-, .'I.

■ "M/mijHpii in*«

AD-783 840
A HYSTERESIGRAPH FOR PLOTTING MAGNETI-
ZATION CURVES
Reinhold W. Kubach
Dayton University

Prepared for:
Air Force Materials Laboratory

31 December 1966

DISTRIBUTED BY:

urn
National Technical Information Service
U. S. DEPARTMENT OF COMMERCE
5285 Port Royal Road, Springfield Va. 22151

iiutM,.»^^. ,..-v^--,.. .-■ .,■.,-:,.„ gj -:-'..-J»aaM«lMlMJiliJ^.Ji^»i—t^^.J..»,J.^-ww,-j.-j-. ■■■■..■ ..,->-...,... II 'r!il-f^iiiiV|-":""J--J- ' ■ ■-■ •ill'Hjr"^"'""--■■»~-~^-^'-'>~—^«"J^
 NOTICE

When Government drawings, specifications, or other data are

used for any purpose other than in connection with a definitely related
Government procurement operation, the United States Government
thereby incurs no responsibility nor any obligation whatsoever.: and
the fact that the Government may have formulated, furnished, or in
any way supplied the said drawings, specifications, or other data, is
not to be regarded by implication or otherwise as in any manner
licensing the holder or any other person or corporation, or conveying
any rights or permission to manufacture, use, or sell any patented
invention that may in any way be related thereto,

ttCESIC* far

MS »»its >\;CtlM

«UIIIRVEI
xmiunw

ST
ÜSTRISüTSW/iMlUBUhf C ES

"""flit ~kTnlLiii f.""if.-.uf

ft,

Copies of this report should not be returned unless return is

required by security considerations, contractual obligations, or notice
on a specific document.
-'"MK-

UNClASSIFJEJi.
Security Classification
DOCUMENT CONTROL DATA • R&D
£±Z£*L£JZ£
(Smcurilr elm filiation o/ Wtfa. body of «tamcl and imfoain« annotation mutt b» mntmnd »*«n 0» ovaraif »port it claititltd)
1 ORIGINATING ACTIVITY fCoiponl« author.) 2«. REPORT SECURITY CLASSIFICATION
University of Dayton Research Institute UNCLASSIFIED
300 College Park 2 6 SROUP
Davton. Ohio 45409
3- REPORT TITLE

A HYSTERESIGRAPH FOR PLOTTING MAGNETIZATION CURVES

4 DESCRIPTIVE NOTES (Typm of and IncJtiaiw dmf)

Tecbni»,«! Report 1 July 1966 to 31 December 1966
S AUTHORS (Latt >. tint MM, Inilimt)

KUBACH, REINHOLD W.
6 REPORT DATE
December 31, 1966
7«. TOTAL NO, Of PACES
57
7b. NO.
w
or »EPS

8a. CONTRACT ÖD «RANT NO • a. ORiaiNATOR'S REPORT NUMBERfSJ

AF 33{615)-Vr^3
6. PROJECT NO. J ' I •>

7367
c »6. OTHER REPORT NOfSJ (Any otharnunjxil 4illMy6tliil(nif
tfw» report)

M 736703
10 AVAILABILITY/LIMITATION NOTICES
This document has been approved for public release and sale; its distribution
is unlimited.
11 SUPPLEMENTARY NOTES 12. SPONSORING MILITARY ACTIVITY

Air Force Materials Laboratory

Research and Technology Division
13. ABSTRACT
Air Force Syatema Command
-!■■■

Wright-Patterson AFB, Ohio 45433

The objective of this study is to design, build, and calibrate a hysteresigraph

for plotting magnetization curves of small cylindrical rod samples of dilute fine
r. particle magnets. The magnetization of the sample and the field near its surface
are measured with a set of balanced sensing coils. The voltages induced in these
coils when the magnetic field is slowly changed are amplified and integrated
electronically by means of two operational amplifiers (Miller integration), whose
outputs drive an X-Y recorder.
The instrument must record hysteresis loops at peak field strengths up to
20,000 Oe with the expected values of coercive forces ranging from 100 to 10,000
Oe. The peak induction values may be expected to be in the range from 300 to
22,000 Gauss.
The instrument has been used to measure hysteresis loops of various permanen
magnets, and coercive forces of powdered rare earth-cobalt alloys which are
under development as new magnetic materials.
""NATIONAL TECHNICAL
INFORMATION SERVICE
U S Department of Commerce
Springfield VA 22151

DD FORM
1 JAN 64 1473 UNCLASSIFIED
Security Classification
 ■r^*??*&*mmMmmmm*m

UNCLASSIFIED
Security Classification
u LINK A LINK LINK C
KEY WORDS
BOLE

Hysteresigraph
Ferromagnetic Materials
Magnetization curves

INSTRUCTIONS
1. ORIGINATING ACTIVITY: Enter the name and address imposed by security classification, using standard statements
of the contractor, subcontractor, grantee, Department of De- such as:
fense activity or other organization (corporate author) issuing (1) "Qualified requesters may obtain copies of this
the report. report from DDC"
2a. REPORT SECURITY CLASSIFICATION: Enter the over-
(2) "Foreign announcement and dissemination of this
all security classification of the report. Indicate whether report by DDC is not authorized."
"Restricted Data" is included. Marking is to be in accord-
ance with appropriate security regulations. (3) "U. S. Government agencies may obtain copies of
this report directly from DDC. Other qualified DDC
2b. GROUP: Automatic downgrading is specified in DoD Di- users shall request through
rective 5200.10 and Armed Forces Industrial Manual. Enter
the fcroup number. Also, when applicable, show ihat optional
markings have been used for Group 3 and Group 4 as author- (4) "U. S. military agencies may obtain copies of this
ized. report directly from DDC Other qualified users
3. REPORT T"iL,E: Enter the complete report title in all shall request through
capital letters. Titles in all cases should be unclassified.
If a meaningful title cannot be selected without classifica-
tion, show title classification in all capitals in parenthesis (5) "All distribution of this report is controlled. Qual-
immediately following the title. ified DDC users shall request through
4. DESCRIPTIVE NOTES If appropriate, enter the type of
report, e.g., interim, progress, summary, annual, or final. If the report has been furnisheJ to the Office of Technical
Give the inclusive dates when a specific reporting period is Services, Department of Commerce, for sale to the public, indi-
covered. cate this fact and enter the price, if known.
5. AUTHOR(S): Enter the name(s) of author(s) as shown on 1L SUPPLEMENTARY NOTES: Use for additional explana-
or in the report. Entei last name, first name, middle initial. tory notes.
If military, show rank *nd branch of service. The name of
the principal ° <thor is an absolute minimum requirement 12. SPONSO iwG MILITARY ACTIVITY: Enter the name of
the departmental project office or laboratory sponsoring (pay-
6. REPORT DAT^ Enter the date of the report as day, ing fcr) the research and development Include address.
month, year; or month, year. If more then one date appears
■^n ihe report, use date of publication. 13. ABSTRACT: Enter an abstract giving a brief and factual
summary of the document indicative of the report, even though
7a. TOTAL NUMBER OF PAGES: The total page count it may also appear elsewhere in the body of the technical re-
should follow normal pagination procedures, i.e., enter the port. If additional space is required, a continuation sheet shall
number of pages containing information. be attached.
76. NUMBER OF REFERENCES: Enter the total number of It is highly desirable that the abstract of classified reports
references cited in the report. be unclassified. Each paragraph of the abstract shall end with
8a CONTRACT OR GRANT NUMBER: If appropriate, enter an indication of the military security classification of the in-
the applicable number of the contract or grant under which formation in the paragraph, represented as (TS). (S), (C). or (V).
the report was written. There is no limitation on the length of the abstract. How-
86, 8c, 8t 8J. PROJECT NUMBER: Enter the appropriate ever, the suggested length is from ISO to 225 words.
military department identification, such as project number,
subproject number, system numbers, task number, etc. 14. KEY WORDS: Key words are technically meaningful terms
or short phrases that characterize a report and may be used as
9.i. ORIGINATOR'S REPORT NUMBER(S): Enter the offi- index entries for cataloging the report. Key words must be
cial report number by which the document will be identified selected so that no security classification is required. Identi-
and controlled by the originating activity. This number must fiers, such as equipment model designation, trade name, military
be unique to this report. project code name, geographic location, may be used as key
96. OTHER REPORT NUMBER(S): If the report has been words but will be followed by an indication of technical con-
assigned tmv other report numbers (either by the originator text. The assignment of links, rules, and weights is optional.
or 6v the sponsor), also enter this numberts).
10. AVAIL ABILITY. LIMITATION NOTICES. Enter any lim-
itations on further dissemination of the report, other than those

AFLC-WF»AF B-OCT hf> fiVI

w UNCLASSIFIED
Security Classification

■^jmk I ^■ftg»^:*"'-^.- ;..^.,.;■.'.■:.■■-..;■ ■ ■ ■■■ ■ —-- ., , ■■■■■V^

umiil IMliMili - - - _^ ; —
A HYSTERESIGRAPH FOR PLOTTING MAGNETIZATION CURVES

Reinhold W. Kubach
University of Dayton

This document has been approved for public

release and sale; its distribution is unlimited.

lt>

. ..,.,^.„_-^^_i^-_^^. ; ■;■■ -., , ., .. ; , _, _, ■•■■■■■■ ^<v^-^«^

 »«■»»«»«»«•sgnsssWM'SW'W?

