventilation and this question is, of course, vitally concerned
with all the conditions of the air breathed, particularly temper-
ature, air purity, air motion, humidity, and freedom from dust
(impurities from breathing, skin exhalations, dust, etc., being
constantly released in large quantities in every audience hall).
With a view of suggesting minimum requirements that are
practical to secure, the followingrecommendations are made as
standards for legislation to cover this important phase of the
needed general regulations for motion picture show places.
MINIMUM VENTILATION STANDARDS

1. Floor Area per Occupant.
A minimum of 4Vs sq. ft. of floor area, as a seating space, per
occupant, exclusive of aisles and public passageways, shall l)e
provided in the audience hall.
2. Cubic Space per Occupant.

A minimum of 80 cu. ft. of air space, per occupant, shall be
provided in the audience hall.
564
PROPOSED MOTION PICTURE THEATER LEGISLATION
3. Quantity of Outdoor Air.
A positive supply of outdoor air from an uncontaminated
source shall be provided the audience hall at all times while the
show place is open to the public, and -the quantity of this posi-
tive supply of outdoor air shall be based on a minimum require-
ment of 15 cu. ft. per minute, per occupant.*
4. Temperature.
The temperature of the air in the audience hall shall at all
times, while the show place is open to the public, be maintained

throughout at the breathing line (persons being seated) within
the range of 62 F. to 70 F. (except w^hen the outside tempera-
ture is sufficiently high not to require the air supply for ventila-
tion to be heated). The temperature, distribution and diffusion
of the supplied outdoor air shall be such as to maintain the tem-
perature requirement without uncomfortable drafts.
5. Direct Heat Sources.
Any good heat source which does not contaminate the air
will be accepted to supplement the warmed outdoor air supply.
Gas radiators are prohibited.
6. Machine Booth Ventilation.

Enclosures or booths for the motion picture machines shall
be provided with special exhaust ventilation with a capacity to
exhaust at all times not less than 60 cu. ft. of air per minute
through a one-machine booth, not less than 90 cu. ft. of air per
minute through a two-machine booth, and not less than 120 cu.
ft. of air per minute through a three-machine booth.
This requirement shall include a number of small metal

screened openings (equipped with special dampers and automatic
appliance with fusible link to automatically close tight in case
The ordinance in force in the City of Chicago at the present time requires
that the air in the auditorium in the class of buihlings in which motion picture
show places are included, shall be changed so as to supply for each person for
whom seating accommodation is provided, at least 1500 cu. ft. of air per hour
for new buildings, and at least 1200 cu. ft. of air per hour for buildings con-
structed prior to the passage of the ordinance, which requirements the Illinois
Chapter of the Society considers practical to obtain and desirable to require by
legislation formotion picture show places.
Higher standards of ventilation than set forth as minimum in the committee's
report are urged wherever possible to obtain.
665
of fire in thebooth) on the sides of the booth near the bottom,
aggregating 180 sq. in. for a one-machine booth, 210 sq. in. for
a two-machine booth, and 240 sq. in. for a three-machine booth;
and this requirement shall also include a metal or other fireproof
flue, extending from the top or side at the top of the booth, and
carried to a proper place of discharge outdoors. The ventila-
tion should be augmented by mechanical or other means, so as
to exhaust at least the quantity of air as herein stated.
^ The size of this special fireproof vent flue shall be not less
than 96 sq. in. clear area for a one-machine booth, not less than
120 sq. in. clear area for a two-machine booth, and not less than
144 sq. in. clear area for a three-machine booth, and this special
vent flue shall be provided with an adjustable damper, operated
from the booth, and equipped with an automatic appliance and
a fusible link to operate so as to open the damper wide auto-
matically in case of fire in the booth. The machine booth venti-
lation shall be kept in operation at all times when the booth is in
use.
be noted that the foregoing regulations are simple,

It will
and that violations may be readily detected, also that care has
been exercised to leave large latitude for design of the ventilating
apparatus.
It should be especially noted that the foregoing regulations
call for a minimum of all requirements as compulsory, and that
it should be the aim of the administrative department having
enforcement of the regulations in charge to encourage motion

picture show owners and managers to provide as comprehensive,
liberal, and high-class equipment as possible, wuth a view to
catering to the comfort and health of the patrons and thus add
to the popularity of the show place as compared w4th others
which may have barely come within the legal requirements.
Elimination of dust from the air supply by means of air fil-

washers is desirable under the best conditions and is
ters or air
imperative under some conditions of especially dusty air supply.
This question is dealt with by suggestion in the following
general clauses.
566
PROPOSED MOTION PICTURE THEATER LEGISLATION
The controlling of relative humidity
is desirable, whenever
possible,but the Committee decided to omit from the regula-

