The Strength of Filler Joist Floors.

THE STRENGTH OF FILLER JOIST FLOORS.-By Ewart S Andrews, B.Sc. En@, M.I.Struct. E., A.M.1nst. C.E .
structural have for a M ANYcasing steelengineers in concrete longtimebeenconvincedthatthe effect of beams in manner the which obtained has for many years with the filler-joist floor, has a
&'A QC.
A.
D.
G.
L.
inch increase strength (an in of 33 per cent.), up the but to present there has beenbutlittlepublished of experimental tests upon which we could base a demand to have the specifications and bye-laws revised so as to enable us to design the steelwork for such floors more economically. In many cases thefiller-joist type of floor has considerable advantages; it is quick in construction and avoids the trouble attendant supporting upon centering from below ; moreover,theordinarycomparativelysmallbuilder,whoperhapshasnot on his staff any men who have had sufficient knowledge and experience for ordinary reinforced concrete can work, undertake t h i s kind of floor satisfactorily. I t is, therefore,withgreatinterestthat we have studied the report just published byMessrs.Redpath,Brown & Co., Ltd., on the " Strength of Steel Joists Embedded in a Solid Concrete Floor Slab." is prepared by M r 2 Thereport,which J . R. Sharman, M .Inst .C .E ., describes the uponmade resultsteststhe at National Physical Laboratory upon a machine was whichdesigned by Mr. Sharman, in consultation Mr. with J. Mitchell Moncrieff, C.B.E., hl.I~:st.C.E., M .I .Struct.E ., todealwithtestsupon horizontal and vertical specimens, and was Messrs. presented Redpath, by Brown OL Co., Ltd., to the National Physical Laboratory.
SCOPE AND METHOD O F TEST.
K.
strengthening effect upon the floor and that it is safe todesignthesteelworkinsuch floors upon higher stresses than the extreme fibre stressof 7.5 tons per square inch which is usually specified. Some designers havefollowed the rule of taking the stress up to I O tons per square
Two series of tests were madeI . W i t h 4 x 19 R..S.J. 9 ft. span. 2 . W i t h 6 x 3 R.S.J. 14 ft. span. Frames were made with two tie-bolts in the length and weretestedplain ; similar frames filled with concrete flush on the bottom sides and of the various depths shown in Fig. I were also tested. A similar frame with concrete flush top and bottom was also tested, but the -report states that this showed little gain strength of
THE STRUCTURAL ENGINEER.

over the plain joists, and is not included in ; the table accompanying the reportwe refer to this in greater detail later. L 33 .
39
T h e calculated stress in the plain joists at the test-load (practically the elastic-limit load) was tons 13.8 per sq. in. the for 4 X 19 beams and 12.1 tons per sq. in. for' the 6 x 3 beams, the ultimate strength of the material in the two cases being 29 and 28 tons per sq. inch respectively. These comparatively low results seem to A The method of testing was as follows :of the compression show buckling that T h e load was applied over the whole width flange was the causeof failure, as it usually of the slabs, either at the centre or at two is in beams without lateral support; the points and diagrams were made of the deratio of span to breadth flange for the two of flections plotted against loads; the the cases was 62 a n d 56 respectively. diagrams indicated clearly the limit of proExtensometer giving actual tests, the portionalitywherethestraightline comof the R.S.J. tension in the bottom flange menced to curve away and the loads at these and compression in the concrete, were taken elastic limit points were called the test loads for specimens J, K a n d L. At the limit of and were reduced in every case to equivalent proportionality tension steel the in the uniformlydistributedloads. 15.2 tons averaged for the three slabs was The report states that " the concrete was per sq. in., and the compression in theconcomposed of 4 parts Thames gravel,2 parts crete 920 lbs. per sq. in. of sand to I part of cement, mixed in the In regard todeflection, it should be noted ordinary way, worked between the joists of that the concrete slabs are much stiffer than the frames with a shovel and then levelled the plain joists. Although carrying heavier off. No special precautions taken, were loads, the deflection of the concrete slabs is with the idea of reproducing the conditions approximatelyonly half that of theplain of work on an ordinary building contract. joists. Tests of concrete cubes and cylinders made COMPARISON O F RESULTS WITH THEORY. atthesametime 'as thefloorshavean Mr. Sharman states that " by calculation average crushing strength only 1,443 lbs. of similar slabs, made with concrete having a of persq.inchandanaveragemodulus crushing strength of 2,400 lbs. per sq. in. elasticity of 1,250,000, showing the cona n d a modulus of z,ooo,oooindicate a n incretetobebelowaveragequality.This, crease of only about 5 per cent. in the load however, did not materially affect the carried before a similar stress would be strength of the slabs. reached in the steel. We refer to this point later. W e think that Mr. Sharman must have SUMMARY O F RESULTS. based his calculationsuponthetheoryof reinforced concrete slabs which assume the % decrease in desteel as concentrated at its centre and takes flection due to concrete (in spite of no account of the depthof the steel, and not increased load). upon the theory which we have advocated for many years. As the latter theory does not appear to be very well understood it may be interest to of work out in detail one of the cases tested and to give the results all the others. of We will take case L , in which we have 6 x 3 I beams at 2 ft. 9 in. centres and con-
~~
-~
~
40
crete
2
THE STRUCTURAL ENGINEER:The Journal

results for various percentageso f rcinforcement ; the diagrams usually given do not extend sufficiently far to deal with the presentcasesinwhichthepercentage of reinforcement is higher than in ordinary reinforcedconcretedesign. I n ourcasethe
in. above the top flange. 2 illustrates this, together with the stress diagram.
Fig.
% reinforcement
is
253 = 2 - 14, x 100 33 x 5