FOREWORD

The work described in this report was carried out in the labora-
tories of the Material Physics Division under USAF Contract No.
AF 33(615)-3793 entitled "Research on Intermediate Phases of Rare
Earth Metals". This contract was initiated under Project No. 7367,
"Research on the Characterization and Properties of Materials",
Task No. 736703, "Electronic and Magnetic Properties of Materials".

The woik was administered by the Air Force Materials Labora-

tory, Research and Technology Division, Air Force Systems Com-
mand, Wright-Patterson Air Force Base, Ohio, Dr. Karl Strnat,
Project Engineer.

This report covers work conducted during the period Julyl, 1966
to December 31, 1966. The manuscript was released by the author
in December 1966 for publication as a technical report.

The author is indebted to Dr. Karl Strnat of the Air Force

Materials Laboratory for helpful discussions and to Mr. John C. Olson
for his assistance in building the apparatus and preparation of the
samples. Special thanks are due the management of the Electro-
magnetic Materials Branch for the use of its facilities.

This technical report has been reviewed and is approved.

AW
Major David J. Iden
Chief, Electromagnetic Materials Branch
Materials Physics Division
Air Force Materials Laboratory

_-i-r ve rriij-tfMi-
i;tf-MfflWff^fr'---:"
Wl\

ABSTRACT

The objective of this study is to design, build, and calibrate a

hysieresigraph for plotting magnetization curves of small cylindrical
rod samples of dilute fine particle magnets. The magnetization of the
sample and the field near its surface are measured with a set of bal-
anced sensing coils. The voltages induced in these coils when the mag-
netic field is slowly changed are amplified and integrated electronically
by means of two operational amplifiers (Miller integration), whose
outputs drive an X-Y recorder.

The instrument must record hysteresis loops at peak field strengths

up to 20, 000 Oe with the expected values of coercive forces ranging
from 100 to 10, 000 Oe. The peak induction values may be expected to
be in the range from 300 to 22, 000 Gauss.

The instrument has been used to measure hysteresis loops of

various permanent r.iagnets, and coercive forces of powdered rare
earth-cobalt alloys which are under development as new magnetic
materials.

iii

■,, MBMii-aiiirtiüitfcii i • I I
- -- ■- ■ -•■ -'■■■ ■■-:—
\
^-^.^M-^*»**«'^**^^

TABLE OF CONTENTS

Title Page

SECTION I INTRODUCTION 1.

1. HYSTERESIS LOOPS AND MAGNETIC MATERIALS 1.

2. SPECIFIC AIM OF PROJECT 1.

3. DESIGN SPECIFICATIONS 2.

SECTION II BASIC CONCEPTS OF MAGNETIC

MEASUREMENTS 4.

SECTION in THEORY OF HYSTERESIGRAPH 8.

1. SCHEMATIC OF A HYSTERESIGRAPH 8.

2. SENSING COILS 8.

3. ELECTRONIC INTEGRATION 10.

4. ANALYSIS OF INTEGRATOR ERROR 17.

SECTION IV DESIGN AND CONSTRUCTION OF THE

HYSTERESIGRAPH 19.

1. RECORDER AND AMPLIFIERS 19.

2. DETERMINATION OF INTEGRATING NETWORKS 19.

3. INTEGRATOR PERFORMANCE 21.

4. COIL CONSTRUCTION 23.

5. COIL COMPENSATION 26.

SECTION V CALIBRATION OF THE HYSTERESIGRAPH 31.

1. CALIBRATING PROCEDURE 31.

2. CALIBRATING THE H COIL 31.

3. CALIBRATING THE (B-H) COIL 31,

iv

■inii iii I-IIIIM-^

■ . . . . .- ->.*■ . '.■ .- .- ■ . — '■ ■ »" ■ " ■' ' '
II

TABLE OF CONTENTS (con't)

1.C
TiÜe Page

4. OVERALL ACCURACY 33.

SECTION VI OPERATION OF THE HYSTERESIGRAPH 34.

SECTION VII SUMMARY 38.

1. SIGNIFICANCE OF THIS CONTRIBUTION 38.

2. RECOMMENDATIONS FOR FURTHER STUDY 38.

3. FUTURE USE 39.

BIBLIOGRAPHY 40.

ir~ii'iiiiäiti'--'""'-,,""-i'iii"fcr«iiii-'"—«— -■■'!iiiiiifn'iiiMW-iitinwiMiti''iifci»iim'nr^*',*"~''"-- -'"---j*-j--^-»-^»-™~-«---------'-'-":*-j-iii-iiiiiiaiir^ifttriitiiTrii-T ,,>■!■,.-.i-.ijayufiiMialiliatiaüllili

n ..... ^.,.^,,. ■>.^:.-;,;^»--^^ :.■:-„ ,.^ ,.,

LIST OF FIGURES

Figure Page

1. Typical Hysteresis Loops (a) B Versus H Loop

(b) 4wM Versus H Loop 2.

2. Basic Circuit for Measuring B Versus H Loops of

Ring Shaped Samples

3. Simplified Schematic of Hysteresigraph for the

Measurement of B Versus H Loops 5.

4. Schematic Arrangement for Measuring Surface B

and H Fields of Rod Samples

5. Elementary Hysteresigraph for Plotting B Versus

H Loops of Rod Samples 7.

6. Elementary Hysteresigraph for Plotting 4FM Versus

H Loops 8.

7. Arrangement of Sensing Coils for Measuring Surface

B and H Fields of Rod Samples 9.

8. General Operational Amplifier 11.

9. Approximate Equivalent Circuit of Operational

Amplifier 13.

10. Approximate Equivalent Circuit of Integrating

Operational Amplifier 14.

11. Integrating Operational Amplifier Circuit 14.

12. Equivalent Circuit of General Operational Amplifier 15.

13. Equivalent Circuit of Miller Integrator 16.

14. Passive Integrating Circuit 16.

15. Integration of a Step Function 17.

16. Response of Miller Integrator to Step Input 22.

(Long Time Integration)

vi

•III"*'*-'
■ .. .—^—■
%&J»s»oto^.

LIST OF FIGURES (con't)

Figure Page

17. Response of Miller Integrator to Step Input

(Short Time Integration) 24.

18. (a) Cross Section of Coil Arrangement

(b) Coil Form 25.

19. Remaining Coil Imbalance 27.

20. Schematic of Hysteresigraph 29.

21. Magnet and Hysteresigraph Set Up for Testing 30.

22. Coil Calibration ( urves 32.

23. 4TTMVersus H Loop of a Typical Ce (Ce-Rich

Mischmetal) Co,. Sample 35.

24. 4irM Versus H Loop of a Typical Alnico VI Sample 36.

25. 4wM Versus H Loop of a Typical SmCoc Sample 37.

vii

. ■iiil>MiMi»BiifMiin>iiriiiiltiit,»<inlriiftTirin-'"~- "*- —'- "'-■«i-"^—' --— -■■.-»■* • mi ■niirt—m ii i ii

— ——-—■
 .~««vv.

SECTION I

INTRODUCTION

1. HYSTERESIS LOOPS AND MAGNETIC MATERIALS

Hysteresis loops provide essential information which quantitative-

ly describe the properties of ferromagnetic materials as indicated in
Figure 1. Consequently the measurement of hysteresis loops consti-
tutes one of the major tasks in the development of magnetic materials.
Reference to Figure 1 also shows that there are two ways of plotting
hysteresis loops. In one version, loop (a), the magnetic induction B
(measured in Gauss) is plotted as a function of the exciting field H
(measured in Oersted). In this form the hysteresis loop is most use-
ful to the engineer engaged in the design of magnetic circuits. In the
other version, loop (b), the intrinsic induction, 4TTM (in Gauss) is plotted
versus the field H (in Oersted). M, the magnetization or the magnetic
moment per unit volume is measured in ÜIBB, = \-jr Gauss and is related
to the induction B by the equation cm"

B = H + 4TTM (1)

Since the magnetization M and the coercive force *HC are closely re-
lated to the microscopic properties of the material, 47rM versus H plots
are most important to the physicist engaged in research and development
of magnetic materials.