tions any humidity requirement.
The machine booth ventilation, as per recommended regula-
*****
tions, would be greatly improved, especially for summer condi-
tions, by providing a duct connection from out of doors to the
bottom of the booth, for the introduction of outdoor air directly
to the booth.
*****
in
Strong emphasis is placed on the need of having the adminis-
trative feature of legislation of the kind here advocated, placed
the control of a responsible department, *
and that such de-
partment be supplied with a special inspector or inspectors, ex-
perienced in heating, ventilation, and sanitation, and that such
*
department be given reasonable latitude by legislation, such as

to require approval of plans preceding installation or to require
special extra equipment for special cases, such as dust filters for
air supply where the air supply is especially dust laden exhaust ;
ventilation of toilets where building laws do not properly cover

this matter; fans in the auditorium, to keep the air in motion
where diffusion is insufficient, etc., it being made clear in the
legislation that such latitude should in no case include the right
to reduce the stated minimum requirements. The administra-
tive department should also be given the support of other local
or state departments, as the case may be, such as the fire de-
partment, police department, health department, etc.
* * * * * . *
5ti7
INDEX
PAGE
Absolute humidity 28
Air, amount of jjer person in various buildings 59
amount of required for combustion 85
composition of '
11
effectof barom. pressure on velocity of 24
effect of humidity on velocity of 27
effect of temperature on properties of 27
effect of temperature on velocity of 24
effect of temperature on volume of 26
heat required to condition 73
measurement of amount used for combustion 89
measuring quantity of 525
moisture removing capacity-of in fan system dryers 79
properties of dry T 13
properties of saturated 15
purity of 58, 59
quantity of and final temp, required for heating 56
relation of altitude to properties of 24
relation of per cent, excess to boiler eff. and draft requirement . . 87
relative volurnes of at various temperatures 13
requirements due to gas and electric lights 62
requirements due to machinery 62
requirements due to heat given off by occupants ()2
requirements due to sun effect 62
specific heat of 16
velocity of at different pressures 21
velocity of at different pressures and temperatures 22, 23
velocity of through heaters 412-414
weight of 11
weight of dry, at various temperatures 13
weight of saturated and partly saturated 17
Air conditioning apparatus 469
Air connections for Buffalo heaters 452-455
Air density, pressure,and velocity relations in fan work 178
Air ducts Ill
diverging nozzle in 122
friction in 115
pressure losses in 114
Air measurements, coefficients of discharge for 203
Air washers 469
Air washers, dimensions of
Carrier type "A" washer 477
Carrier type "B" washer 479
Carrier type "C" washer 481
Air washers, principles of construction of 469
Air washers, specifications of
Carrier type "A" washer 537
Carrier type "B" washer 539
Carrier type "C" washer 540
Air washing 67
guarantees - 544
Alternating current motor specifications 536
Altitude, relation of to properties of air 24
Anemometer, method of use 197
Apparatus 168
guarantees 543
selection of for heating and ventilating 506-518
568
INDEX
PAGE
Application of heater tables and curves 415-417
Area and circumference of circles 550, 551
Arrangement of fan system apparatus 63
Aspiration psychrometer 29
Automatic humidity control 74
Automatic pumps and receivers
centrifugal 407
duplex 4(jri
Average indoor temp, for various outdoor temp 404
"B" volume blowers and exhausters 334

capacity table 335
Baby Conoidal fans 350
Barometric pressure, effect of on velocity of air 24
Black steel piping, weight of 102
Blast area of Buffalo heaters 458
of piping 120
Blast and exhaust ducts for forge shop equipment 108
Blow-through fan system apparatus 03
Blow pipe work (see exhaust systems) 93-105
Blowers, "B" volume 334
capacity table 335
Blowers, electric 355
Blowers, Planoidal steel plate
capacity tables 217-223
dimension tables 359-362
performance curves 224, 225
Blowers, steel plate pressure 340
Blowers, steel pressure 330
capacity table 337
dimension table 398
performance curve 339
Bodily heat, removal of 59
Boiler eff. and draft required, relation of to per cent, excess air 87
Boiler plant, measurement of air used in 89
Brake horsepower of engines (see steam engines) 488-505
Branch pipes, diameters of 144-152
Buffalo pipe coil heaters (also see heaters) 399
blast area of 458
by-pass proportions for 457
curves 432-438
dimensions of heater case for 450
friction of air through 440
length of connection 457
regular open area pattern 399
return bend pattern 399
sizes and dimensions of 449
steam, drip and air connections of 452-455
tables 418-431
velocity of air through 412-414
Buffing wheels, dust removal from 96
Buildings, heat losses from 46
By-pass proportions for heaters 457
Calculating fan tests, method of 529

