Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
FIGURES
the Commentary on the 1963 ACI Code. Measured and computed strengths were i n good agreement when applied load was assumed to enter the column through the two faces Joining the loaded area. The structure possessed some strength under t e s t load conditions a f t e r each shear f a i l u r e . This capacity could be accounted f o r almost en- t i r e l y by y i e l d l i n e a n a l y s i s . L i t t l e strength was available at the columns a f t e r the i n i t i a l punching f a i l u r e had occurred. The t e s t s involved successful application of many laboratory proce- dures i n the f i e l d . Control of the t e s t and data c o l l e c t i o n were within tolerances somewhat better than those customarily obtained i n the f i e l d . P r i o r to application of t e s t loads, d i f f e r e n t i a l settlement of the structure probably took place; diagonal cracks were v i s i b l e i n a wall and i n edge beams. Analysis indicated that t h i s settlement could have caused added loads on columns. Settlement combined with shrink- age may also have produced t e n s i l e in-plane forces in the slab. These i n t e r n a l forces probably reduced shear capacity, p a r t i c u l a r l y at edge columns. 1-56
CONCLUSIONS The behavior of the structure was i n general accord with e x i s t i n g design theories, though a shear weakness was observed at edge columns. However, damage t o the structure before t e s t i n g began may have i n - fluenced both performance and strength. Performance of the t e s t structure was satisfactory \ander 350 psf applied load (added dead plus about 1.0 l i v e load). Steel stresses were low and deflections were w i t h i n acceptable l i m i t s . No ab- normal increases i n crack width were observed at service load. Deflections computed using an equivalent frame analysis were i n good agreement with measured values. Although the frame analysis i s cvmibersome, i t appears t o be a r e l i a b l e method f o r p r e d i c t i n g deflections at the center of panels. For a l l t e s t s the structure s a t i s f i e d c r i t e r i a f o r eval\xation of load tests stipulated i n the 1963 ACI Code. Flexural capacity was not reached i n any of the three t e s t s . Ul- timate strength was governed by shear at i n t e r i o r and edge columns before a y i e l d mechanism coTild be developed. Shear strength of the slab at the i n t e r i o r column loaded from a l l four sides was about 20 percent greater than that implied by ultimate strength design methods of the I963 ACI Code. Shear strength of the slab at edge coltmins was less than t h a t implied by slab design procedures of the Commentary on the 1963 ACI Code. However, measured and computed strengths were i n good agreement when the shear was considered carried by the edge beams. Shear strength of the slab at an i n t e r i o r column supporting a single loaded panel was about 60 percent of t h a t implied by design methods of 1- 55
I n a l l three t e s t s , load was d i s t r i b u t e d t o adjacent supports a f t e r f a i l u r e . However, load-deflection curves f o r the three t e s t s show that large deforma- t i o n s r e s u l t when a coltmm support i s l o s t t h r o u ^ shear f a i l u r e . I n Test I I I , load capacity dropped substantially a f t e r shear f a i l u r e at both Colimins C5 and D5. Shear f a i l t i r e s had already occurred at adjacent columns i n Line 4. Consequently, an applied load of only 644 psf could be maintained. This low load capacity was due t o lack of load r e d i s t r i b u t i o n since nearby supports had already punched through. Because the loading system was hydraulic, load dropped o f f when a shear f a i l u r e occurred. A hydraulic system i s not capable of immediate response to instantaneous deflections. I f -the structure had been subjected t o gravity loads, severe damage would have resulted a f t e r the f i r s t column punched through the slab. In each of the t e s t s , only a part of the structure was loaded. VJhen dis- tress occurred load was d i s t r i b u t e d t o adjacent parts of the structure. This d i s t r i b u t i o n could not take place i f the whole roof area were sub- jected t o a uniform gravity load. After the f i r s t column punched through, capacity at adjacent columns wovild be exceeded and t o t a l collapse wo\ild occur. 1- 54
Test I I I . This t e s t was intended t o investigate the influence of i n - plane forces on the strength of the slab. Ockleston(5) and other invest- igators have shown th a t in-plane forces are substantial. Ockleston found that f a i l u r e loads were 2.5 t o 3 times the loads computed by y i e l d - l i n e analysis. This excess strength was a t t r i b u t e d t o arching action. In t h i s single-panel t e s t , the slab exhibited some i n e l a s t i c behavior p r i o r t o punching at Col\amn C5. Load versus both deflection and s t r a i n relationships became non-linear under an applied load of 1500 psf. Yield of reinforcement i n the positive moment regions indicated i n c i p i e n t formation of a y i e l d - l i n e mechanism. Shear f a i l v i r e at Column C5 occurred at 1972 psf (added dead load plus 6.4 l i v e loads). Subsequently, shear f a i l i i r e occurred at Column D5 when applied load was again increased t o 1449 psf. Shear capacity at Colimin C5 was not adequately predicted by e i t h e r Moe's equation or the ACI method. However, good agreement was obtained by mod- i f y i n g the ACI method. In t h i s modification, l i v e load was considered to enter the column only through the two sides j o i n i n g the loaded area. The manner i n which load was transmitted t o Column C5 appeared t o be similar t o that f o r a comer colvmin. POST-FAILURE BEHAVIOR Af t e r i n i t i a l f a i l u r e , loading was continued i n a l l three tests t o determine the c a p a b i l i t y of the structure t o continue to carry load. I n each case, the structure had a capacity t o support some load a f t e r f i r s t f a i l u r e . Table X I I I l i s t s applied load i n t e n s i t y at f a i l u r e and the maximum applied load a f t e r f a i l u r e f o r each of the three t e s t s . Data f o r Test I I are from the second part of the t e s t . TABLE X I I I POST-FAILURE STRENGTH Test No. Maximimi Applied Load, psf Maximum Applied Load A f t e r Failure, psf I 895 579 I I 728 594 I I I 1972 1449 1-53
Shear strength of the slab was s a t i s f a c t o r i l y predicted i n Test I by both Moe's equation and the I963 ACI Code.^^^ I n t h i s t e s t , loading was arranged so t h a t v e r t i c a l shear transmitted from the slab t o Column Ck was uniform about the periphery. This condition i s the simplest case of shear tr a n s f e r between slab and column. Test I I . F i r s t f a i l u r e i n Test I I occurred at Col\min E3 under an unex- pectedly low load. In the second part of Test I I , shear strength at Column Ek was investigated. Conditions of loading were very nearly the same i n both parts of the t e s t , and f a i l u r e at Ek was nearly i d e n t i c a l t o t h a t at E3. Calculated strength using methods f o r shear i n slabs overestimated shear capacity at Columns E3 and Ek. Since the slab was extremely s t i f f compared t o the columns, l i t t l e moment was transferred to the edge columns. Shear stresses were p r i n c i p a l l y from v e r t i c a l load. Computed shear strength of the edge beams at Columns E3 and Ek s a t i s f a c t - o r i l y predicted measured values. Although a shear cone was developed, i t appears that strength was governed by beam action. Slab action was not developed since there was essentially u n i a x i a l bending along the edge of the structure. From service load up to about the load that caused shear f a i l u r e at Column E3, the slab responded " e l a s t i c a l l y . " Only when the diagonal cracks opened at E3 was there evidence of distress. Just before f a i l u r e , load-strain and load-deflection relationships became non-linear i n d i c a t i n g r e d i s t r i b u t i o n of i n t e r n a l forces. The slab exhibited one-way action i n Test I I . With three adjacent panels loaded along the edge of the bu i l d i n g , the behavior of the slab was similar to t h a t of a propped cantilever beam. Nearly f u l l restraint was provided by continuity of the slabs along Line D while the edge columns at Line E pro- vided the v e r t i c a l props. 1- 52