The data for plotting hysteresis loops were previously obtained by

a point-by-point measurement method which was tedious and time-
consuming. For this reason instruments known as hysteresigraphs
for drawing B versus H loops automatically were developed.

2. SPECIFIC AIM OF PROJECT

The hysteresigraph to be described here was exclusively devel-

oped for plotting 4TTM versus H loops. None of the hysteresigraphs
reported in the literature were readily adaptable for this use, nor were
*hey commercially available. Furthermore, the particular application
of this instrument created additional design problems which are not
characteristic of those reported in the literature (References 1 - 4).
This new type of hysteresigraph is used in an extensive research program
conducted at the Air Force Materials Laboratory, aimed at finding new
„naterials for possible use as permanent magnets (Reference 5). One
phase of this program deals with the systematic evaluation of the mag-
netic properties of crystal ^nisotropy-controlled powder magnets of the
rare earth-cobalt intermetallic compounds YCoc, SmCo^, Co (Ce-rich
mischmetal) C05 and others. A large number of samples arc to bo

■ftWli^^,...-.^. ...
■-C-- ■-■• "■
m

tested which are typically 15mm long x 11mm diameter, and only a
small packing fraction (typically 5 v/o) is active magnetic material.

B [GAUSS]

4»M [GAUSS] r(o)Bv».H

(b)4rMv*.H
DEMAGNETIZATION
CURVES

/(SATURATION)

MC ' "8
/ —— H [OERSTED]

/' MHCJtBHC(C0ERCIVE F0RCES)

'/ B»H + 4«-M

Figure 1. Typical Hysteresis Loops (a) B Versus H Loop

(b) 4wM Versus H Loop

Thus the new hysteresigraph has to be capable of plotting 4irM versus

H loops of extremely "weak" magnets, but also dense and "strong"
magnets.
I
3. DESIGN SPECIFICATIONS

Because of the wide scop 3 of the entire program the specifications

for this instrument were rather severe. The design specifications
were as follows:
a) Obtain 4irM versus H loops of dilute, i. e., weak magnets
which permit one to measure coercive forces H over the
range of 100-10,000 Oe at H «30,000 Oe. Tne expected
range of 47rMg = Bg = 6000 -21, 500 Gauss for the magnetic
portion of the sample which is present at packing fractions
of«5%. The values of Hc should be readable to + 5 Oe.
Measurement of the small 4trM signals may be highly inac-
curate (+ 25%).
2.

■ ■■■■..-, :^'T v;,-, ■'■■":■-£ mmttWmmä *&*"**** --.-V-—;.;•. -'■■■- ■■ ■-•'■■-''■ - ■■■■ •■^-■■■y-^
 b) Obtain 4vM versus H loops for dense magnets, having packing
fractions of 30-100%, with high accuracy for the H-measuremeni
(+ 5 Oe) as well as the 4irM-measurement (4; 20 G). From
these curves B versus H loops are to be derived and the energy
product determined.

3.

rnWMiMim if'"''''"^^ .... .^.—«. 1,, ,1 I, «maiiBÜttMftü^lflü

 SECTION II

BASIC CONCEPTS OF MAGNETIC MEASUREMENTS

The quantities B and H are conveniently found by measuring the

induced voltage in a coil. The instantaneous voltage, e, induced in a
coil of N turns, when the magnetic flux <f> is changing, is given by
Faraday's law as
d<f> -8
e = - N^-C-. x 10 volts (2)
dt
Faraday's law holds whether the coil is stationary and the flux changes
or the coil is moved through a uniform but time invariant field. The
flux change is found by integrating Equation (2) as

*--#!£-. )

which con be written as

g
A*= *,-*2=-£-/tt£edt (4)
The change in flux density B is obtained from Equation (4) as
A<p 108 rtz
* """"'S
where A is the cross-sectional area of the test sample in cm2. It
follows from Equation (5) thatAB is determined if / edt can be measured.

The classical instrument for performing the integration /edt is

the ballistic galvanometer. Commercial flux meters differ from the
basic ballistic galvanometer on account of their much heavier electrical
damping, the torque compensation of their moving coil suspension, and
their portability.

B versus H loops can be measured with such a flux meter when it

is used in a circuit as shown in Figure 2. In this arrangement the
ring-shaped test sample of magnetic material carries two windings.
The primary winding (P) is connected to an adjustable constant current
supply and the secondary winding (S) is connected to a flux meter. The
field strength H is calculated from the relation
4wNi
H = (6)
10 x L
where N is the number of turns on the primary winding, 'i1 is the current
as measured by the ammeter A, and L is the mean magnetic path length.
When H is suddenly changed from one value to another, the resulting
change in B induces a voltage in the secondary winding which according

4.

- ■ - : ...... .-■..>■■ ■ ■ ■■■ ■■■■ ■ ■■ : ^- ■ -.■.-■ -•■< ■-:'-- JK.-,-..,.;<'.i;!r';"-'"'---v'.",J" '— .,--,■-■ . —
 to Equation (5) is proportional to the change in B, For a detailed
description of the methods which are commonly used to establish enough
points from which a B versus H loop can be drawn, the reader is
referred to Reference 6. The disadvantage«- some of which have been
pointed out in Section I» have led to the development of hysteresigraphs
for plotting B versus H loops.

-RING SAMPLE

MXWSTABLf -vwÄ-
CONSTANT FLUX
CURRENT METER
SUPPLY

Figure 2. Basic Circuit for Measuring B Versus H Loops of Ring

Shaped Samples

Most of the hysteresigraphs for the measurement of B versus H

loops are built around the basic diagram shown in Figure 3. From an
inspection of this diagram it is seen that the field intensity H is meas-
ured in terms of the voltage e„ which is developed by the exciting
current and applied to the X-axis of the recorder. The induction B is
measured by applying the induced voltage e to the integrator and
B
e dt to the Y-axis of the recorder.
/ ß
INTEGRATOR
RING SAMPLE
/ o
Y-AXIS
X-Y
RECORDER

I- + e
H -
X-AXIS
-O

*—VWV—1--
CURRENT
IREGULATOR)

Figure 3. Simplified Schematic of Hysteresigraph for the Meas-

urement of B Versus H Loops
This system as it stands is not suitable for rapid testing of large
numbers of test specimens, because of the time required to machine

5.

-■■■:■■■■■■ ■'■■■ ".^

jftgäijjflifliiiiii
■ ■■■■■ ■■■;;■. ■■..-:'^.--. ■-•■-■■■-■■ >■ .>■■■ ■■>;-"-: •■ '■■ ■---•■■ •■V

■ ■ ■ ■■.:■.-.-v. .-;■ ■■- ■■-■- ■ ■-"■ ^■^■■^

 «,,w^:^w-^'jwwws'i:^T^M^--tlP :^^?':.-^'i^w^,*^^*ggWf^^1,e|#4^^-^!#^g

and wind each ring. Furthermore, large field intensities as are re-
quired for testing permanent magnets cannot be produced by the pri-
mary winding, as the necessary number of turns or the required cur-
rent become prohibitively large. Consequently hard magnetic materials
are being tested in strong and homogeneous fields of the order of
10-20 kOe which can easily be produced between the pole caps of electro-
magnets, thus necessitating cylindrical rod samples rather than ring
samples. Under these circumstances the measurement of the field
intensity H becomes more difficult because the H field between the pole
caps will be disturbed by the sample's own stray field, known as the
demagnetizing field. The H field referred to in Equation (1) is the true
internal field Hj of the sample and not the applied field Ha. These two
fields differ from each other by the sample's demagnetizing field Hj
according to the relation
Hj = Ha - Hd (7)
The demagnetizing field is given by
Hd = DM (8)
where D is the demagnetizing factor and M is the magnetization of the
sample. The demagnetizing factor is well defined for ellipsoids; for
other sample geometries in which M is not uniform throughout, the
demagnetizing factor is not an exact concept. In practice, average
values of O are ascribed to the various geometries in order to indicate
the magnitude of the internal field (Reference 7).

•« -YOKE
MAGNET MAGNET
COIL COIL

»
* -H COIL
,.fw
4
POLE^
CAP

^POLE ■ ■:}-
I \\
/ '-BCOIL
SAMPLE
x
CAP

Figure 4. Schematic Arrangement for Measuring Surface B and

H Fields of Rod Samples

If M is to characterize the material properties independently of

sample shape or size, 4wM must be plotted as a function of Hj and not
Ha. It is thus necessary either to correct mathematically for the
demagnetizing factor or to try to measure Hi directly. The first
method can be applied to ellipsoidal samples but becomes questionable
for cylindrical rods. Thus, for cylindrical rods, Hj should be meas-
ured directly. Since it is impossible to measure Hj by placing a probe

6.