Capacity of fans (see fan cai)acity tables) 205-356
Capacity of pipes 143, 154, 155
Carbon dioxide, amount allowable in air 59
determination of 60
Carrier air washers
dimensions of 470-481
specifications of .537-541
Carrier humidifying system
cooling capacity of 69
temperatures in ..... 71
Carrying capacity of pipes 143, 154, 166
569
PACK
Cases, flimensions of Buffalo heater 456
Cast iron heaters, Vento (also see heaters) 401
curves 432-438
friction through 447
sizes and dimensions of 460-462
tables 439-444
Centrifugal automatic pumps and receivers 467
Chimneys
draft produced by 548
height 548
size of with appropriate horsepower of boilers 549
Circles, area and circumference of 550, 551
Circular equivalents of rectangular ducts 156, 157
Circumference and area of circles 550, 551
Coefficients of discharge for air measurements 203, 204
Coils, condensation in heater 411
Cold storage insulation, heat loss through 53
Collectors, dust 101, 102
Combination fan, engine and heater tables 206, 518-523
Combustion, air required for 85
measurement of air used for 89
Composition of air 11
Condensation in heater coils 411
Contlitioning air, apparatus for 469
heat required for 73
Cone wheels 332
capacity table 333
Connection, length of heater 457
Connections of Buffalo heaters, steam, drip and air 452-455
Conoidal fans (see Niagara Conoidal an<l Turbo-Conoidal fans)
Converging nozzle
use of in fan testing 201
Conversion table, temperature 552
water 563
Cooling 68
Cooling capacity of Carrier humidifying system 69
Cooling effect, relation of to per cent, relative humidity 69
Cooling guarantees 544
Corresijonding pressures and velocities of dry air 21
Corresponding velocity of dry air at various press, and temp 22, 23
Cubic feet air per person for ventilating 59
Cupola service 110
Dehumidifiers 471
spray type 474
Dehumidifying, refrigeration required for 474
1)esign of fans 1 68
of humidifiers 472
Determination of temperature guarantees 463
Dew-Point 28
Dew-Point method of humidity control 72
Dew-Point temp., relation of wet -bulb and dry-bulb temp, to 30
Diameter of branch jnpes 144-152
Dimensions of Buffalo heaters 448-451
of Buffalo heater cases 456
of Carrier air washers 476-481
of engines (see steam engines) 498-505
of fans (see fan dimen.sion tables) 358-398
of Vento heaters 460-462
Direct current motor specifications 536
Directions for fan installation tests 525
Disk fans 348
capacity table 349
Disk fans, multiblade 350
570
INDEX
DiverginK nozzles i-acje
efficiency of 123
in air ducts 122
])erformance of 125
Downward systems of ventilation 64
Draft, air re(|uired for S5
for furnaces S3
losses 84
Draft, mechanical 81
forced 81
induced 82
induced Planoidal fan tables 322-325
induced Niagara Conoidal fan tables 326, 327
induced special steel plate fan and engine tables 328, 330
Draft produced by chimneys 548
requirement 82
requirement and boiler efif., relation of to per cent, excess air ... 87
Draw-through fan system apparatus 63
Drip connections of Buffalo heaters 452-455
Dry-bulb temp., relation ot wet-bulb and dew-point temp. to. 30
Dry-bulb thermometer 28
Dry kilns 559
Dryers, Buffalo progressive lumber dry kiln 559
Dryers, moisture removing capacity of air in 79
Drying, methods of 75
conditions for different materials 77
Duct velocity for public buildings 134
Duct work, guarantees 543
specifications 535
Duplex pumps and receivers 466
Dust collectors 101 102
.
Dust removal from grinding and buffing wheels 96

New York State law concerning 96
proportioning piping 98
recommendations 99
Econom\', friction lo.ss through heater for maximum 415

proportioning piping for maximum 130
Effect of barometric pressure on velocity of air 24
of humidity on velocity of air 27
of temperature on properties of air 27
of temperature on velocity of air 24
of temjjerature on volume of air 26
Efficiency, calculation of fan 530, 531
elements of fan design affecting. 170-172

of boiler and draft requirement, relation of to i>er cent, excess air 87

of diverging nozzles 123
Efficiency tables of Niagara Conoidal fans 274, 275
Elbows, friction in 1 17-120
Electric blowers 355