Test t o calculated r a t i o s i n Table XI show tha t f o r mechanisms A, B, and C maximum load during t e s t was only about one-half of the computed load. This corresponds to t e s t observations since no f i i l l y developed y i e l d pattern was noted. I n each of the three t e s t s , shear f a i l t i r e of the slab prevented development of a f l e x u r a l mechanism. In Test I I I , however, a y i e l d l i n e pattern was p a r t i a l l y developed. As described e a r l i e r , reinforcement y i e l d occurred i n the positive moment region of the slab. I n Mechanisms A, B, and C, i t was assiamed th a t y i e l d l i n e s were w i t h i n the te s t areas. Cracks that formed i n the slab during the tests indicated y i e l d l i n e s could develop outside the loaded panels. Flexural capacities f o r Test I a f t e r f i r s t f a i l u r e and Tests I I and I I I before f i r s t f a i l u r e were evaluated assuming y i e l d l i n e s outside the loaded areas. In Test I , loading was continued sifter punching occurred a t Column Ck. A sketch of cracking on top of the roof i s shown i n Fig. 3Ì . These cracks circumscribe the test area and are located i n the v i c i n i t y of cu t - o f f of negative moment reinforcement. Only temperature steel reinforced the cracked sections f o r negative moment at these locations. The crack pattern indicated t h a t negative moment y i e l d l i n e s had formed at the locations of the cracks. Mechanism D shown i n Fig. 58 represents the collapse mechan- ism. Computations were based on the assumption that Column CU carried no load. This pemitted the four panel t e s t area to act as one large panel. The four t r i a n g u l a r areas th a t include the load must rotate about axes through column Lines B, 5* and 3- This requires the slab t o deflect downward within the loaded area and t o raise upward beyond the axes of ro t a t i o n . Slab elements outside the negative y i e l d l i n e s rotate about the positive moment y i e l d l i n e s . This mechanism agrees w e l l with the de- formation of the structure observed near the end of Test I . Using the p r i n c i p l e of v i r t u a l work, t o t a l load t o form Mechanism D was computed to be 690 psf. This quantity compares w e l l with the maximvmi load of 799 psf that could be applied a f t e r Column Ch punched through the slab. The t e s t t o calculated r a t i o i s I.16 f o r t h i s mechanism. This r a t i o i s not unduly high since s t r a i n hardening and a x i a l forces i n the slab were ignored. 1-51
are taken i n t o account. S i m i l a r l y , the influence of unequal span lengths are considered. I n the discussion t o the paper on frame analysis, (15) i t was suggested that an approximate method be used f o r computing deflections. This prodedure uses a one f t wide s t r i p along one of the diagonals of a panel. Deflection at midspan of the s t r i p i s calculated assuming f i x e d ends. Load applied t o the beam i s one-half the design load. Moment of i n e r t i a of the beam i s based on the uncracked section. Deflections were computed using t h i s approximate method and taking i n t o account the difference i n moment of i n e r t i a between the s o l i d portion and the waffle portion of the slab. Deflections computed by t h i s shortcut method underestimated deflections measured i n the loaded panels of the t e s t struotxjre. Calc\ilated deflections were about one-half t o one-quarter of the measured deflections. Even considering the slab cracked, deflections are s t i l l somewhat less than computed. I t appears that f o r t h i s structtire the approximate method does not adequately predict behavior. STRENGTH UNDER OVERLOAD Strength under overload was governed by sudden shear pvmching i n a l l three t e s t s ; f l e x u r a l capacity was not reached. Only i n Test I I I was substantial y i e l d i n g noted before shear f a i l u r e occurred. Test I . The structure behaved " e l a s t i c a l l y " u n t i l f a i l u r e occurred. Above service loads, load-deflection and load-strain relationships remained very nearly l i n e a r . Even at an applied load of 843 psf (dead load plus 2.7 l i v e loads), deflections a t the center of loaded psmels were less than 0.70 i n . This d e f l e c t i o n corresponds to a r a t i o of about L/500. S i m i l a r l y , r e i n - forcement stresses were low. The reinforcement stress increased by 20,000 p s i under an applied load of 843 psf at only one location. At most other locations, reinforcement stresses were 10,000 psi or less. 1- 50
by Eq. 6. I n Tests I I and I I I , the measured deflections were about twice those specified by the Code. The t e s t load was not held f o r 2k hours. However, because of age of the structure and previous loading, additional deflection from creep over a 24-hour period would not be large. TABLE X I I CENTER OF PAMEL DEFLECTIONS Measured Recovery, Test Panel F i r s t , X I I . Second L^/20,000 t . F i r s t =11 b Second No. No. Loading Loading i n . Loading Loading 3,iÌ -B,C o.6o o.hk 0.24 67 93 T 4,5-B,C 0.i^6 0.36 0.24 92 92 1 3,U-C,D 0.42 0.37 0.22 86 95 It,5-C,D 0.29 0.29 0.22 86 93 2,3-D,E 0.66 0.59 0.25 67 87 I I 3,U-D,E 0.65 0.64 0.25 82 83 lf,5-D,E â â 0.25 ââ¢â â I I I 5,6-C,D 0.49 0.46 0.22 97 98 Measured recovery of deflection from the short term load i s also l i s t e d i n Table X I I . Recovery i s given i n percent of the measured deflection at the center of the panel. I n Panels 3,4-B,C and 2,3-D,E recovery was less than 75 percent of the maximum deflection f o r f i r s t loading. However, a l l panels exhibited recovery more thaji 75 percent f o r the second loading. Although comparisons made are based on short term deflection data, indications are th a t the structure would s a t i s f y load t e s t requirements of the 1963 ACI Code.(20) Deflections computed according to frame analysis(l5) agreed s a t i s f a c t o r i l y with t e s t r e s u l t s . This method was developed f o r slabs under uniform load- ing. However, f o r these t e s t s i t appears that the procedure i s applicable also when only a portion of the structure i s loaded. By considering frame rotations at the ends of loaded spans, ef f e c t s of adjacent unloaded areas 1-49
reasonably w e l l . Deflection l i m i t a t i o n s are given i n Chapter 9 of the ACI Code (ACI 3l8-63).(20) por roofs intended t o support other con- st r u c t i o n l i k e l y t o be damaged by large deflections, the allowable l i m i t f o r deflections i s L/360. This allowable de f l e c t i o n includes immediate deflection from l i v e load plus de f l e c t i o n due t o creep and shrinkage. For the panels tested the value of L/360 ranged from O.9O i n . t o O.96 i n . when L was taken as the short span dimension. Deflections measiired i n the three t e s t s a t 350 psf (added dead plus about 1.0 l i v e load) ranged from 0.15 t o 0.37 i n . These deflections are about O.166 and 0.333 of the ACI l i m i t . Even with time-dependent ef f e c t s added, the structure should easily s a t i s f y the c r i t e r i o n f o r deflections given i n Chapter 9 of "the ACI Code. (20) Chapter 2 of the ACI Code^^^ contains provisions f o r load t e s t s of struc- tures. When a load t e s t i s required by a bu i l d i n g o f f i c i a l , superimposed load s h a l l be equal to 0.3 times the dead load plus 1..7 times the service l i v e load. For the te s t structure t h i s i s 590 psf. Under t h i s load, max- imum defl e c t i o n should not exceed: L^/20,000 t (6) where L = the shorter span of slabs t = t o t a l depth of member I f the maximum deflection exceeds L^/20,000 t , recovery of deflection w i t h i n 2k hours a f t e r load i s removed s h a l l be at least 75 percent of the maximum deflection. The t e s t load must remain i n positio n f o r 2k hours. I f recovery of deflection i s less than 75 percent, a second t e s t may be conducted 72 hours a f t e r removal of the f i r s t t e s t load. Measured deflections and deflections computed by Eq. 6 are l i s t e d i n Table X I I . Measured deflections at 590 psf were taken from load-deflection curves. Zero deflection f o r f i r s t loading was that at zero applied load at the s t a r t of each t e s t . Zero deflection f o r second loading was that at zero applied load following application of load t o added dead plus I.5 l i v e loads. I t i s seen that measured deflections were greater than those computed l.k8