: : '•■■■■■.■• ■ ..
_^._ __ v ■ ■■■■■.■: '■-. ■-■ ■
 ^sutsamtmummmmmB^mimmK^ MHMinpapHi
'Mmm^vmmmmmmim

within the sample, another alternative makes use of the fact that the
tangential component of the H field goes unchanged through the interface
of two substances. This follows from Maxwell's equations in the elec-
tromagnetic field theory (Reference 7). Thus in the center of the rod
length where the tots1 H is tangent to the surface, the field in the air
just outside the sample is identical »vith the H field in the material. A
schematic for measuring the surface B and H fields of rod samples in
an electromagnet is shown in Figure 4.

Utilization of a coil arrangement consisting of a B coil and an

annular H coil to be described later in Section III, makes it possible
to measure these fields quite accurately in terms of the induced voltages
eg and ejj respectively. If each of these induced voltages is applied
to its own integrator and recorder channel, the result is a hysteresigraph
for plotting B versus H loops of rod samples as shown in Figure 5.

Figure 5. Elementary Hysteresigraph for Plotting B Versus H

Loops of Rod Samples

To plot 4TTM versus H loops of rod samples requires that the annular
H coil and the B coil be interconnected and matched to measure 4TTM=
B-H.

ffijjjjUPjUfrjfritejftHaiflfcBiii^ jjjjMÜ juMaaüüüifii

l^-rw. «■*.<„*,r,„-.,*iaMm f^jgj^jgj^g^ggg^KimiilSiMUSiMMttlStMM^tlKml^m
w*T-rr™--^*—— BWWWBIBFgF**"fe-"**-' '-" -■ -j-w ■**"*üBWiBgJ?1'-*-^'-UM tf-P*-1".■ ^f"t-''W"IJ^»w.'JW ^ ? ■üjpwiH^twvV';■** '- w;-igjW*w;ipfBJB;iifW^wwij ^»*p,"'»y J*J.™n.Ujtf ipilJl J| IJIJUp

SECTION in
THEORY OF HYSTERESIGRAPH

1. SCHEMATIC OF A HYSTERESIGRAPH

From the development of the previous section a tentative schematic

of a hysteresigraph for plotting 4irM versus H loops can be Hrawn as
shown in Figure 6. Whereas the system of Figure 6 is in principle
straightforward, its realization is not, because of problems arising from
the measurement of 4TTM and problems arising from the application of
this system to weak permanent magnets.

-ox

1
INTEGRATORS
n RECORDER

<B-H)
4

-OY

Figure 6. Elementary Hysteresigraph for Plotting 4*M Versus

H Loops

Since 4TTM cannot be measured directly, its determination hinges

on the ability to measure the "true" B and H fields on the surface of
the sample. The design and calibration of a coil arrangement capable
of producing voltages proportional to H, B and B-H contributes the
most important, and as it turns out, the most difficult part of the en-
tire hysteresigraph construction.

2. SENSING COILS

The coil arrangement of the hysteresigraph shown in Figure 6

consists of two coils referred to as the annular H coil and the B coil
and used for the measurement of the B and H fields on the surface of
the sample. If these two coils are interconnected in series opposition,

.8

miiii iiinii-i«iiitiirii III Mil"- tsümmummmätiiimmmm

II.«! !■■. J'.Liw,ijw|j^!_BjMBi!p>)B|.i <miwmw^$**>'k*<imvJtm»!''uii*™t*»i!*mvi-.g ^F*^«H»!PM»4^Myjl »fSEimwuMhi [J»!iapjW!Sjwy, •wv^mw ■ *s ^.f^ismmmmt

a voltage proportional to 4irM is obtained according to the relation

4irM = B - H (9)

H is here the internal field H^, which is reasonably well represented

by the field near the surface of the sample as measured by our coll.
B is the magnetic induction of the sample induced by the H field. The
annular H coil if formed from two individual windings, having identical
numbers of turns N , referred to as outer H winding (Ho) and inner
H winding (Hi) respectively, as shown in Figure 7. If these two coils
are connected in series opposition, the voltage induced in them is
given by d
!
H = e (XHo - XHi) (10)
Ho Hi dt
where 4t (XHo) and 4r (XHi) are the time rate of change of the flux linkages
of the Ho and Hi windings respectively. Since in general X=N#=NBA
and for a uniform field in air B=H, Equation (10) can be written as
d
e = H(Ao - Ai)N.H (ID
H " dt
where Ao and Ai are the cross-sectional areas of the Ho and Hi windings
respectively, and Nu are the number of turns each of the Ho and Hi
windings. It is seen from Equation (11) that the terminal voltage of the
annular H coil in Figure 7 is proportional to the change in flux linkages
through the annulus formed by the two windings, and can therefore be
used to measure the H field at the surface of the sample.

OUTER H-WINDING (Ho)

B-COIL
INNER H-WINDING (Hi)

SAMPLE

Figure 7. Arrangement of Sensing Coils for Measuring Surface B

and H Fields of Rod Samples

The B field is sensed by measuring the flux through the sample

with a coil referred to as B coil, which is closely wound to the sample
surface. The induced voltage in the B coil is given by

d
e
B = [NBASB + NB(AB - A ) H] (12)
dt

9.

'iTOiüfrrffli TmTTiff «fi fiji^-a*™«™'* ....,.-, ■.-„...:■■■ ,,^^^^^-^. ,,,V: ■: . M|[,

• •iiliMrtBÜÜlil'irii I
 ^painBimmmmMmmaj^mvm^lWiimmMWlV^itßmjtli-ivii»^** lamjij^wwiiii^iiug.jH^P^^^

rearranging
d
e„ = - ^-[NBAS(B-H)+NBABH] (13)

where Nß is the number of turns of the B coils, Ag is the cross-sectional

area of the sample, and Aß is the average cross-sectional area of a
turn of the B winding.

When the B and annular H coils are interconnected in series opposition,

we have from Equations (11) and (13)
e =
(B-H) " "dT [NBAs(B-H)+NBABH-NH(Ao - Ai)H] (14)

If the two coils are balanced such that

NBABH = NH(Ao - Ai)H (15)
or
N A = N A
B B H H (16)
where Ajj=(Ao-Ai) is the average annular area between a pair of turns
of the H coil, Equation (14) becomes

e
(B-H) ■ " TT B H
[NBAS< - >1 (»>

e =
(B-H) '-^tNBAS(4,rM)1 <18>
and finally after integration
e = _ N A (47rM) (19)
(B-H) B s
The requirement of equal area-turns of the annular H coil and B coil
expressed in Equation (16) is very difficult to realize initially in actual
coil winding, and it is therefore necessary to arrange the individual
windings in such a way that they can be compensated afterwards.

3. ELECTRONIC INTEGRATION

The integration of /erjdt and je. Mdt is performed by two inte-

grating operational amplifiers commonly referred to as Miller integrators.
The term, operational amplifier, is a generic term applied to amplifiers
whose gain functions are such as to enable them to perform certain
mathematical operations such as summation, differentiation, integration,
and others. This outstanding versatility of operational amplifiers is
obtained from the application of negative feedback. It is recalled from
circuit theory that negative feedback in general tends to improve gain
stability and linearity, to reduce output impedance, and in some con-
figurations to increase the input impedance. But the extent to which

10.

..v>_u,.^.^.^^L.^.L^.-^^/^N.^,:^^.-:i:.V^_.^.^^„^;.-._ i-, mm* --'—■-

 iLjji-.ippii.!j»m*'l-'1 w.v&mmm**it* ~xi'■■ ■ **■?*■'-(v"*-■ *-.m*?!miM>tywm.*r>-mwmj^mß^M^iJMwmj.*>.<.*■ A

^W^B8Sä<

closed loop performance is improved depends on the magnitude of loop

gain (Reference 8). Another useful property of negative feedback which
is the basis of all operational amplifier technology is that with enough
gain, the closed loop anplifier characteristics become a function of only
the feedback components. This can be shown by analyzing the general
operational amplifier shown in Figure 8. Applying Kirchhoff1 s current
law to the summing point, yields
= i
*1 i ' H (20)
or
(ej-e« )Yl = e€Yi + (e€ - eQ)Y£ (21)

rewrite and solve for eQ:

ee(Y1 + Yt + Yf)
e = (22)
o

however, the nominal gain A is defined as

A = (23)

rSUMMING POINT

AMPLIFIER
GAIN A

Figure 8„ General Operational Amplifier

and the resultant gain with feedback is

e
o
A,f = e (24)
1
Using these relations in Equation (21) results in
(Y + Y + Y
1 i f> * 1 (25)
A
f =
or
<Y1 + Y
i + Y
f> A
f
(26)

11.

iiiÄHii*ifciifci^^^-^ --*^^ Ä «5i« •MH^NttlttlMI

 wfN»r«m»gwmM|»mV "" «.w;».««^'11-" ■' "

solving for Af . v + Y + V Y
i 1 £ , 1
A [1 ( * )J = i (27)
A Yf Yf

or Y
A -- ' '
Y
t 1 _ j_ , i + Y. * Y Y
f)
<28)

A Y
' £
finally: Y
1
A
f=
Y 1 1 Y +Y (29)
1 (_i 1_)
A A Yf
or equivalently:
Z,
A£=-_L
Zj 1 Z + Z 1 (30)
1 (Zf —i L_) _
A ~ Z1ZJ A
Since Zj is always large, Z^lO^ß in some designs, we may assume
infinite input impedance, then

Af=.A i
Zj 1 Zf 1 (31)

A Z A

so that Equation (31) can be considered an exact expression for the

closed loop gain of the general operational amplifier circuit shown in
Figure 8. Some important results can be obtained from this expression
as will be developed subsequently.