~~
n - = -545, so that d n = -545 X 5 = 2.72in, We next find the equivalent moment of inertia (IE) of the section about t h i s neutral axis of the formula bn3 I E =m (I, A (d - n2) 3 where IB = moment of inertia of steel beam about its own N.A.
and fromthediagram
x 2 * 7 2 3 + 1S ( 2 0 - 2 3-53~2.28~ 3 = 221 l 5 X 38.5 = 798 inch units. T h i s if BC and Bs are the safe bending moments in inch pounds which will induce stresses of c in the concrete in compression a n d t in the steelintension, we have for c = 600 and t = 16,000.
- 33
In this method we first find the neutr axis by the ordinary formula for reinforced concrete beams, viz. :b n2 - = m A (d-n) ... ... ( l ) 2 we take m = 15, and for the 6 x 3 I beam A = 3.55 * . inserting our values from Fig. 2 we have = 15 x 3.53 (5-11) 2 16.511~ 26S-S3n = n2+3-21n-16-1 = 0
"'
x 798 = 176,000 in.-lb. 2.72
3 3 n :
n =
- 3-21
+
2
2/3-212
2
+ 64.6
= - 3.21
+ 8.65
by means of the diagram Fig. 3, which gives
= 2 - 7 2 in. The above formula can solved be
= 161,000 in.-lb. 5-28 For the I beam alone we have 16,000 X 20.2 B 3 ( 2 0 - 2 being the moment of inertia and 3 the half depth), = 108,000 in - lb. . . Increaseduetocasing = 161,000 108,000 = 53,000, . . increase in strength due to casing =
- 16'ooo
IS
. __ 798
~~__.__.____
-_ J 3
108
x l00 = 49.
o THE INSTITUTIONSTRUCTURAL f OF ENGINEEKS. 41

carriedtothepracticalfailure point-say thepoint at whichthedeflectionbecame 5 in.-then the uncased joists would show considerably strength those lower than cased flush top and bottom. We think also thattherelativeincreases of strengthfor % increaseinstrength. the other cases would be higher if the pracMark. Bv Theorv. l Bv Test. of the tical failure load were taken instead D 60 elastic limit load. 40 F 140 90 Atsomelaterdate,perhaps,the load29 38 deflection diagrams will be published, and J L 49 38 we shall then be able to have an interesting K 79 -~ l 85 discussion of them and shall probably be -~ able to learn much from them. Onthewholethiscomparisonshows a In many respects it a pity that theconis fair agreement between theory and practice, crete was clearly below par in quality, but and although Mr. Sharman thinks that the it is reassuring that even with poor concrete IOW strength of the concrete did not materithe increased strength due to the casing was ally affect strength the of the slabs, we so great. believe thattheresultswouldhavebeen There is one point which we do not follow higher had the concrete been of the usual from the report, but which is of importance strength. forthetrueinterpretation of theresults; COMMENT UPON RESULTS. that is the manner in which the weight of theconcretecasing itself has beentaken It should be noted that the results given into account. If it should transpire that the are for the increase of load for the elastic loads given the are superimposed loads, limit as determineddeflection by the then it is clear that the relative strengths measurements, and no record is given as to for cased and plain beams are really higher the relative ultimate strength of the specithan the figures given. mens. We hope this research work will be conThisbrings u s to a pointuponwhich tinued; there are many moot points in the there is a wide difference of opinion among design of structural steelwork that can only engineers; we believe that those engineers We believe that be settled by experiment. who have most experience of experimental there are many directions in which econowork are agreed that is, on the whole, best it mies in design could be effected if we had tobaseworkingstressesupontheelastic the experimental evidence would which limit, the but factor of safety one that enable us to design uponsureknowledge, adopts should depend upon ratio the of and thus to reduce that child of ignorance theelasticlimittotheultimatestrength. that we call the Factor Safety. of T h a t is why we use a lower basis stress for sufficientevidence It is quiteclearthat compression members for than tension has now been collected to show that a conmembers (usually 6.5 a n d 7.5 tons per sq. crete casing has a marked effect in increasin. for mild steel). ingthestrength of structuralsteelwork; Although,therefore,wethinkthatthe a n d we should see to it that those who %it elastic limit strength is of great importance, o n the local authorities over us shall make we think that the ultimate strength should suitable revision in the official regulations also be considered. No figures as to these that hamper efficient design and make steel rrltimate strengths are given in the report. We expect, however, thatif the tests were structures cost more than they need. T h e test result was 38 per cent. increase. a comparison The following table gives between the results of the tests and of the above theoretical treatment for the various cases.
____~.
~~

The Strength of Filler Joist Floors.

Uploaded by

Copyright:

Available Formats

The Strength of Filler Joist Floors.

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

The Strength of Filler Joist Floors.

Uploaded by

Copyright:

Available Formats

What type of floor structure is discussed in the document?

What were the results of the tests conducted on the floor structures?

THE STRENGTH OF FILLER JOIST FLOORS.-By Ewart S Andrews, B.Sc. En@, M.I.Struct. E., A.M.1nst. C.E .

THE STRUCTURAL ENGINEER.

THE STRUCTURAL ENGINEER:The Journal

253 = 2 - 14, x 100 33 x 5

x 798 = 176,000 in.-lb. 2.72

by means of the diagram Fig. 3, which gives

= 2 - 7 2 in. The above formula can solved be

o THE INSTITUTIONSTRUCTURAL f OF ENGINEEKS. 41

You might also like