DISCUSSION OF RESULTS Structural behavior of the t e s t structure may be divided i n t o three cate- gories: (a) service load behavior, (b) strength under overload, and (c) post - f a i l u r e behavior. Service load i s defined as added dead load plus one l i v e load. Behavior vinder overload i s defined as that under load greater than service load; i t includes the load at which the f i r s t punch- ing f a i l u r e occurred i n each t e s t . Post-failure behavior i s defined as that a f t e r the f i r s t punching f a i l u r e had occurred. The tests were short-term, covering a period of about two days f o r each t e s t . Long-term effects were not measured. However, the b u i l d i n g was about two years old at the time of t e s t . The structure showe_d evidence of dam- age apparently caused by shrinkage and d i f f e r e n t i a l settlement. These long-term effects may have influenced r e s u l t s of the short-term t e s t s . SERVICE LOAD BEHAVIOR Response t o service load was detemined by measuring strains i n both the reinforcement and the concrete and by measuring deflections. Prior t o application of the t e s t loads, the concrete was severely cracked. However, no additional cracking was observed i n loading t o dead plus one l i v e load. Cracks that were present remained quite narrow. Strain gages were located where maximum stresses were expected. Measured strains showed that reinforcement stresses were low at service load. Max- imum change i n stress under application of added dead load plus l i v e load was 13,500 p s i . Most strains recorded indicated an increase i n stee l stresses less than 10,000 p s i . Measured and computed deflections at the center of loaded panels are shown i n Fig. 53 f o r a l l three t e s t s . I t i s seen th a t the two quantities agree
After Colvmms C5 and D5 had punched through, a load of about 650 psf could be applied to the panel. I n t e n s i t y of applied load did not drop off with continvied deformation of the slab. I t appears that a y i e l d mechanism was forming. However, the damaged condition of the structure prevented deter- mination of the location of a well defined mechanism. The structure, even though severely damaged, did possess the capacity to r e s i s t moderate loads. Shear f a i l u r e s prevented development of f u l l y i e l d - l i n e mechanisms i n each of the three t e s t s . Consequently, no information was obtained on possible strengthening e f f e c t s from a x i a l force in the slab. l-k6
There i s r e l a t i v e l y l i t t l e difference i n load between Mechanisms E and B for Test I I before shear f a i l i i r e occurred. Positive moment y i e l d l i n e s for both mechanisms are assumed to be located within the loaded area. Neg- ative moment y i e l d l i n e s f or E are located i n panels adjacent to the t e s t area at the bar cut-off l i n e . Positive moment y i e l d l i n e s are located adjacent to the wall along Lines 1 and E and at the face of columns along Lines C and 6 . For Mechanism E, a f a i l u r e load of iklO psf i s calculated. This i s somewhat l e s s than the 1 5 5 0 psf computed for Mechanism B. Both the loads axe about twice the load i n t e n s i t y at shear f a i l u r e i n the t e s t . Various mechanisms were investigated for Test I I I . In these patterns, the negative y i e l d l i n e s were extended beyond the four sides of the loaded panel. The lowest applied load was found for Mechanism F shown i n F i g . 60, the negative moment y i e l d l i n e s being p a r a l l e l to the panel diagonals. The computed load of 2^*80 psf i s only s l i g h t l y greater than the maximum t e s t load. Test to calculated r a t i o f o r t h i s mechanism i s O . 8 9 . During the t e s t , y i e l d s t r a i n s were measured in much of the positive moment reinforce- ment. I t i s apparent that the y i e l d l i n e mechanism had almost completely formed when Column C5 punched through the slab. Cracking observed on top of the slab a f t e r Test I I I , i s shown i n Fig. 5 1 . At the time Test I I I was started, shear f a i l u r e had occurred at Colimns Ck, Bk, Ek, and E 3 . Consequently there was l i t t l e or no support of the slab at these coliamns. Crack 1 was observed before f a i l u r e occurred at Column C5. After C5 punched through the slab. Cracks 2 , h, and 5 were noted. F i n a l l y , Crack 3 formed a f t e r both Columns C5 and D5 had punched through. Except for Crack 2 adjacent to Coliarins C6 and D6, cracks formed at the cut-off of negative s t e e l . Considerable positive moment cracking was noted on the underside of the slab a f t e r Test I I I was completed. The column s t r i p along Line C between Lines k and 5 was severely cracked. Along t h i s section no positive mo- ment reinforcement was provided. Consequently, moment capacity of t h i s section was only that of the p l a i n concrete section before the crack formed and was zero a f t e r the crack formed. 1 - 4 5
The y i e l d - l i n e mechanism that corresponds to the f l e x u r a l strength of a slab i s the one produced by the lowest load. SeveraJ. t r i a l s may be necessary to se l e c t the governing mechanism. I n most cases, load determined by t e s t w i l l be greater than that computed from y i e l d l i n e a n a l y s i s . This r e s u l t s from ignoring i n the analysis both s t r a i n hardening of the reinforcement and a x i a l forces induced i n the slab. Y i e l d l i n e analyses were made for each of the three t e s t s . I n computing mo- ment along hinging l i n e s , y i e l d s t r e s s for a l l reinforcement was assimied to be 42 k s i . V/here possible, e f f e c t i v e depth was based on measured section thickness. Average concrete strength of 5 2 4 0 p s i was used for a l l c a l c \ i l a - t i o n s . Ultimate moment calculations were made according to provisions i n the 1 9 6 3 ACI Code. The strength reduction factor, cp, was taken as 1 . 0 . For a l l sections, temperature reinforcement was included i n cal c u l a t i o n of moment. Patterns of y i e l d l i n e s are shown in F i g . 5 6 , 5 8 , 59> and 6 0 . Calculated f l e x u r a l strengths and maximum loads i n the t e s t s are l i s t e d i n Table XI. These loads are applied loads plus the actual dead weight of the slab and loading equipment. Y i e l d - l i n e mechanisms A, B, and C are patterns commonly assumed for the three t e s t conditions. Negative moment y i e l d l i n e s are located at the periphery of the loaded panels. Positive moment y i e l d l i n e s i n t e r s e c t within the loaded panels. TABLE XI FLEXURAL STRENGTH Test Calculated Y i e l d Strength Maximum Total Load during Maximum Test Load No. Mechanism* Load, psf Test, psf Calc. Y i e l d Load I A D 2l40 6 9 0 1 1 1 5 7 9 9 0 . 5 2 1 . 1 6 I I B E 1 5 5 0 l4lO 842 8 4 2 0 . 5 4 0 . 6 0 I I I C F 3 9 8 0 2 4 8 0 2 2 0 2 2 2 0 2 0 . 5 5 0 . 8 9 1 - 4 4
By the ACI method, moment e f f e c t s were ignored, and i t was assumed that shear s t r e s s was uniformly distributed around the c r i t i c a l section. I t should be expected that shear s t r e s s e s from dead load can be approximated c l o s e l y assigning a uniform d i s t r i b u t i o n . However, a t a comer column i n the single panel t e s t , the s t r e s s d i s t r i b u t i o n would not be escpected to be uniform for applied load. To get into the coliann, load wooild come p r i n c i p a l l y through the two sides facing the loaded area. Shear strength using the ACI s t r e s s l i m i t a t i o n was recomputed for the slab at Column C5. I t was assumed that dead load supported by the column was uniformly distributed at d / 2 from the column. Live load was then assiuned to be distributed uniformly on a section at d / 2 from the two sides of the column jo i n i n g the loaded area. The difference between the l i m i t i n g s t r e s s , h^i, and the nominal dead load s t r e s s , "Vjyjjlod, was assumed to be available to carry l i v e load. The shear capacity computed using these assumptions was 7 0 0 kips. This value i s i n good agreement with the mea- sured ultimate shear. While seemingly conservative, the assumptions made in t h i s a n a l y s i s c l o s e l y predict the r e s u l t s of t h i s t e s t . FLEKURAL STRENGTH In Test I I I , y ielding of reinforcement i n the positive moment region i n d i - cated that f l e x u r a l capacity was approached. Shear f a i l \ i r e prevented de- velopment of negative moment yi e l d i n g i n the slab during t h i s t e s t . After shear f a i l u r e i n Test I , load required to deform the structure became l e s s as deflections increased. This behavior plus observed cracking indicated formation of a f l e x u r a l mechanism. Since the t e s t slab was underreinforced at a l l sections, f l e x u r a l strength was governed by yielding of the reinforcement. F l e x u r a l strength of a slab i s reached under a load that produces f l e x u r a l y i e l d i n g along an adequate nvmiber of l i n e s to produce a collapse mechanism. An ana l y s i s based on y i e l d l i n e mechanisms was developed by I n g e r s l e v ( 2 2 ) and modified by Johansen. ( 2 3 ) In t h i s a n a l y s i s , i t i s assumed that the y i e l d - l i n e sections possess adequate rota t i o n a l capacity for the mechanism to form completely. 1- k3