To begin with, let A—00 in Equation (31), which then reduces to

Z
f
Af « (32)
Z
l
This means that as long as ee is negligibly small (which is implied by
A—GO), the gain is solely dependent on the ratio of the impedances
Zf/Zj. If Equations (24) and (32) are combined, the resulting expression
can be written in the form
e = [ZfYjJej (33)
o

12.

^jMamaiäaiflfflüfi jfcffl ü ■-•■*&&***>•&**

 ag^w^^iy «f ^g^^art^a^^w^m^Tg^l^'^^^-^^^^

Bf
^^.'HMfllMlMMwH,

This expression is very important, as it contains the explanation of the

name "operational amplifier" which has been applied to the circuit of
Figure 8.

So far in the development leading to Equation (33) no conditions

had been imposed as to the nature of the potentials. Consequently the
impedance functions that appear in the equations are functions of the
general differential operator p(=d/dt). As a matter of fact, the foregoing
development has actually been a manipulation of the controlling differential
equation relating the output potential to the input potential, and Equation (33)
is the approximate form of this controlling differential equation. If in
the special case the applied potential is sinusoidal, then p(=d/dt) is re-
placed by jw; Z^ and Y are functions of jw and e and e. become E_
and Ej, respectively, where Ej and EQ are the complex amplitudes of
the input and output potentials (Reference 9). In this case the steady
state relations of Equation (33) is the
E
o = Z Y E
f l l (34)

Equation (32) permits a very convenient approximate circuit to be

drawn of the operational amplifier. This approximate circuit is given
in Figure 9. The use of the expression "virtual ground" emphasizes
the fact that the presence of feedback forces the voltage ee to zero.
This is symbolized by drawing a short from the summing point to
ground; no current actually flows through this short, however.

h 1 1
h f\

+ —■♦'l *if •i

e
.

iN=0
1 e
c

1 1
\J
n

Figure 9. Approximate Equivalent Circuit of Operational Amplifier

If the impedance elements in the circuit of Figure 9 are selected
as shown in Figure 10, then ZJSRJ and Z, is the operational expression
of Z£=l/C,p, where p(=d/dt) is the time-derivative operator and (1/p)
denotes integration with respect to time.

An application of Equation (32) requires that

1
e = - (35)
o
VfP
13.

I I "I- Ill i-'" -- --'-■ ^

|||pl!|IIWpW!i|^IW,UI^UiiHWt^ ■PHQMJPpnpp jUIMtuM». i' ipupjp I : I |" 1PB

which is
1
e
o = / r* e dt (36)
RlCf

This shows that the output potential is related to the integral of the
input potential.

Cf
O JWMr

e,
S
I
'N 0

Figure 10. Approximate Equivalent Circuit of Integrating Opera-

tional Amplifier
If we substitute Z, = R. and Zf = 1/C p into Equation (31) we obtain
f
1 i
V RlCfp (37)
1 1
1 - (1 + )
RlcfP
or
A
f = (38)
1 + (1 - A) RjCfP
Equation (38) becomes then an exact expression for the closed loop
gain of the integrating operational amplifier shown in Figure 11.
R, Cf
0 /V*AM/— -* 1 (-

Figure 11. Integrating Operational Amplifier Circuit

14.

i.... _*.**.—H^Kigi mmtM ■ a ^...„., a, mm, M - -—.. manigjinK„„.^„.w.^-,-r aaaaüM^a n'iiiii'fWrrt

 mn!**'-w' ''ft <m&fQHn
^^-T»,.--^*-*.^*^-*"" - ■,1pW«mnw'!Bra!!sws^^

Additional insight of operational amplifier characteristics is obtained

by considering the current in the feedback element i, in Figure 11.
From inspection of this circuit, i, is given by

<e< - e
o>
e«(l - A)
\ = (39)

where
z' =

Obviously, nothing would be changed at the input of the amplifier if the

impedance Z, were shunted between the summing point and ground,
since the same currents would still flow; this is shown in Figure 12.

z,

z'f = l-A

Figure 12. Equivalent Circuit of General Operational Amplifier

The only requirement is that A is known. If we substitute Z =R and

Z,= l/Cp as was done previously, another equivalent circuit of tire
integrating operational amplifier is obtained as shown in Figure 13.
This particular equivalent circuit is extremely useful for calculating
the integrator performance. The fact that a capacitor C, located
between the input and output terminals of an amplifier has the same
effect on the amplifier as a capacitor (l-A)C shunted between input and
ground, is known as the Miller effect (Reference 10), It is for this
reason that the integrating operational amplifier shown in Figure 11 is
often referred to as Miller integrator. The closed loop gain of the
Miller integrator shown in Figure 13 is found by recognizing that
1
"l Cfp(l - A)
e« = j (40)

R +
l
Cfp(l - A)

15.

■■■• - ~-~~r-" ■■■■—-" ■ :?--■..■-..■■ rrir"J-w-~

- ---'--'-• --- -- J^^M«AaM«riMMr-rtMmi)lll1filt[rfri.rr
 or ec = (41)
1 + (1 - A) R^p

Ae,
and
e =Ae« = (42)
1 + (1 - A) RjCfp

and finally

f c
(43)
l 1 + (1 - A) R^p

which is the same result as that obtained from an analysis of Figure 11;
hence circuits shown in Figures 11 and 13 are equivalent.

R.

ß
B
+
Cf(l-A)1r e
♦ [\ A^b— f\
l * \r +
e
\s °_
o— < I— I 0

Figure 13. Equivalent Circuit of Miller Integrator

An interesting comparison can now be made between the integrating

circuit shown in Figure 14 (Reference 11) and the Miller integrator
shown in Figure 13. It is apparent that the input circuit of the equivalent
Miller integrator is of the same form as the passive integrator of
Figure 14, hence the Miller integrator can be treated as if it were a

R
-VWvV-
|+ e, i + RCp
e. Czz e,
i

o- k e,
WHERE T = RC*TIME CONSTANT
l + Tp

(a) (b)
Figure 14. Passive Integrating Circuit

16.

riiiiriiiriliiiiiiiiiWiiiiiiiiWif'' riBimrir-^^-"-""- A(I ni T -ii-rriMiiiitwüiiir '"" *""*""** mi*m

yjff aiittfitlfttiMiiiMiii^^
 passive integrator. It is also seen that the use of a feedback capacitor,
that is the use of a negative feedback, around an operational amplifier
has effectively increased the time constant by a factor (1-A) and the
. same time multiplied the output of the integrating network by the factor A.

4. ANALYSIS OF INTEGRATOR ERROR

If a step input voltage E. is applied to the passive integrator of
Figure 14, the output response with no initial charge on the capacitor
is given by
-t
"RX (44)
eo = El
•V(l- e )

Since the equivalent Miller integrator of Figure 13 is of the same

form, we have analogously
-t
(45)
e0 =AEl(l-e (1 ' A)RIC* ,
or
t
T (46)
eQ = AEj(l - e )

where T=(1-A)R C, is the effective time constant of the equivalent Miller

circuit. After expansion of the exponential Equation (46) becomes
t .2
= A E. [1 ] (47)

It is seen from Equation (47) that the integrator output voltage begins to
change linearly with time at a rate equal to AE,/T , which corresponds
to correct integration of the input step. With increasing integrating time
the output voltage deviates more and more from the ramp function given
by the first term of Equation (47) and tends to approach AE. as a limit

CORRECT
INTEGRATION

i£«tr* 15. Integration of a Step Function

17.

:.- ä^ädämaämmammeitimaiaää&
:.^^:i„^:^:*i:«^~^ iaa i&mmfrZnammüiia jemmm-**^^^"*'^" laüttaiai
?!

as shown in Figure 15. It should be noted that the amplifier saturates

long before any appreciable deviation from the initial slope occurs» see
Section IV. The resulting integrator error is given by the second term
of Equation (47). The absolute value of this error is
2
(48)
j«|<AEl—,
2T
and the percentage error
AEjt2

c(%)< 22 2 x 10
100 (49)
AEjt
T

«(%)<„ *t x
x 10
100
° << 50 tt
50 (50)

So if the integrated output voltage is required to be within «% of the

true integral AE.t, the integration time must be limited to

t<^j- (51)

Equation (48) does not account for any errors as a result of stray
pickups, drift, and noise which may be introduced at the summing
junction or generated within the amplifier.