Electric motors, .specifications for 536
Emery wheels, dust removal from ^ 96
Engines, fans, and heaters, combination tables of 206, 518-52li
Engines (see .steam engines) 482
brake horsepower of 488-497
dimensions and maximum horsepower of 498-505
guarantees for 543
mean effect ive pressure of 485
specifications for 535
water rates of 486, 487
Equalizing friction for unequal lengths of pi)>jng 139
Equivalent pressures in inches and pounds . 553
571
PAGE
Exhaust systems 93
dust collectors 101, 102
friction loss 1 02
hood construction 93
pipe sizes, table 94
proportioning piping in 129
size of main pipe 95
size of piping 93
velocity required 95
Exhaust-through fan system apparatus 63
Exhausters, "B" volume 334
capacity table 335
Exhausters, Planoidal steel plate
capacity tables 207-213
dimension tables 359-373
performance curves 214, 215
Exhausters, planing-mill, slow speed
capacity tables 345, 346
dimension tables 396, 397
Exhausters, planing-mill, standard
Factors, useful 562

water conversion 563
Fans, design of 168
guarantees for 543
horsepower of 175
measuring pressure produced by 526
miscellaneous applications of 106
'
performance laws of 179

t pressure characteristics of 172
relation of total, static and velocity pressure in 176
selection of 181
special types and features of 174
Fan capacity tables 205
"B" volume blowers 335
"B" volume exhausters 335
Baby Conoidal 356
cone wheels 333
disk fans 349
electric blowers 355
muJtiblade disk fans 350
Niagara Conoidal, induced draft 326, 327
Niagara Conoidal, static pressure 232-275
Niagara Conoidal, total pressure 228-231
y)laning-mill exhausters, slow speed -. .345,346
planing-mill exhausters, standard 342,343
Planoidal blowers 217-223
Planoidal exhausters 207-213
Planoidal induced draft 322-325
propeller wheels 352, 353
special steel plate induced draft with engines 328-330
steel plate pressure blowers 340
steel pressure blowers 337, 338
Turbo-Conoidal 278-319
Fan dimension tables . 358
Niagara Conoidal, double width 382-385
Niagara Conoidal, full housing 375-378
Niagara Conoidal, overhung whee! 386-393
Niagara Conoidal, three-quarter housing 379-381
planing-mill exhausters, slow speed, double 397
planing-mill exhausters, slow speed, single 396
planing-mill exhausters, standard, double 395
^
planing-mill exhausters, standard, single 394
572
N DEX
Fan dimension tables (continued) paoe
Planoidal blowers, full housing 359-362
Planoidal exhausters, full housing 359-362
Planoidal exhausters, overhung wheel 366-373
Planoidal exhausters, three-quarter housing 363-365
steel pressure blowers 398
Turbo-Conoidal, double width 382-385
Turbo-Conoidal, full housing 375-378
Turbo-Conoidal, overhung wheel 386-393
Turbo-Conoidal, three-quarter housing 379-381
Fan efficiency, elements of design affecting 170, 171
method of calculating 530
Fan, engine and heater tables, combination of 206, 518-523
Fan engines (see steam engines) 482-505
Fan engines, specifications for 535
Fan installation tests, directions for 525
F'an sizes for forge shop equipment 107
Fan system dryers 77
moisture removing capacity of air in 79
Fan system heaters, specifications for 534
Fan systems for heating and ventilating . . , 62
blow-through arrangement 63
draw-through arrangement 63
upward and tlownward ventilation 64
Fan test, directions for making 525
method of calculating 529
Fan testing 190
anemometer 197
converging nozzle 201
installation 525
orifice 198
orifice at end of pipe 200
pitot tube 190
pitot tube, pipe traverse for 196
pitot tube, use of in an air duct 195
short length of pipe 201
Fan work, relation between pressure, velocity and air density in.. . . 178
P'inal temjierature and air required for heating 56
Forced draft 81
F'orge fires, application of steel pressure blowers to 338
Forge shop equipment 107
blast and exhaust duct sizes 108
fan sizes 1^7
Foundry blower practice 109
PViction, allowing for in public building ducts 134
equalizing for unequal lengths of pipes 139
in air ducts H^
in elbows 117-120
loss in exhaust systems 102
proportioning piping to allow for 138
through Buffalo pipe coil heaters 446
through Vento cast iron heaters 447
Furnace draft 83
Galvanized iron pipes, heat loss from 54

weight of 160, 161 ,
163-167
Gas heaters 468
Gauges of iron used for pij)ing lj>8
Grinding wheels, dust renu)val from 9b
Guarantees, determination of temperature 463
Guarantees for ^
air washing ^44
apparatus ^43
cooling ^44
ducts and outlets ^43
engines o43
fans o*"*
673
PAN ENGINEERING BUFFALO FORGE COMPANY
Guarantees for (continued) page
heaters 543
heating 543
humidity control 544
mechanical draft 544
ventilating - 543
Heat
latent 31
sensible 31
total 31
Heat losses
due to infiltration 55
from buildings 46
from galvanized iron pipes 54
from various materials 51 52
,
through cold storage insulation 53