s = spacing of s t i r r u p s i n a direction p a r a l l e l to the longitudinal reinforcement = ultimate shear c a r r i e d by web reinforcement In both equations the strength reduction factor, cp, was taken equal to unity. Width of the beam, b, was taken as i n . Measured quantities were used for f J and d. The r a t i o V/M was determined from the frame a n a l y s i s for column loads. Y i e l d strength of the s t i r r u p s was assumed at 42 k s i . Shear reinforcement was very l i g h t , consisting of two closed-loop No. 3 bars spaced at 12 i n . Computed shear strengths, V^^^, at Columns E3 and Ek are l i s t e d i n Table IX. Computed values are the sum of edge beam strength at a distance, d, from the narrow faces of the columns. For both locations, E3 and Ek, strength com- puted using Eq. k and 5 more cl o s e l y predicts measiared capacity than the slab shear equations. Test to calviated r a t i o s for both E3 and E4 approach unity when analyzed as beams. This indicates that beam action predominated i n shear strength. Although the shear f a i l u r e s were conical about the columns, i t appears the diagonal cracking was precipitated by beam action. Subse- quently, s t r e s s e s were redistributed so that v e r t i c a l load was transferred through the long face of the col\imn. The extent to which the low shear strengths observed i n Test I I were caused by damage to the structure be- fore the t e s t began i s not known. Further laboratory t e s t s regarding the manner in which v e r t i c a l shear i s r e s i s t e d at edge columns are i n progress at the PCA Laboratories. Test I I I . Measured shear c a p a c i t i e s at Column C5 were l e s s than predicted. Table IX indicates that t e s t strength was only 6 0 to 6 5 percent of that c a l - cvaated by e i t h e r the ACI Code with no reduction for moment or by Moe with the moment e f f e c t determined by Eq. 2. Moment e f f e c t s analyzed by Moe were only for the condition of bending about one a x i s . From the arrangement of loading i n Test I I I , i t i s obvious that moment i s transferred to columns by bending about two axes. Thus, i t i s not surprising that strength i s underestimated using Moe's procedure. 1- k2
f a i l u r e surface of the t e s t slab with that of a laboratory specimen designed to be weak in torsion. Cracking at the free edge of t h i s laboratory speci- men i s shown i n Fig. 57* Diagonal cracks run i n the opposite d i r e c t i o n from those of the t e s t structure shown in F i g . 38 and 40. In the laboratory , specimen, the diagonal cracks are outward from the column beginning at the junction of the column and the top of the slab. These cracks are the r e s u l t of t o r s i o n a l moment. Observed behavior of the slab i n Test I I indicated one-way action. There was l i t t l e b i a x i a l bending at the edge colimins, and moments across Line E were small. Panel loads were transferred to the colianns p r i n c i p a l l y through the edge beam along column Line E. Consequently, shear strength i s more clo s e l y represented by a beam than a slab. Shear strength for the edge beam was computed using Eq. 17-2 of the ACI Code. (20) Ultimate shear strength i s given by: = hdi[l.9yfj + 2500(PâVd/M) ] (4) where A = Area of tension reinforcement s b = width of compression face of f l e x u r a l member d = distance from extreme compression f i b e r to centroid of tension reinforcement fc = compressive strength of concrete M = bending moment Pâ = As/bd V = t o t a l shear at section = t o t a l ultimate shear Strength due to v e r t i c a l s t i r r u p s was computed by using Eq. 17-4. This equation i s : = (Avfyd)/s (5) where A^ = t o t a l area of web reinforcement i n tension within a distance, s, measured i n the d i r e c t i o n p a r a l l e l to the longitudinal reinforcement fy = y i e l d strength of reinforcement l - 4 i .
COMPARISON OF MEASURED AMD COMPUTED SHEAR STRENGTHS Test I . Ratios of t e s t to calculated values are l i s t e d i n Table IX for the f i r s t column that punched through during each t e s t . I t i s seen that shear strength a t Column Cl* i s conservatively predicted by the ACI Code method. Strength predicted by Moe's equation i s somewhat higher than that measured. This, however, i s not beyond the range of sc a t t e r encountered in the t e s t s eval\xated by Moe. Moe's equation i s based on a " b e s t - f i t " curve and not a lower bound curve. Test I I . Colimuis E3 smd E^^ were subjected to the same magnitude of force in the t e s t s . Although Column E3 had a somewhat greater cross section than Coltann Ek, shear f a i l u r e occurred f i r s t a t E3. The f a i l u r e load f o r E3 was s l i g h t l y l e s s than that f o r Ek. Observed shear strength a t E3 was about 50 percent of that predicted by Moe's equation and about U5 percent of that using the ACI equation. Shear strength at Elt- was also l e s s than that predicted by both methods; i t was nearly 60 percent of that predicted by Moe and about 55 percent of the ACI prediction. Application of Moe's Eq. 2 indicated that the e f f e c t on shear strength of moment tr a n s f e r was small. Shear capacity was 20 kips l e s s at E3 and 30 kips l e s s at Eh than that for "concentrically" loaded columns. The low strength exhibited i n the t e s t s at E3 and Ek was probably inflvienced by the condition of the structure. As shown i n F i g . k, diagonal cracks were v i s i b l e a t E3 p r i o r to t e s t i n g . Although looA from settlement was added to the t e s t value for E3, t h i s increased the column load a t f a i l u r e by only about 10 percent. Tensile s t r e s s e s i n the plane of the slab r e s u l t i n g from settlement and shrinkage may also have contributed to the unusually low strength. Diagonal cracking at Columns E3 and Ek i s shown i n F i g . 38 and kO. Since l i t t l e moment was transferred to the columns, s t r e s s e s from torsion were not s i g n i f i c a n t . The small Influence of torsion can be seen by comparing the 1-ltO
applied from two directions, across Line C and across Line 5- The moment used i n calculations was that across Line C, the greater of the two. ^ f l e x ^ term in Eq. 1 to indicate the condition of the slab r e s u l t i n g from bending moments. The term accounts for cracking, depth of neutral a x i s , compressive s t r e s s e s , and other states of damage. In these calcu- l a t i o n s , Vâ was taken as the column reaction found by the frame an a l y s i s with an applied load equal to the f l e x u r a l capacity determined by y i e l d l i n e a n a l y s i s . Loads determined by y i e l d l i n e a n a l y sis are shown i n Fig. 56. In the 1963 ACI Code, a method for computing shear strength of slabs i s contained i n Section 1707 of Chapter 17. Provisions for t r a n s f e r of moments are specified i n Chapter 9, Section 920 of the Code. Additional information i s included m the Code Commentary. (21) For v e r t i c a l loads only, ultimate shear strength i s given by: = 4bdv/I7 (3) Notation i n t h i s equation i s s i m i l a r to that for Eq. 1. In Eq. 3, however, the c r i t i c a l section, b, i s defined as the perimeter located at d/2 from the column faces. When moment i s transferred, the c r i t i c a l section b i s d i f f e r e n t from that given above. I n the direction of moment, the c r i t i c a l section remains equal to the column width plus d /2 each side of the column. The section peirpendi- cular to the direction of moment extends I . 5 times the slab depth to each side of the column. The Code Commentary ^^^^ s p e c i f i e s that moment transferred by torsion i s that part of the unbalanced moment exceeding the f l e x u r a l capacity of the c r i t i c a l sections. Unbalanced moments at Columns E3, E4 , and C5 were i n each case l e s s than f l e x u r a l capacities at the c r i t i c a l section. Therefore, no part of the unbalanced moments added to the v e r t i c a l shear s t r e s s e s . 1-39
TABLE X SHEAR EQUATION PARAMETERS Moe 's Equation ACI Moe's and ACI Equation Location Test No. Column Size, i n . X i n . b, i n . r , i n . V f l e x , kips Equation, b, i n . d, i n . p s i Unbalanced Moment, k-in. Colimin C4 I 26 X 26 104 26.0 2190 2240 179 18.8 72.4 â Column E3 I I 12 X 32 56 26.6* 620 740 113 28.3 72.4 530 Column E4 I I 12 X 24 48 20.0* 620 740 102 27.1 72.4 630 Column C5 I I I 26 X 26 104 26.0 l l 4 0 1200 109 21.5 72.4 4510 * Equivalent value of r for a rectangular column was taken as x^ + y^/x + y, where x and y are the column dimensions.