18.

-......,-.,....aaaaMttMttMaMiaiaaaaia a .*...;...-.^~<.-^.^^"-*^^^^^ i ■ • ■■--•■-^- -- ■■-■■ .---.•■.^.t^^-Tr--n.rt'<w»Tt-ariürriM

n

SECTION IV

DESIGN AND CONSTRUCTION OF THE HYSTERESIGRAPH

1. RECORDER AND AMPLIFIERS

The material presented in Section III forms the basis of the actual
design and construction of the instrument. Since the operational
amplifiers and the X-Y recorder are purchased parts, the actual design
consists basically of determining the integrating networks and the
calculation, winding and balancing of the sensing coils. Following this
the system can be assembled and its performance checked and com-
pared with the specifications of this application. The X-Y recorder,
a Moseley Model 7000 A "Autograf", was already available, and no
further consideration is needed. The amplifiers were selected on the
basis of large open loop gain, low noise, and low offset. Among the
large variety of available types, two Philbrick Model SP 656, chopper
stabilized amplifiers having an open loop gain of |A| = 5x10? were
selected for the integrators.

2. DETERMINATION OF INTEGRATING NETWORKS

The noise generated within the amplifier is one of the parameters

determining the lower limit of the input signals which may be integrated
and amplified. For the amplifiers selected, the noise referred to the
input is lOfjL Volts peak-to-peak. The minimum input signal which could
be tolerated was somewhat arbitrarily set at 100/x. Volts.

Assuming a step input to the integrator of 100/x. Volts, we have

from Equation (47) for correct integration
A
e = e
o l^2 " V
(1 - A)RC

If we let A/(1-A)RC = 100 and if the total integration time were 10

seconds, then
-3
e = 100 x 0. 1 x 10 x 10 = 0. 1 Volts
o

Since the recorder has 0. 5m Volt/inch deflection on the most sensitive

range, adequate deflection is assured under these conditions.

An effective gain of 100 can easily be realized by letting R=10K

and C=l/if, then
A 100A
A
eff
(1-A) 104xl0-6 (1 -A)

19.

a fa^a^-^iato
i^^l^*^^«*^«:*^- W',. ..w^:--^^:.LWL^ieia- ..«Siiäifc -^:^A_^ >^i' ^t^^Ä» rri-tfi'ir-*ü1»iii-iii.JMiii--":i'-iTi-
fci^:^..:^*::«^,^^^:.- , ■*■: „.^„^-i-'
 and since A=5xl0 and A is negative, then A/1-A&-1 and A ..»-100.

The flux linkage sensitivity is determined from a consideration of

the magnetic characteristics of the sample and the characteristics of
the electromagnet in which the samples are tested. From Faraday's
law of induction J .
9 -8
e = - N x 10 Volts (2)
dt
Since dp/dt depends on how fast the applied field is changed from
H J+) to H J[-). A minimum time of 10 seconds was considered
a realistic value, consistent with the inductance-limited reversal time
of the electromagnet and the response of the recorder, then from
Equation (2)
-8
eAt = - NA£ x 10" Volt-sec
or _g
edt = - Nd?X10 Volt-sec

Using the minimum voltage of lOOfi Volts as determined before, the

equivalent flux linkage sensitivity is then
0. 1 x 10 x 10 = 1 m Volt-sec
Assuming a coil of 500 turns, the equivalent flux change which is
measurable with sufficient accuracy is then
1 x 10"3 _6
A/£ = = 2jtVolt-sec = 2 x 10" Weber
500
From the specifications of Section I, the minimum saturation value
for the magnetic portion of the sample is approximately 4wM =B =
6000 Gauss, and the typical packing fraction is 5%. The latter can
be taken into account for the purpose of this calculation by assuming
that we deal with a sample having the full saturation value of the massive
magnetic material, but a cross section of only 5% of the actual composite
area, hence

A ' = x A = —— x (1. 1cm)2 = 0.048cm2

3 100 S
100 4
The peak-to-peak flux change is then

Ap= z4>oS = 2A 5 ' x B S = 2 x 0. 048 x 6 x 103 = 575 Maxwell = 5. 75 x

-6
10 Weber

Comparing this value with the theoretical value obtained from the
amplifier characteristics, it is seen that a flux linkage sensitivity of
lm Volt-sec is more than adequate to measure 4wM of low induction
materials.
20.

^^^^^^^.„^^^
 ■ <*-

To measure peak-to-peak values of MHC*100 Oe at applied fields

=15, 000 Oe, assume an area for the annular H coil of 1. 8 cm2,
of Hmax
then
A> = 2<t> =2x1.8x15x10 =54x 103 Maxwell
max

or
= 5.4x 10'' Weber
Since the flux change as seen by the H channel is about 100 times that
obtained c he 4irM channel for the dilute samples, an effective gain of
10 was considered adequate. The corresponding values of R and C
were found to be R=100 Kfland C=lfä.

3. INTEGRATOR PERFORMANCE

Each integrating amplifier was calibrated by applying a known

step input voltage to the amplifier input. The amplifier output was
connected to the Y-axis of the recorder. The X-channel of the recorder
was operating in the sweep mode, which was set at 100 sec /inch. A
step input E. (+)=0. 59m Volts was applied while the amplifier was
saturated atv t(-). With the recorder turned on, the integration
time was measured from the time the amplifier became unsaturated.
After the amplifier reached V .(+)E was reversed until the amplifier
reached V .(-) again. One such cycle is shown on the graph of Figure 16.
On the up-scale integration, the integration time was found to be
t = 432 sec
the calculated output for correct integration

E O = AE.t
in
= (100) (0.59 x 10~3) (432) = 25.48 Volts

measured output from Figure 16

E„o = 25.5 Volts

On the down-scale integr«.4 on the integration time was found to be

t = 427 sec

and the calculated output for correct integration

Eo = (100) (0.59 x 10"3) (427) = 25.20 Volts

the measured output from Figure 16

E = 25. 5 Volts
o
While on the up-scale integration the integrating error is <0.01%, on
the down-scale integration this error is approximately 1%.

21.

aHrtMüatttüttai wääim .^^a^ j>a* »«^»^ .... -»i rtiTur.iiiHi - -

^^ü^h^iuiiiir^^ ■*******&&&&&"
 22.

;
" riiMfrftfctfW"-"^"--^'^1^
i^müikttilinriiMr-irr
±»z^<^,xM4äÜ. iaata^^r^^^;^ 1
 iH
♦»
u
o

*
ft

CO

i O
tt
u
00

• '43
« .3
a n)
o h
(X bo
* 'S

DO
•H

(A W CO

J
23.

^t.-.-rJÜ;;.-,^«^*,.*.
■■..,.-, :.,..-.....^ .-.M-.^-... -... . ■-.-, .:,....,...... .,' ■..- £UBiHH^Bii£i iii iIIMdiiii*^ffiamfaitattttiWiM
 This performance was considered quite satisfactory since it takes
approximately 1 minute for plotting one complete 4irM versus H loop,
after which the integrator is reset. Figure 17 shows the response of
one of the integrators to the same step input voltage but with different
sweep speeds. Again the observed integrating error is negligibly small.

According to Equation (51) the theoretical integration time should

be equal to or less than 1000 sec for an integration error of 0. 1%. The
deviation from this theoretical result must be attributed to drifts and
noise in the amplifier.

4. COIL CONSTRUCTION

In Section III we saw that the criteria for obtaining a voltage pro-
portional to 4TTM from the annular H coil and the B coil was expressed
by the relation
N A
B B " VH
where AH=(A0-Aj) is the average area of the annular H coil. This
equation is difficult to satisfy because of rather large and unpredictable
discrepancies which exist between the calculated area-turns of a coil
and the area-turns obtained after the coil has been wound. It is there-
fore necessary to arrange the individual windings which make the
annular H coil and B coil in such a way that the area-turns of one coil
can be precisely adjusted to satisfy Equation (52). Since Equation (52)
requires also that the number of turns of the HQ and rL windings are
equal, the annular H coil was chosen to be the coil whose individual
turns as well as area-turns could be adjusted after completion of winding.

A winding arrangement which could serve these purposes is shown

schematically in Figure 18. In this schematic the H. winding is wound
on the bobbin core, followed by the B and H0 windings respectively. The
bottom layer of the H. winding and the top layer of the HQ winding are
provided with additional turns which serve as the pull turns for compen-
sation purposes.