Heat radiated by
gas and electric lights 62
machinery 62
occupants 62
sun 62
Heat required to condition air 73
for ventilation 56
Heat transfer, steam to water 404
Heat transfer through metal surfaces 403
longitudinal flow, steam to air 403
transverse flow, steam to air 408
Heat transmission (see heat transfer) 403
Heated buildings, humidity in 67
Heater cases, dimensions of Buffalo 456
Heater connection, length of 457
Heater curves, Buff'alo pipe coil 432-438
Vento cast iron 432-438
Heater, engine and fan tables, combination of 206, 518-523
Heater tables, Buffalo pipe coil 418-431
Vento cast iron 439-444
Heater tables and curves, application of 415-417
Heaters, Buffalo pipe coi! 399
blast area of 458
by-pass proportions for 457
curves 432-438
dimensions of cases for 456
length of connection 457
regular open area pattern 399
return bend pattern 399
sizes of 449
steam, drip and air connections for 452-455
tables 418-431
velocity of air through 412-414
Heaters, condensation in coils of 411
gas 46(S
guarantees for 543
indirect 4r).S
specifications for 534
temperature attained with 411
Heaters, Vento cast iron 401
curves 432-438
sizes 460-462
tables 439-444
574
INDEX
__ PACK
Heating 46
airquantity and final temp, recjuiroil for 56
fan system for 62
guarantees for 543
relation of humidity to 30
Heating and ventilating, selection of apparatus for 506-518
Height of chimneys 548
High speed multiblade fan specifications 533
Horizontal engines (see steam engines) 482-505
Horsepower of a fan 175
of engines (see steam engines) 482-505
Humidifiers 47I
design of 472
power required to operate 474
Humidity 28
absolute 28
on velocity of air
effect of 27
relation of to heating 30
relative 28
Humidity control
automatic 74
dew-point method of 72
guarantees for 544
si)ecifications for 542
Humidity in heated buildings 67
in manufacturing establishments 68
Indirect heaters 458

Indoor temijeratures, average for various outdoor temperatures 464
Induced draft 82
Induced draft fan tables 321
Niagara Conoidal fans 326, 327
Planoidal fans 322-325
special steel plate fans and engines 328-330
Industrial buildings, proportioning piping for 135
ventilation of 65
Infiltration, heat losses due to 55
Installation tests of fans, directions for making 525
Jones underfeed stokers, mechanical draft with 91
Latent heat 31
Laws of fan performance 179
Layout of piping 141
Legislation recommended for motion picture shows 564-567
Length of heater connection 457
Lights, gas and electric, heat radiated by 62
Localities, mean monthly temperatures in 465
Logarithms 560, 561
Longitudinal flow-heat transfer 403
Lumber dryers 559
Machinery, heat radiated by 62

Manufacturing establishments, humidity in 68
Maximum economy, friction loss through heater for 415
proportioning piping for 130
Maximum horsepower of engines (see steam engines) 498-505
Mean effective pressure of liigh speed steam engines 485
Measuring air quantity 625
Measurement of air used in boiler i)laiit 89
of pressure produced by a fan . 526
Measures of pressure and weight 563
575
. -
Mechanical draft gj*

forced draft oj
induced draft 09
induced draft Planoidal fan tables '.'.
. 322-325
induced draft Niagara Conoidal fan tables .
'.
326 327
induced draft special steel plate fan tables .328-330
Mechanical draft guarantees caa
Mechanical draft in connection with mechanical stokers on
Mechanical stoker.s, mechanical draft with .**
gX
Miscellaneous applications of fans '
Moisture removing capacity of air in fan svstem dryers

, 206
, * 70
Monthly temperatures in different localities ' ' '
Motion picture shows, legislation recommended for 4^5