Moe developed h i s equation from a large number of t e s t s on slabs. Additional data was evaluated using r e s u l t s from other investigations. From a t h e o r e t i c a l examination of shear strength i n slabs an equation was derived with constants determined from t e s t data. The equation takes into accovint column s i z e , slab thickness, and t e n s i l e strength of concrete. For a concentrically loaded colimin, shear strength i s given by the expression: V^ = hdyl7[i5(l - 0.075 7 < a ) ] / [ l + 5.25(bdyi7/V^^^J] ( l ) where b = periphery of c r i t i c a l section d = e f f e c t i v e depth f^ = compressive strength of concrete r = length of side of square column or equivalent for rectangular column Vâ, = shear force a t f l e x u r a l f a i l u r e f l e x V = ultimate shear force for concentric load u When both moment and shear are transferred at a slab-column j o i n t , shear strength i s given by: V = V^ - M/r (2) where M = moment transferred to the coliamn V = ultimate shear force for combined moment and shear In Eq. 1, the c r i t i c a l section b xs the perimeter of the column. For rectangular columns, the length r i s equal to the quantity (x^ + y ^ ) / ( x + y ) , where x and y are the dimensions of the column. Sheax strengths computed by Eqs. 1 and 2 are l i s t e d i n Table IX. Parameters used i n the calculations are given in Table X. Moments used i n Eq. 2 are those determined from the frame a n a l y s i s . For Column C^i, no moment was transferred. I n Colimins E3 and Ei*-, moment transferred was only i n the direction of column Lines 3 sxid k, respectively. Moment i n Column C5 was 1- 37
TABLE IX SHEAR STRENGTH OF SLAB Location Shear Strength, kips Test Column Reaction Test No. Frame Analysis t e s t by Moe's Equation ^Moe by ACI's Equation ACI by ACI Beam Equation Beam t e s t Moe t e s t "ACI t e s t Beam Column C4 I C 4a 1130 1150 1420 1420 970 â 0.81 0.81 1.16 1.19 Column E3 IIA 3 E 390 440 800 870 930 460 0.49 0.51 0.42 0.47 0.85 0.96 Column E4 I I B 3 E 390 460 710 770 800 430 0.55 0.60 0.49 0.57 0.91 1.07 Column C5 I I I C 5 700 730 1090 1120 1180 0.64 0.65 0.59 0.62
Test loads on Columns Ck, E3, E4, and C5 at the time of f a i l u r e are l i s t e d i n Table IX. These loads were determined as described below. Although Columns Bk and D5 also punched through, t h e i r f a i l u r e loads are not included in Table IX. F a i l u r e a t Dk took place a f t e r adja- cent Columns Ck and Ek had punched through. S i m i l a r l y f a i l u r e occurred at D5 a f t e r punching had occurred at Column C5. In both cases the structure was damaged extensively when the secondary punching occurred. For these conditions, forces acting at Bk and D5 at failvire would be impossible to assess. Column loads, designated as V^est' include both dead load and applied l i v e load. For each t e s t , the average dead load i n t e n s i t y l i s t e d i n Table IV was used. Also included was weight of loading e'quipment. This amounted to 6, 7, and 10 psf for Tests I , I I , and I I I respect- i v e l y . The nominal dead load car r i e d by a column was defined as that within the rectangular area bounded by the centerlines of panels surrounding the column. ' S i m i l a r l y , nominal applied l i v e load transferred to a column was de- fined as that within the dead load contributory area. The col\amn reac- tion due to l i v e load was determined by multiplying the nominal column load by the r a t i o s of computed to nominal l i s t e d i n Table V I I I . These mult i p l i c a t i o n r a t i o s were determined by the frame a n a l y s i s as described i n an e a r l i e r section of t h i s paper. The r a t i o for Column Ek was assumed to be the same as that for Column E3. Only the raaaimum and minimum multiplication r a t i o s were recorded for Colimin Ck. Added load of 39 kips from settlement was included i n the t e s t load for Col\min E3. Shear strength based on material properties was computed for each punch- ing location by two methods. One method was that developed by Moe. (19) The second procedure was that given by the ACI (3l8-63)(20) and described i n the Commentary on the ACI Code (SP-IO). Ì -ÌÌ ^ 1-35
Based on a n a l y s i s of the frame described above, d i f f e r e n t i a l settlement may have induced an additional load of 39 kips i n Colimm E3. I t must be pointed out that t h i s computation of the e f f e c t of settlement i s only a rough e s t i - mate. The time-dependent nature of settlement and l a c k of s p e c i f i c know- ledge concerning the support i n t e n s i t y and d i s t r i b u t i o n of the s o i l support preclude a precise estimate of the induced reaction. However, t h i s a n a l y s i s shows that the d i f f e r e n t i a l settlement observed coiild place additional load on Column E3. Shear strength woxild c e r t a i n l y be influenced by such an additional column load. SHEAR STRENGTH OF SLAB Shear f a i l u r e occurred i n a l l three t e s t s . Only i n Test I I I was there evidence of flexuraJ. d i s t r e s s ; reinforcement i n the positive moment region began to y i e l d before punching. None of the meas\jred reinforcement s t r a i n s approached y i e l d i n Tests I and I I when the f i r s t shear f a i l u r e occiirred. Maxlmm applied load was s u b s t a n t i a l l y below the f l e x u r a l capacity of the slab i n each t e s t . I n Test I , the punching f a i l u r e occurred a t Colvimn Ck. This column was surrounded by the four loaded panels. The symmetry of loading and geometry of structure resulted in an e s s e n t i a l l y concentric column load. Consequently, there was l i t t l e or no moment t r a n s f e r between the slab and the column. A somewhat d i f f e r e n t shear condition was produced i n Test I I . Here, the slab f a i l e d i n shear at the edge Gol\mins. These edge coliamns were located at the inte r s e c t i o n of two loaded panels. Since the columns had very low f l e x u r a l s t i f f n e s s compared to that of the waffle slab, there was l i t t l e moment tr a n s f e r between the slab and the edge columns. The columns acted primarily as v e r t i c a l props for the slab cantilevered out from column Line D. A t h i r d type of shear condition existed i n Test I I I . I n t h i s t e s t , the i n i t i a l pionching f a i l u r e occurred at Column C5. Only one panel was loaded, and some moment was therefore transferred from the slab to the four support- ing columns. Since only two faces of the columns joined the loaded area, i t would be expected that v e r t i c a l load was transmitted to the columns p r i n c i p a l l y through these two column faces. l-3h
TABLE V I I I COLUMN UNIT LOADS Column ⢠No. Frame Analysis No. Test No. Computed Column Unit Load,* Ib/psf Nominal Column Unit Load,** Ib/psf Computed Nominal C4 C4 C4 4a 4b C I I I 1057 1066 1031 908 908 908 l . l 6 .1.17 1.14 E3 E3 3 E I I I I 383 479 479 479 0.80 1.00 C5 C5 5 c I I I I I I 264 248 228 228 l . l 6 1.09 * From frame analysis ** From area bounded by panel centerlines not known. I n order to evaluate probable e f f e c t s on colvimn loads, i t was assumed that the structure was supported only beneath the two center bays between Lines 3 and 5* For t h i s support condition column loads r e s u l t i n g from settlement can be computed. Only the dead weight of the foundation slab was considered m these calculations since the reaction due to the weight of the roof slab i s calculated i n the same way as that due to l i v e loads. Load i n Column E3 induced by settlement of the structure was estimated by analyzing the frame shown i n Fig. 55. Since the frame i s symmetrical about column Line 4 only one-half of i t i s shown i n F i g . 55. F l e x u r a l s t i f f n e s s e s of the struct\ire along column Line C were used in the computations. Again an uncracked concrete section was used. Uniform load along the foundation was taken equal to the dead weight of the slab. This load was supported e n t i r e l y from below by a uniformly d i s t r i - buted load i n Span 3-4. Because of symmetry, end spans at Line 4 were consi- dered fixed. Dead weight of the 28-in. deep foundation slab was 350 psf. 1- 33