The physical dimensions of the bobbin were determined from the

sample size, the number of turns and layers for a predetermined winding
length and are shown in Figure 18. In order to obtain adequate signal
output from the coils, 500 turns for each of the three windings were
chosen on the basis of the calculations in Section IV. The free winding
length was chosen to be 1/3 of the sample length. This was the best
compromise between the desired number of turns and a manageable
wire size.

The Hj and H0 windings were wound with #42 AWG Formvar wire
and the B winding with #38 AWG Formvar. The following winding data

24.
 fH&fam&mmmm

^PROTECTIVE RING

REMOVABLE FLANGE

SPACER

■5.0 *« 5.0-
■3.5-

23 11.8
1
.8 12.6 30 21.3

itr **arM"A u
ALL DIMENSIONS
7.5 * IN MILLIMETERS
10- NOT TO SCALE

Figure 18. (a) Cross Section of Coil Arrangement

(b) Coil Form

25.

jmfjjg^jgg^HillUgjjgj jj w;..^|-:t, lamgi—aMMjafajfaamateiaa .^.'--vi»«u^«a«iM«mvii«jriiMi-'jiM«ibii nnnw».it,<ni*mwwrimiiM amBMBftaua ^^a&iaiuaMi

p, ^wwi,« i!?W7^^r?«!'^F^y^rr^;-^.-r.'^'

were obtained while actually winding the coils according to Figure 18:

Inner H winding PL:

total number of turns NTT. = 480

inside diameter D = 13. 0 mm

i

outside diameter D = 14. 0 mm

o

average diameter D = 13. 5 mm

B winding:

total number of turns N = 470

B
inside diameter D
i = 14. 0 mm

outside diameter D = 17. 5 mm

o

average diameter Dx, _ 15. 8 mm

Outer H winding H o*

total number of turns Nu = 480

inside diameter D^ = 20. 1 mm

outside diameter D = 21. 1 mm

average diameter D = 20.6 mm

The effective area of the B winding is then

Aß = —2— Dß = .785 x 1.582 = 1.97cm2

5. COIL COMPENSATION

After the coil was wound the detachable flange was removed from
the coil form and the pull ends of the individual H windings were joined
together to form the annular H coil. The coil was then checked for
unbalance at the center of a long solenoid and subjected to a 420 Hz
ac-field having H = 67. 5 Oe. The initial unbalance between the
max
annular H coil and the B coil was found by measuring eR and eH individ-
ually and eg_pj directly. The RMS values of these voltages were
EB = 1. 18 Volts, EH=1.21 Wits and E(B_H)=28. 2 m Volts. Since EH

26.

imiirT»-! Yi i mm^^m | iiuiinMii-nffiiw-*-*" ---"— -- aMünü i -■■-■• tm^mmämiiatfMim>miü»iimm |jg^j^|MMHMi

 -- "-;:r.^i:«-!v --•■;;. ■■■ ■ :*T?"fl
;.-:.,-.". ;.--■■--.= -;iT-i-.— (•-.-,-rr- -r--:::,^--^ ^■■■;'-:.'-,;-.--ä=^- r--'"i- ^»Jirf.-Vr,,».,, W*^J^*?W ' ' ' '
mf LfflUMfffCTrTt'^^■-■T*>~w*^

fc«8»Sfto »^»^^»»B^.'a^jt^w.wt."^™^«---^.^

exceeds Eß by 28. 2 m Volts or 2. 3% the excess number of H turns

was found as
2.3x480
excess NTT = 11.0 turns
100
After pulling 10 annular turns, E/jj_m was measured again, giving
E/g_jx\ -" 0. 68 m Volts. Because of obviously present stray-pickup
voltages no further improvement was expected by this method. Instead
the coil was placed in the electromagnet and the compensation was
checked in a slow changing dc-field using the hysteresigraph. Without
a magnetic sample in the coil, E/g_pjv = E^v* = 0 and for a perfectly
balanced coil, the X-Y recorder should draw a straight line while the
applied field is changing. By setting the X-axis attenuator (H-channel,
A ££=10) to 0.2 V/inch and the Y-axis attenuator (4TrM-channel A ££=100)
to 0.02 V/inch, the 47rM-channel is then 100 times as sensitive as the
H-channel. Under these conditions any remaining unbalance of equal
area-turns is precisely readable on the X-Y plot. The result is as
shown in Figure 19.

Y n

UNDER
♦ COMPENSATED

i1OVER
COMPENSATED

Figure 19. Remaining Coil Imbalance

The error in area turns as obtained from the hysteresigraph is then

0.2:
« 0.02%
*NA"
12.5 x 100
With Ng=470 turns this apparent error amounts to

8NA = 0. 0002 x 470 « 0. 1 annular turn

This compensation for equal area turns of the B and H coils was
considered satisfactory and no further improvements were attempted.

The area of the B coil can now be determined by placing the coil
again in the same ac-field of the solenoid as before. The voltage induced

27.

- ■ ■'l~~---'
n ^i n.r^-tth*tf"J-i^-
r. fan ii.r': ■— r i HI urn.
 in the B coil by an ac-field is given by

g
E = 4. 44 x f x NR
D x 4> x 10 Volts x(53)
max '

Since no iron is involved here, <f> =AgxH and substituting numbers

in Equation (53), we have

1. 18 = 4. 44 x 420 x 470 x Aß x 67. 5 x 10'8

if tiiis expression is solved for AR the result is

1.18
An =' = 1.99crn
B
.592
By comparing the value for the area of the B coil obtained from the
physical coil dimensions with fcuat obtained from the ac measurement,
it is seen that they are in good agreement. Since the annular H coil
and the B coil are balanced with an error of approximately 0. 02% the
area of the H coil is then, A =1. 99 cm^ also.
ri
To protect the coil from damage in everyday use, the coil was
inserted into a holding fixture and sealed on both ends with epoxy, see
Figure 21.

The entire system was then wired according to the schematic shown
in Figure 20.

28.

-■■■■■- ■ ■-• ■ • • *—■ —■- ..,--■--. ..-....■■ —— .lllM -,,, - - niM^^M^an

T\ *mmmmmmm>«im

-a
(II
u
00
•rl
»
01
<U
■w
a>

K
O

u
CO

o
<M
V
a
bo

29.

■k .'^uJta- -iBrf'iJi- itfi -r-Li¥f—1n iM\tiätoAimtfMHätMäimikT*titMGtm na 'i ■ rti ¥ ■ inrinMiiai^Birfarttü-itirinaii

 '•-..

Im
v
H
u
o
«H

4»
co

■a
h
00
•i-i
09
V
V
*>
W
>.

S
0)
Ö
n)
2
M

bo

30

■ i ■- - - — - —- ■■ - i
,mM«iia^-„»„.,v,.. ■ili-itirTi-fl-'rfiW"""^*^*-" rii«
 SECTION V

CALIBRATION OF THE HYSTERESIGRAPH

1. CALIBRATING PROCEDURE

The calibrating procedure of the completely interconnected hys-

teresigraph consists of separately calibrating the H and (B-H) coils.
The calibration of the H coil was derived from the known applied field
between the pole caps of the electromagnet. The calibration of the
(B-H) coil was derived from the known saturation magnetization 4wM8
of a standard test sample of pure iron.

2. CALIBRATING THE H COIL

A Rawson rotating coil Gaussmeter was used to meaturf. the applied

field between the pole caps of the electromagnet which were separated
by a distance equal to the coil length. The electromagnet was then
excited and cycled several times through a maximum field of either
polarity. The maximum field excursions in both the positive and
negative directions were recorded and averaged out to get the peak-to-
peak applied field. Without changing the pole separations nor the exciting
magnet current, the Gaussmeter probe was now replaced by the coil
assembly without a test sample and the magnet was again cycled. Since
it is under these conditions 4TTM=0, the hysteresigraph will then draw
a straight line along the X-axis, which is now a measure of the peak-
to-peak applied field.

From the Rawson Gaussmeter the measured peak-to-peak field

was
H
= 33, 100 Oe
P-P
With the coil in the magnet and the recorder sensitivity set at X=0. 2
V/inch a line 14.67" length was obtained as shown on Figure 22. The
calibration constant for the H field is now

33, 100 Oe
C
H = 11,280 I 1
14.67 x0.2 VoltJ
This completes the calibration of the H coil.

3. CALIBRATING THE (B-H) COIL

The (B-H) coil was calibrated by recording hysteresis loops of a

pure iron standard of accurately known dimensions and saturation
magnetization value. The number used for the latter is 4i»Mg=21, 510
Gauss. It was determined at the Air Force Materials Laboratory on

31.