=ifi4-'ifi7
Motor specifications coc
Murphy automatic stokers, mechanical draft with o?
.Multiblade disk fans o^k
Multiblade fan specifications .................'. 532
Niagara Conoidal fan.s
227
efficiencies of 974 r,yi
outlet velocities of '.'.'.'.'.'.'.'.'..'. '
927
performance curve of " 27fi
selection of fX^
specification tables of .....'
. . . . .
545 547
Niagara Conoidal fan capacity tables 227-27^
induced draft 32^ 097
static pressure
232-273
total pressure
228-331
Niagara Conoidal fan dimension tables
double width 00.^ oc
full housing '.'.'...'.....
'-^-r-a-s
overhung wheel '^86-393
three-quarter housing
379-381
Nozzles (see diverging and converging nozzles)
120-126
Occupants, air required by -n

heat radiated by ^9
Orifice, pres.sure losses in ,
2n
use of in fan testing
198-201
Outdoor temperatures, average indoor temperatures for 464
Outlet guarantees .1^
Outside wet-bulb temperature, relation of room temperature ito. . .
^72
Parsons mechanical stoker, mechanical draft with on

Per cent, e.xcess air, relation of to draft requirement and boiier
Performance curves, u.se of
eff 87
20"
Niagara Conoidal fans
276
Planoidal blowers 294 995
Planoidal exhausters o'ia.' o1\
propeller wheels
slow speed planmg-mill e.xhausters 3o4
347
steel pressure blowers oon
Turbo-Conoidal fans . . . . .
'.
. .
'300
Performance of fans, laws of ^ ^
'1
-n
Pettenkofer method of analy.(is for carbon dioxide
6]
Pipe coil heaters (see heaters and Buffalo heaters) 300
blast area of ;
4^^
by-pass proportions for '
A-
curves '
432-4.3S
dimen.sions of heater case for 45^;
friction of air through 446
length of connection ' 45-
regular open area pattern '
399
return bend pattern .......... 399
676
INDEX
Pipe coilheaters (continued) page
sizesand dimensions of 449
steam, drip and air connections of 452-455
tables 418-431
velocity of air through 411
Pifje traverse for pitot tube 196
Pipes, carrying capacity of 143, 154, 155
diameter of branch 144-152
size of steam 554-557
Piping, gauges of iron used for 158
layout 141
pressure to overcome friction of air passing through 112, 113
proportioning to allow for friction 138
proportioning for exhaust systems 129
proportioning for industrial buildings 135
proportioning for public buildings 132
proportioning various losses in 127
sizes and weights of 158
weight of black steel 162
weight of galvanized iron 160, 161, 163-167
Pitot tube 190
pipe traverse for 196
use of in an air duct 195
Planing-mill exhaust fans, slow speed 344, 345
Planing-mill exhaust fans, standard 341 342
,

dimension tables 394 395
,
Planing-mill exhaust systems (see exhaust systems) 93

design of 103
dust collectors 101, 102
friction loss 102
hood construction 93
pipe sizes for 94
size of main pipe 95
size of piping 93
velocity required 95
Planoidal fans
performance curves of blowers 224, 225
performance curves of exhausters 214, 215
selection of 182
specification table of 545
Planoidal fan capacity tables
blowers ^^ ^~?
o
exhausters ~?^~ooc
induced draft fans 322-325
Planoidal fan dimension tables
blowers, full housing 359-362
exhausters, full housing 359-362
exhausters, overhung wheel 366-373
exhausters, three-quarter housing 363-365
Power to operate humidifiers 474
Pressure, velocity and air density relations in fan work 178
Pressure and weight, measures of 563
Pressure, measurement of produced by a fan 526
relation of to velocity 16
relation of total, static and velocity in fans 176
Pressure characteristics of fans ' 172
Pressure losses in air ducts ..
\]yf.
in diverging and converging nozzles and in orifices l^-O
Pressure to overcome friction of air passing through piping 112, 113
577
PAGE
Prupeller fans 351
Projicrties of air 11
effect of temperature on 27
relation of altitude to 24
Properties of dry air 13
of saturated air 15
of saturated steam 558
Proportioning piping practical applications 131
Proportioning piping in exhaust systems 129
in industrial buildings 135
in public buildings 132
to allow for friction 138
for maximum economy 130
Proportioning various losses in piping 127
Proportions of heater by-pass 457
Psychrometer, aspiration 29
sling 29
Psychrometric charts and tables, use of 32
charts 35-37
tables 38-45
Public buildings, allowing for friction in 134
duct velocity in 134
register velocity in 133
sizes of registers and risers in 133
Pumps and receivers, automatic
centrifugal 467
duplex 466
Purity of air 58
Quantity of air and final temperature required for heating 56
Rectangular ducts, circular equivalent of 156, 157