At most locations, there i s good agreement between measured and computed deflections. However, there i s a wide difference between measured and computed deflections along colisnn Line £. Computed deflections substant- iaJ.ly underestimate the measured deflection, and t h i s was also r e f l e c t e d i n the center-of-panel deflections discussed e a r l i e r . The high measured deflections were probably influenced by the diagonaJ. cracking caused by settlement of the structure p r i o r to tes t i n g . Column Loads. An equivalent frame an a l y s i s was also vised to calculate coltmin loads. Loads were determined for columns a t which the slab f a i l e d i n shear during each of the three t e s t s . I n addition, e f f e c t on column loads of settlement of the structure was evaluated. Column loads determined by the frame an a l y s i s for a 1 psf applied l i v e load are l i s t e d i n Table V I I I . These quantities are compared to nominal column reactions. Nominal reactions are based on the assumption that the column supports a loaded area bounded by the centerlines of adjacent panels. Column loads based on t h i s assianption would e x i s t i n the i n t e r i o r of a multi- panel slab subjected to a uniform load. The colvnnns chosen for comparison are those where shear punching f i r s t occurred i n each of the three t e s t s . I t i s seen that for Columns Ck and C5 there i s l i t t l e difference i n the computed column loads using different frames. For both columns, computed values Eire somewhat greater than nominal column loads. The r a t i o of com- puted to nominal loads ranges from I . 0 9 to I . 1 6 for Column C5 and l.lk to 1.17 for Coliann Ck. As expected load computed for Column E3 i s somewhat l e s s than the nominal value. As shown previously i n F i g . 15, considerable d i f f e r e n t i a l settlement seemed to have occurred each side of column Line h. Maximum settlement was at column Lines 1 and 7- Prom contours of deflection, l i t t l e d i f f e r e n t i a l settlement was noted along Column Line k. Ihe structure appeared to be supported mainly i n t h i s i n t e r i o r area. History of t h i s settlement i s 1- 32
TABLE V I I COMPUTED DEFLECTIONS AT 300 psf APPLIED LOAD Test Panel Deflection i n Inches No. No. Frame Cantilever Plates Total I 3.4- B,C 4.5- B,C 3.4- C,D 4.5- C,D 0.170 0.170 0.170 0.170 0.065 0.068 0.061 0.068 0.031 0.031 0.029 0.029 0.266 0.269 0.260 0.267 I I 2.3- D,E 3.4- D,E 4.5- D,E 0.113 0.048 0.090 0.170 0.170 0.170 0.035 0.034 0.034 0.318 0.253 0.294 I I I 5,6-C,D 0.176 0.090 0.031 0.297 Measured and computed deflections at the center of the panels under one l i v e load are compared m Fi g . 53 for a l l three t e s t s . I n general, agree- ment between measured and calculated values i s s a t i s f a c t o r y . Computed values are about the same f o r each of the four panels i n Test I . As noted e a r l i e r , measured deflections In Panels 4,5-B,C and 3,4-C,D were about equal, while deflection measured m Panel 3,4-B,C was higher and that In 4,5-C,D was lower. Computed deflections agree c l o s e l y with those recorded m the t e s t for Panel 4,5-B,C. Computed deflections are somewhat l e s s than those mea- sured m Test I I but somewhat more than those measured in Test I I I . In the a n a l y t i c a l procedure, the equivalent frame I s taken along column l i n e s In the long span direction of the panels. Although the a n a l y t i c a l procedure was developed to obtain deflections at the center of a loaded panel, i t I s i n t e r e s t i n g to compare measiared and computed deflections at midspan of the column l i n e s . This comparison I s shown In F i g . 54. Deflec- tions at an applied load of 350 psf (added dead plus about 1.0 l i v e load) are l i s t e d f o r a l l three t e s t s . Deflections along Line D for Test I I are l i s t e d below column Line D. Deflections for Tests I and I I I along Line D are shown above Line D. 1-31
ANALYSIS OF RESULTS EQUIVALEHT FRAME ANALYSIS Analyses of the structure were made using two di f f e r e n t equivalent frame methods. One was used to compute deflections a t the center of the loaded panels for each t e s t . The second a n a l y s i s was applied to compute column loads for each of the three t e s t s , and also column loads resti l t i n g from settlement of the structure. In both methods of analysis the three-dimensional structure was f i r s t re- duced to an equivalent two-dimensional frame. Moments and shears from the applied loads were then obtained for the "ersatz" frame by use of the Cross moment d i s t r i b u t i o n procediire. (â¢'â¢Ì^ B r i e f descriptions of the analy- ses are presented i n Appendix A. Detailed descriptions of the procedures are given i n other references.(I5*l6,17*l8) Deflections. Since the structure was observed to be cracked a t the time of t e s t , deflections were computed on the ba s i s of a cracked section. Deflections were f i r s t computed using the uncracked f l e x u r a l s t i f f n e s s of the slab. These deflections were multiplied by the r a t i o of the s t i f f n e s s of the uncracked section to that of the cracked section. For t y p i c a l sections, t h i s r a t i o was equal to 2.J. Deflections of the frame, cantilever and plate computed as described i n Appendix A are l i s t e d in Table V I I . These deflections are for an applied load equal to 350 psf (added dead plus about 1.0 l i v e load). Values for the can t i l e v e r deflection include e f f e c t s of rotation of the supported end. Quantities l i s t e d for the frames and ca n t i l e v e r s are the average of the two frames bounding the loaded span. For the panels i n Test I and I I I , deflection i s p r i n c i p a l l y from deformation of the frame. In Test I I , deflection comes primarily from the cantilever beam. This indicates the contribution of rotation of the edge beam. 1-30
was simply pushed down since the columns offered very l i t t l e support. F i g . 50, taken at the end of Test I I I , shows that the slab was pushed down nearly a foot at Colvmin D5. Major cracks on top of the slah a f t e r Test I I I are shown i n Fig. 51. A l l of the cracks are associated with negative moments induced during the t e s t . Except for the cracks at Columns C6 and D6, crack locations were very near cut off points f or negative moment reinforcement. Crack- ing on the bottom of the slab a f t e r the t e s t i s seen i n the composite photo, Fig. 52. Extensive cracking i s v i s i b l e i n the slab area near Columns C5 and D5. The coliamn s t r i p between Columns C5 and D5 was severely cracked also. These cracks were formed a f t e r the shear f a i l u r e at Columns C5 and D5. 1-29
that negative moment reinforcement over the column was pulled from the slab. The close view of Colimin C5 i n F i g . kQ shows that some of the reinforcement was spliced d i r e c t l y over the column. This was not observed a t any other location. Load-strain data f o r the f i n a l stage of Test 111 are given i n Appendix B. At most gage locations, reinforcement s t r a i n s were proportional to load u n t i l applied load was about 1000 psf. Beyond t h i s load i n t e n s i t y , s t r a i n s increased more rapidly as load was increased. Yielding of reinforcement i n the positive moment region of the slab was f i r s t detected at Gage No. 71 under an applied load of about 15OO psf. This gage was mounted on the bottom reinforcement of the column s t r i p along column Line 5 as indicated i n F i g . 24. At an applied load of about 1900 psf. Gages 72 and 73 a l s o indicated reinforcement y i e l d . Just before shear f a i l u r e at Column C5, Gage jk indicated that y i e l d i n g was imminent. None of the s t r a i n s measured i n the top layer of reinforcement indicated y i e l d i n g of the negative moment reinforcement. After slab punching occurred at Column C5, load was dropped to zero. Load was again applied a f t e r the slab had been examined. At 14^9 psf applied load (added dead plus 4.6lO l i v e l o a d s ) , the slab f a i l e d i n shear a t Column D5. Applied load immediately dropped to about 6kh psf (added dead plus 2.0 l i v e loads). This was the maximum load that could be maintained a f t e r punching at Columns C5 and D5. F i g . 49 indicates deflections under an applied load of 644 psf Just be- fore load was removed and the t e s t was ended. The pattern of deflections shows the t e s t panel was s t i l l i n t a c t along column Line 6. Although load was applied only to Panel 5,6-C,D, substantial downward deflections were measured i n the previously tested panels adjacent to the axea of Test I I I . These deflections occurred as the r e s u l t of punching a t Columns C5 and D5 and at columns along Line 4. In the aorea along Lines 4 and 5, the slab 1-28