'^""■''-"■miiiiiii - ■■ -mriinUiiii ..... ,.,... .,...,,-.,-^,:^.^„..... aw.,.a^aaagggjjgg^iij^gil^lMia

 00

►
u
3
ü
Ö
o
•H
■M
flj
h
43
■-•
(4
U
0

(M
M
4)
U
a

32.

it^iiiMiiitii^»j^^:^"'«:i<«ifiiitiiifW'»rj. ,-,-' •'- ■' ■ ■ .-..V-J-^^-^^MJ^--'"^'^^-^-^^™*^' BmTilinyrfnÜiilfÜliM — "■■" i.r in« i»r t i.-nriMMtfclirtttimift
 iron from the same piece, using a pull coil technique and fields up to
50 kOe (Reference 12). This number is considered more accurate
than the often used literature value of 4wM =21, 580 (Reference 13).
Placing this sample into the coil and cycling the applied field as de-
scribed above resulted in the 4wM versus H plot shown in Figure 22.
The Y- sensitivity of the recorder was set at 2 Volts/inch and the
measured distance between the lines representing the peak-to-peak
magnetization (2x4irMg) as found from Figure 19, was dM=7. 9 inches.
The calibration constant for the 4irM - channel can now be determined as

2x21,510 [uauss-i
r Gauss-
<• = 2720
i»M
7.9x2 Volt J
If a sample under test has a cross-sectional area A different from
the cross-sectional area Ag of the iron standard, the new magnetization
constant is then

'4TTM
Gauss -i
[uauss

'4irM A', Volt J

where As is the cross-sectional area of the standard.

For the test samples used here, A =1.00 cm' hence,

2720 2720 [uauss

r Gauss -j

'4TTM
= 2400
1.000 1. 132 Volt J
0.882
Thi i completes the calibration of the (B-H) coil.

4. OVERALL ACCURACY

The accuracy of the instrument is determined by the accuracy of

the quantities which were used for its calibration. The accuracy of
4wM of the high purity iron sample is estimated at a few tenths of 1%.
The accuracy of H is determined by the _+ 1% accuracy of the Rawson
rotating coil Gaussmeter. The overall accuracy oi the entire hysteresi-
graph is then estimated at about 1%.

33.

*""»"'•-^ ■■■■■■
.„amamm.,..,... ,.. .„.„„;.. ,,,j -.... j,.,::,,^,„..^ . . . .. „..,,., .,^-,^.,^ .■„,,,.... ....- ..... r.^..**j. i laitBüiiafrtii
 SECTION VI

OPERATION OF THE HYSTERESIGRAPH

The only preliminary work prior to drawing 4wM versus H loops

consists of eliminating the drift by means of the offset adjustment
provided for each amplifier. An initial warm-up period of about 20
minutes is required to stabilize the amplifier outputs before these
adjustments can be made. Thereafter only occasional readjustments
are necessary. Following this procedure and with the coil and sample
positioned between the pole caps, the magnet can now be excited. Before
bringing the recorder pen into writing position, the applied f* '* is
cycled several times to allow adjustment of the proper 4iriw T
scales on the recorder attenuator. Upon completion of a £«. ial
cycles, the integrating capacitors are discharged and the pe-i is
brought into writing position, and the 4wM versus H loops can be
recorded.

Once a complete loop is drawn the coordinate axis can be estab-

lished and 47TM =B and wHc can be evaluated using the calibration
constants given in Section V. Examples of drawn 4wM versus H loops
are shown in Figures 23 through 25.

34.

•_..._..._ — ■■
35.

jiiyjj^l^^^^Ä
 O
u

M
<
u
•H

'S
a
o
o

aa>

fM

36.

Ma jma aaüü »^^ a ■ - -.-.■-.-■^■.. ■ ....-—. mmmmmum ■■■■-"- ^^-^-^.C^..,....

 r
W
Ifl
o
Ü
I
t-l

EH

0
O

to
CO
u
>

4?
m
<M
V
U
00

37.

^...iiaW*-.^-.^ ääj^»äilu^ ^^jA^Ute^r-i'i^k' üä»

«1

SECTION VII

SUMMARY

1. SIGNIFICANCE OF THIS CONTRIBUTION

The hysteresigraph described in this paper was designed to plot

4nM versus H loops of dilute fine particle permanent magnets. As
such, this instrument has made valuable contributions toward the eval-
uation of magnetic properties of materials.

The fact that specialized instruments of this nature are not avail-
able on the commercial market places the burden on the individual
investigator to design and build his own test equipment. To assist
other investigators in this field, this paper has been presented as a
case study of a hysteresigraph design. The novel feature of this par-
ticular instrument is the coil arrangement which allows the measurement
of the B - and H fields at the surface of cylindrical samples. As a
result the instrument described here is not limited to drawing 4TTM
versus H loops, but can also be adapted to plot B versus H loops of
cylindrical samples. Among all hysteresigraphs which have come to
the attention of the writer, only the one reported by Cioffi (Reference 1)
can be used to plot B versus H loops of rod samples. However, the
technique used there is quite different and much more involved than the
one used here. In addition, the instrument described here uses only
"off the shelf" items, with the exception of the coil. The necessary
flexibility for designing an instrument for a particular application is
thus preserved.

The main objective of this particular application was the evaluation

of j^Hc of large numbers of fine particle permanent magnets. This
objective has been accomplished quite satisfactorily, since over 200
samples have been tested thus far without difficulties.

2. RECOMMENDATIONS FOR FURTHER STUDY

Experience with this instrument has shown that further studies
are needed toward the improvement of long-time electronic integration.
As was shown in Section IV, accurate integration is possible over time
periods of 7-10 minutes. This performance was more than adequate
for this application. However, more and more magnetic testing of hard
magnetic materials is being done in superconducting solenoids which
require up to 30 minutes or more to finish one complete cycle. For
this reason hysteresigraphs of the type discussed here or in the literature
cannot be used with superconducting solenoids. In this latter application,
the sample sizes can be only a small fraction of the ones used at the
Air Force Materials Laboratory, thus necessitating much smaller

38.
 sensing coils. Additional studies are therefore required regarding the
sensitivity requirements.

3. FUTURE USE

The development of the instrument described here has closed an

existing gap in the available types of hysteresigraphs. It is hoped
that this instrument can serve as a basis for further developments
Itfiding to an extension of its use to magnetic testing in superconducting
rolenoids.

39.

^ ii ,'■,-,; r:,j ■,t'i lfflMiiiifaliflii'iii^ifiTir'i tifin ••-—■---MijfjM •-•--"-' "'■ läUTain.. itMUSüm' fiäifii ■ • i i - '-'i:- &Ei ■ i ■ r-NiriifiiMlilii^^"-^-'
 BIBLIOGRAPHY

1. P. P. Cioffi, "A Recroding Fluxineter of High Accuracy and

Sensitivity11, Review of Scientific Instruments, vol. 21,
number 7, pp. 624 - 628, July, 1950.

2. P. F. Elarde, "All Electronic Magnetic Hysteresigraphs",

The Western Electric Engineer, pp. 8-14, January, 1965.

3. R. R. Bockemuehl and W. E. Sargeant, "Practical Hysteresi-

graphs", Journal of Applied Physics, supplement to vol. 31,
number 5, pp. 180 S-182 S, May, 1950.

4. P. Mazzetti and P. Soardo, "Electronic Hysteresigraph Holds

dB/dt Constant", Review of Scientific Instruments, vol. 37,
number 5, pp. 548-552, May, 1966.

5. K.J. Strnat and G.I. Hoffer, YCo_, A Promising New Mag-

netic Material, AF Technical Report AFML-TR-65-446,
May, 1966.

6. E. W. Golding, Electrical Measurements and Measuring

Instruments, Sir Isaac Pitman Publishing Co., London, 1941.

7. E.M. PughandE.W. Pugh, Principles of Electricity and

Magnetism, Addison-Wesley Publishing Co., Reading, Mass.,
1962.

8. J. Millmann, Vacuum Tube and Semiconductor Electronics,

McGraw-Hill, New York, 1958.

9. J. L. Stewart, Circuit Theory and Design, Wiley and Sons,

New York, 1956.

10. J. M. Miller, "Dependence of the Input Impedance of a Three-

electrode Vacuum Tube Upon the Load in the Plate Circuit",
National Bureau of Standards Scientific Papers, vol. 15,
number 351, pp. 367-385, 1919.

11. M. E. VanValkenburg, Network Analysis, Prentice Hall,

Englewood Cliffs, New Jersey, 1959.

12. J. L. Mouton, Magnetic Properties of Rare Earth-Iron Alloys,

Thesis, Air Force Institute of Technology, August, 1964.

13. . R. M. Bozorth, Ferromagnetism, D VanNostrand, Inc.,

New York, 1955.

40.

iriirlii-fr-hM--"1-""'-^-' "
■tx&imäiuiätmmttmmm