Refrigeration required for dehumidifying 474
Refuse exhaust system (see exhaust systems) 93
Register velocity in public buildings 133
Registers, sizes of for public buildings 133
Regular open area pattern heaters (see Buffalo heaters) 399
Relation of altitude to properties of air 24
of boiler eff. and draft requirements to per cent, excess air 87
of cooling effect to per cent, relative humidity 69
of dry-bulb, wet-bulb and dew-point temperatures 30
of humidity to heating 30
of pressure, velocity and air density in fan work 178
of room temperature to outside wet-bulb temperature 72
of total, static and velocity pressure in fans 176
of velocity to pressure 16
Relative humidity 28
relation of cooling effect to per cent, of 69
Relative volumes of air at various temperatures 13
Removal of bodily heat 59
Return bend pattern heaters (see Buffalo heaters) 399
Risers, sizes of for public buildings 133
Room temperature, relation of to outside wet-bulb temi)erature 72
Saturated steam, properties of 558

Schoolhouse ventilation . . . . 65
Selection of apparatus for heating and ventilating 506-518
Selection of a fan 181
Niagara Conoidal 186
Planoidal ; 182
Sensible heat 31
Short length of pipe, use of in fan testing 201
578
1
INDEX
PAOE
Size of steam pipes . 554-557
Sizes of piping 158
Sizes and dimensions of Buffalo heaters 449-451
of Vento heaters 460-462
Sling psychrometer 29
Slow speed planing-mill exhaust fans 344, 345
Special steel plate induced draft fans and engines 321, 328-330
Special types and features of fans 174
Special ventilation requirements 61
Specific heat of air 16
of various substances 78
Specific weight of water vapor 14
Specification tables of
Niagara Conoidal fans 546, 547
Planoidal fans 545
Turbo- Conoidal fans 546, 547
Specifications for
air washer 537-540
duct work 535
fan engine 535
fan system heater
534
.
high speed multiblade fan 533

humidity control 542
motors 536
multiblade fan 532
steel plate fan 53
Spiro turbines 484
Spray type dehumidifiers 474
Standard planing-mill exhaust fans 341 342 ,
capacity tables ".

342, 343
Standards of ventilation 58
Steam, drip and air connections of Buffalo heaters
regular open area pattern 452, 453
return bend pattern 454, 455
Steam engines 482
mean effective pressure of high speed 485
water rates of 486, 487
Steam engines, brake horsepower of
Buffalo class "A" 488-491
Buffalo class "A" at various speeds and pressures 496-497
lûffalo class "I" (cvlinder below shaft) 494
Buffalo class"N", long stroke 492
Buffalo class"S", long stroke 492
Buffalo double vertical double acting 493
Buffalo low pressure 495
Steam engines, dimensions and maximum horsei)ower of
Buffalo class "A" horizontal center crank 498
Buffalo class "A" horizontal side crank 499
Buffalo class "N" horizontal side crank 500
Buffalo class "S" horizontal side crank 501
Buffalo class "A" vertical 503
Buffalo class "I" vertical 504
Buffalo class "O" vertical 502
Buffalo double vertical double acting 505
Steam engines direct-connected to special induced draft fans 328-330
Steam engine guarantees 543
specifications 5*j^
Steam tables 558
Steam turbines 484
Steel plate fan specifications 531
Steel plate pressure blowers ^^^
579
PAGE
Steel pressure blowers 336
application to forge fires 338
capacity table 337
dimension table 398
Stokers with mechanical draft 90
Jones underfeed 91
Murphy automatic 91
Parsons mechanical 90
Taylor 91
Taylor stoker with mechanical draft 91

Temperature, effect of on properties of air 27
effect of on velocity of air 24
effect of on volume of air 26
Temperature attained with indirect heaters 411
Temperature conversion table 552
guarantees, determination of 463
in Carrier humidifying system 71
Temperatures, mean monthly in different localities 465
relation of dry-bulb, wet-bulb and dew-point 30
Testing fans (see fan testing) 190
Tests, directions for fan installation 525
method of calculating fan 529
Theoretical outlet or blast area 126
Thermometers, dry-bulb and wet -bulb 28
Tile sizes for forge shops 108
Total heat 31
Transmission of heat (see heat transfer) 403
Transverse flow-heat transfer 408
Traverse, pipe 196
Turbines, Srîro steam 484
Turbo-Conoidal fans 277
outlet velocities of 277
performance curve of 276
selection of 277
Turbo-Conoidal fan capacity tables 277 319
Turbo-Ccnoidal fan dimension tables
double width 3S2 385
full housing 375-378
overhung wheel 386-393
three-quarter housing 379-381
Types of humidifiers 471
Upward tiystem of ventilation 64