and 6. The curves are very nearly l i n e a r for loading up to 6hh psf. Resi- dual deflection measured when applied load was released a f t e r the added dead load plus one, plus 1-3, and plus two l i v e loads were nearly zero. Strains measured by gages on the concrete also showed l i n e a r response to load up to 6kk psf applied load. Load-strain data are presented i n F i g . 45 f o r gages on the bottom of the slab adjacent to Columns C3, D5, and D6. Measured s t r a i n s shown i n F i g . U5 did not exceed 0.0002 at a load of 6hh psf. Load-strain data for gages mounted on reinforce- ment in the loaded panel are given i n Appendix B. Some of the gages on the top la y e r of reinforcement showed e r r a t i c behavior. However, gages on the bottom reinforcement responded as expected. S t r a i n s i n the bot- tom reinforcement at the center of the panel were about O.OOOU for an applied load of 6kh psf. F i r s t s i g n i f i c a n t cracking was observed at an applied load of 10U6 psf. Cracking was seen on the top of the slab at Column D6. Similar cracking was observed at a l l the columns when loading reached 1U89 psf. These cracks were located i n the loaded panel a few inches inside the bound- ary of the colxanns. Negative moment cracking was v i s i b l e i n Panel U,5-C,D at lk89 psf applied load. This cracking was si m i l a r to that observed during Test I . The crack was p a r a l l e l to the edge of the loaded panel and was located where negative moment reinforcement ended. Crack width was about 0.125 m. under an applied load of 1730 psf. Deflections measured at an applied load of 1730 psf (added dead plus about 5.6 l i v e loads) are given m F i g . k6. At the center of the loaded panel, deflection was over 2 i n . There was also a downward deflection at the center of adjacent Panels 5^6-B,C and 5,6-D,E, while the three panels bounded by column Lines U, 5, B, and E, were deflected upward. A punching f a i l u r e occurred a t Column C5 at an applied load of 1972 psf. This load i s equal to added dead plus 6.4 l i v e loads. A view of the f a i l u r e area at the top of the slab i s shown i n Fig. k j . I t i s seen 1-27
reached, a shear f a i l t i r e occurred at Column Eh. A photograph of the f a i l u r e zone i s shown i n F i g . ko. Load at f a i l i i r e was equal to added dead plus about 2.3 l i v e loads. The load-deflection data i n Appendix B show greatly increased deformation as load i n t e n s i t y increased be- yond 607 psf. A f t e r f a i l u r e , i t was possible to maintain a load i n t e n s i t y of ^66 psf. The slab was deflected f a r t h e r u n t i l a secondary shear f a i l u r e occurred at Column Dl* under an applied load of 59^ psf. To prevent damage i n the Test I I I area. Part B of Test I I was terminated when Column Bk punched through. Deflections at the conclusion of Test I I are shown i n F i g . hi. The r e s i d u a l deflections indicate considerable separa- t i o n of the slab from the columns both along the edge of the building and at Column Dk. Cracking on the underside of the slab a f t e r the t e s t i s shown i n F i g . 42. Test I I I . The f i n a l t e s t i n the s e r i e s c a r r i e d out on the waffle slab consisted of loading Panel 5^6-C,D. This panel was the second bay i n from the nearest edge of the building. From previous t e s t s , the four columns along column Line k from Line B to Line E suffered damage from shear. In Tests I and I I , i t was observed that e f f e c t s of loading were small outside the loaded areas. Consequently, i t was not expected that the condition of the slab along column Line k- would have a major influence on the re- s \ a t s of Test I I I . Deflections at an applied load of 3^5 psf (added dead plus 1.0 l i v e load) are shown i n F i g . 43. At the center of the t e s t panel, deflection was 0.26 i n . Deflection of the slab at midspan of column Lines C and D was about 0.18 i n . , twice that measured at midspan of the boundaries at column Lines 5 and 6. Load deflection relationships for the t e s t panel are shown i n F i g . kh for center of the panel and at midspan of the two column s t r i p s along Lines 5 1-26
i n s t r a i n at t h i s load. Load-strain curves for Gages 66 and 68 shown i n F i g . 37 are t y p i c a l . (Location of the gages i s shown in Fi g . 24.) However, Gages 42 and 37 continued to indicate increased s t r a i n . These gages were on negative moment reinforcement across Line D. Since decrease i n s t r a i n was exhibited by gages in both positive and negative moment, regions, the pattern of changing s t r a i n indicated r e d i s t r i b u t i o n of moments. I t i s evident that i n i t i a t i o n of the shear f a i l u r e at Column E3 reduced the column reaction. This increased the amount of negative moment i n the slab along column Line D. Shear f a i l u r e a t Column E3 occurred when the applied load reached 607 psf. This load i n t e n s i t y i s equal to added dead plus about 1-9 l i v e loads. A view of the failiare zone at Column E3 a f t e r the t e s t i s shown i n F i g . 38. The diagonal cracks extended from the junction of the column and edge beam to the top of the slab at an angle measured to be about 35° from horizontal. Deflections a f t e r f a i l u r e had occurred and the load was removed are shown in Fi g . 39- I * i s apparent that r e s i d u a l deflections were sub- s t a n t i a l . Part B of Test I I was conducted to evaluate shear strength at Column E4. This column was sim i l a r i n location and in dimensions to Colimin E3. Colimn E3 was 12 x 32 i n . i n plan, while Column E4 was 12 x 24 i n . Deflections and st r a i n s were measured, but t h i s information i s used only for q u a l i t a t i v e purposes. The proximity of the f a i l u r e at Column E3 precludes t h e i r use as a true indication of behavior of a monolithic reinforced concrete structure. Load was applied i n one increment to 607 psf, the load at which f a i l u r e occurred at Column E3 i n Part A. Load was then Increased i n increments of about 40 psf. When an applied load i n t e n s i t y of about 728 psf was 1-25
Test I I . Test I I was divided i n t o two parts. I n Part A, the three panels along the edge of the b u i l d i n g between column Lines 2 and 5 were loaded. This part of the t e s t ended with a shear f a i l u r e at Column E 3 . I n Part B, Panel 4,5-D,E and the adjacent h a l f of Panel 3,4-D,E were loaded. Part B was tenninated a f t e r the slab f a i l e d i n shear at Column Elt-. Part A of Test I I l e d t o a deflection of the waffle slab at 3^7 psf applied load (added dead plus 1 . 0 l i v e load) as shown i n Fig. 3 5 . I t can be seen t h a t deformation outside the loaded panels was small. Along the middle of the slab between Lines D and E, deflection at midspan between columns and at panel centers was about equal; deflec- tions at the centers of loaded panels were only s l i g h t l y higher than those at column s t r i p midspan. Such a defle c t i o n pattern indicates e s s e n t i a l l y one-way action of the loaded panels. Deflections of the edge beam along column Line E were about twice those of the column s t r i p s along column Line D. The edge beam was 3 0 . 3 7 5 - i n * deep, 6 - 3 7 5 i n . deeper than the slab, and had a width of 4 5 i n . This extra depth was not s u f f i c i e n t t o compensate f o r the smaller e f f e c t i v e width of the slab along column Line E. Load-deflection curves were l i n e a r u n t i l load exceeded 4 9 9 psf (added dead plus approximately 1 . 5 l i v e loads). Cvirves f o r def l e c t i o n at the middle of the loaded panels are shown i n Fig. 3 6 . A f t e r t h i s load stage, deflections i n Panels 2,3-D,E and 3,U-D,E increased s i g n i f i c - a n t l y with load. I n p a r t i c u l a r , load-deflection curves became nearly- horizontal f o r Locations 5 1 and 5 9 i n Fig. 2 2 as loading approached added dead plus two l i v e loads. Strain records also indicated distress at Column E 3 . Load-strain relationships changed markedly at an applied load of 56O psf (added dead plus I . 7 l i v e loads). Gages i n Panel 3,4-D,E showed a decrease 1 - 2h