Useful factors 562
Velocity of air at different pressures 21

at different pressures and temperatures 22, 23
effect of barometric pressure on 24
effect of humidity on 27
effect of temperature on 24
in ducts for industrial buildings 129
in ducts for public buildings 134
through heaters 412
through piping for maximum economy 130
Velocity, pressure and air density relations in fan work 178
Velocity to pressure, relation of 16
Ventilation, fan system for 62
heat required for 56
of industrial buildings 65
of schoolhouse 65
standards of 58
upward and downward systems of 64
Ventilating guarantees '
543
580
INDEX
PAGE
V ento cast iron heaters 401
curves 432-438
sizes and dimensions of 460-462
tables 439-444
Vertical engines (see steam engines) 482-505
Volume of air, effect of temperature on 26
Water conversion factors 563

Water rates of steam engines 486, 487
Water vapor, specific weight of 14
Weight of air 11
of dry air at various temperatures 13
of saturated and partly saturated air 17
Weight of piping 158
black steel piping 162
galvanized iron piping 160, 161, 163-167
Weight and pressure, measures of 563
Wet-bulb temperature, relation of dry-bulb and dew-point to 30
Wet-bulb thermometer 28
581
O
CD
"1
CD
Q-
o
CD
crq
CD
CD
3
CD
S3
Q
no
o
o
CD
o
=3
OP
CD
CD
=3
3
o
r3
o
;- ^.-^

Fan Buffalo

Uploaded by

Copyright:

Available Formats

Fan Buffalo

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fan Buffalo

Uploaded by

Copyright:

Available Formats

o

OF TABLES, CHARTS AND DATA ^-^

WILLIS H. CARRIER, Chief Engineer

the architect become familiar with the fundamental principles

This book is intended to be used as a guide in the selection

The information herein presented is in complete and reliable

Section V Exhaust Systems.

Section VI Miscellaneous Applications.

Section II Fan Testing.

Section IV Fan Dimensions.

Section VI Air Conditioning Apparatus.

Section VII Steam Engines.

Section VIII Practical Applications and the Selection of

sidered as consisting of oxygen and nitrogen, but also containing <^

FAN ENGINEERING BUFFALO FORGE COMPANY

where t = temperature, deg. Fahr.

The relationship temperature and specific

It may be noted from the curve on page 14 that the value of

PROPERTIES OF DRY AIR

Specific Weight of Water Vapor

The table on page 13 gives the properties of dry air for

Example As an example of the use

PROPERTIES OF SATURATED AIR

'3UI '83Q Jdd ooo ooo ooo

applied to the flow of air as are commonly used for th^flow of

where Vb = velocity in ft. per second, or

where V = velocity in ft. per min.

where h'= head expressed in in. of water

Then at 70 F. and 29.92'' barometer and with dry air

Thus we see that the velocity at standard conditions stated for

4005 a/i.734 = 5273 ft. per min. (14)

The weight of dry or saturated air at other temperatures

where V= velocity in ft. per min.

The quantity of air discharged through an orifice or nozzle

p = pressure head in in, of water causing flow of air

For values of coefficients of discharge see "Coefficients of

and Q = 1444.5 CA-v/J^

Example . As an example of the application of the above we

at 29.40 inches. What will be the velocity of this air at a

The above formulae are sufficiently accurate for low pres-

V2 = 6552^Ti[l- (^)'''] (19)

where V2 = velocity in ft. per min. at discharge.

where Q = cu. ft. free air per min.

CORRESPONDINQ PRESSURES AND VELOCITIES OF DRV AIR AT 70

CO-* 00 <*?0<N Mt-O Tt^M-* OOOM J^MO ^^--^OJ

rfiMM OOIMI^ Tt^t^C^I t^00 ICM

ooow MOCO 00 (MOO

^^O M.^^ OOt^O ^MOO Tt*iC'-' COO

r^^ lor^o OCC-* Tft^iO iOOO QCt>-

(NXt^ OCOt <NcOt1< CDOiO M<N0 ^'<^22 S'^

'înê <ôi "^"^"J

Effect of Temperature and Barometric Pressure on Velocity

or where we wish to correct for both temperature and barometer

In the above formulae V represents the velocity of the air at

Relation of Altitude to the Properties of Air

AdnD^Bt^i JO 93H0NI Nl 3dnSS3dcl Dldl3jM0dV8

Effect of Temperature on the Volume of Air

The effect of temperature on the various properties of air is

Actual vel. Cu. ft. per lb. air as observed

in the atmosphere, and the weight of water vapor a given space

CO-* 00 <?0<N Mt-O Tt^M- OOOM J^MO ^^--^OJ