punching occurred and d i f f e r s i g n i f i c a n t l y ; the four-panel area was "dished" with maximum deflections at Column Cl*-. I n the v i c i n i t y of Column Cl*-, f l e x u r a l cracks on the bottom of the slab formed a f t e r punching. These cracks are v i s i b l e i n the composite photograph i n Fig. 32. A f t e r f a i l u r e , t h i s region was subjected t o positive moments with the bottom of the slab i n tension. Positive moment reinforcement i n the slab was not continuous across the column l i n e s . Thus, no reinforcement was located i n the bottom tension zone of the slab, and moment resistance was nearly zero. Loading was continued a f t e r the i n i t i a l shear f a i l v i r e . Load carrying capacity of the structure became less as deflections increased. Load- deflection curves shown i n Fig. 28 I l l u s t r a t e t h i s behavior. The t e s t was terminated when the load necessary t o increase deflections was re- duced t o hjk psf. Spalling of concrete i n compression on the south face of Colimin Bl*- was observed j u s t before the t e s t was stopped. Spalling occurred i n the column concrete a t the bottom face of the roof slab. On the mezzanine side of Column B U, combined torsion-flexure cracking was v i s i b l e on the v e r t i c a l face of the slab where the'depth decreased from 2h to 8 i n . This cracking i s shown i n Fig. 33. During the t e s t , both before and a f t e r shear f a i l u r e at Column Ch, negative moment cracks were observed on top of the slab outside the t e s t area. These cracks circijmscrlbed the four panels and were located about 10 f t beyond the loaded area. The cracks were i n the v i c i n i t y where negative reinforcement over the colimn l i n e s was cut o f f . An outline of cracking on the top surface of the slab I s shown i n Fig. 3I*. Crack widths were not measured but were on the order of one-eight I n . wide j u s t a f t e r punching occurred. A f t e r the t e s t was complete and a l l load was removed, the cracks closed somewhat, though not more than about one- ha l f the maximum width observed during the t e s t s . 1- 23
Measured deflections at selected locations under 3k6 psf applied load (added dead plus 1.0 l i v e load) are shown i n Fig. 29. I t can be seen th a t deflections at the centers of the loaded panels were not the same. Maximum panel deflection was 0.3^ i n . at the center of Panel 3,4-B,C, over twice the 0.15 i n . deflection of Panel 4,5-C,D. Panels 4,5-B,C and 3,4-C,D each deflected about 0.25 i n . Relative magnitude of these panel deflections remained the same u n t i l the end of the t e s t . Deflec- t i o n s outside the loaded area were neglible. Load-strain relationships measured f o r the reinforcement were i n f l u - enced by cracking of the concrete. Gages located i n areas where cracks were present p r i o r to the t e s t showed a s t r a i g h t - l i n e relationship between load and s t r a i n . At gage locations where cracks formed during the t e s t , strains increased sharply with load when the concrete cracked. Strains measured i n the concrete and reinforcement were small under added dead plus one l i v e load. Changes i n reinforcement s t r a i n due t o applied load did not exceed 0.0003 at any location. Gage No. 63 mea- sured the highest steel s t r a i n under t h i s load, O.OOO29O tension. This gage was located on the positive moment stee l at midspan of the column s t r i p between Panels 3,l4-C,D and 4,5-C,D as shown i n Fig. 2k. Gage 60, si m i l a r l y located i n the positive moment region of the column s t r i p between Panels 3,l4-B,C and 3,lt-C,D showed a s t r a i n of O.OOO26O tension. Strains i n the reinforcement at the middle of Panel 3,4-C,D were about one-fourth the values recorded by Gages 60 and 63. A maximum concrete s t r a i n of 0.0004 compression at added dead plus one l i v e load was measured f o r Gage 5 on the bottom of the slab adjacent to Column Ck. The ultimate strength was reached abruptly by punching shear at Column Ck under an applied load of 895 psf. This load i n t e n s i t y i s equal t o added dead plus about 2.8 l i v e loads. Fig. 30 shows the pattern of de- f l e c t i o n s at an applied load of 8^3 psf, the load stage j u s t before punching occurred. Deflections given i n Fig. 31 are those j u s t a f t e r 1-22
TEST RESULTS TV.ST DATA Complete t e s t data are presented i n Appendices B and C. Only the p r i n - c i p a l data are summarized i n the t e x t of t h i s paper. L i s t s of tables and figures precede the Appendices, the major categories of information pre- sented being: 1. Details of slab construction 2. Properties of reinforcement 3. Program of loading 4. Loading system hydraulic pressure 5. Ram ca l i b r a t i o n 6. Measured ram loads 7. Measured deflections 8. Measured strains 9. Slab properties at f a i l u r e areas BEHAVIOR Added dead plus one l i v e load was applied t o permit observation of service load behavior. Response to overload was then observed, and the ultimate strength of the loaded portion of the structure was deter- mined. F i n a l l y , a f t e r i n i t i a l f a i l u r e , loads were reapplied to evalu- ate load carrying capacity. Test I . The four panels i n Test I responded i n an essentially e l a s t i c manner under the applied load u n t i l ultimate strength was reached when Colvimn C4 punched through. Load-deflection relationships, as shown i n Fig. 28, did not deviate s i g n i f i c a n t l y from a stra i g h t l i n e . Some re- sidual deflection remained each time a f t e r the loads were removed. How- ever, residual deflections became smaller with each load cycle. 1- 21
been removed from the structure. This added dead load was held while readings were taken on a l l instnmients. An additional 300 psf (l.O l i v e load) was then applied i n s i x increments. At t h i s stage, t o t a l applied load was equal to that assumed f o r design.* Each of the f i r s t two increments were equal to about 0.25 of the l i v e load, the l a s t four to about 0.125 l i v e load. Readings of a l l instruments were taken a f t e r each increment was applied. A f t e r data were collected at an applied load of 3^1 psf, load was reduced to zero and another set of readings was obtained. Application of added dead load plus 1.5 l i v e loads followed the previ- ous loading sequence. The hf-psf added dead load was applied as before i n one increment, with s i x increments again used t o reach 1.5 l i v e loads. I n t h i s case, each of the f i r s t two increments was equal 0 .5 l i v e load, the next four to 0.125 l i v e load. Applied load was then released, and another set of "zero applied load" readings was obtained. Next, added dead load plus two l i v e loads was applied. Loading f o r the 47-psf design dead load was carried out as described previously. This was followed by three increments of 0 .5 l i v e load each, and fovir of 0.125 l i v e load. Readings were taken a f t e r each increment and a f t e r load was released. F i n a l l y , added dead load plus four increments of 0 .5 l i v e load each were applied. A f t e r an i n t e n s i t y of two l i v e load was reached, loading was continued u n t i l f a i l u r e occurred. Load increments were chosen so that f a i l u r e was reached i n a convenient nvimber of steps. After f a i l u r e had occurred, hydraulic pressure was again applied t o the rams t o detennine the resistance of the struct\ire t o additional deformation. * Total dead load i s equal to 4? psf plus the slab dead loads l i s t e d i n Table IV. 1- 20
ways. Each pump had an i n t e g r a l Bourdon type d i a l indicator pressure gage. These gages were used i n c o n t r o l l i n g pressure during the t e s t s . Bourdon-type s t r i p chart recorders were placed i n the hydraulic l i n e s from each pump. These recorders were used p r i m a r i l y t o check pres- sure levels and determine pressure f l u c t u a t i o n during hold periods of loading. A t h i r d check of pressure l e v e l was provided by a pressure c e l l placed i n the hydraulic system. PressTore i n the c e l l was sensed by e l e c t r i c a l resistance s t r a i n gages and a s t r a i n indicator. A l l equipment, the rams, load c e l l s , and pressure indicators, were c a l i - brated at the PCA Laboratories before and a f t e r the t e s t s . Detailed information concerning the equipment and i t s c a l i b r a t i o n i s given elsewhere.(9) Two methods were used t o determine thickness of the t e s t slab. Level sightings above and below the slab were taken at r i b intersections and at i n t e r v a l s along the edge beam on column Line E. These readings were referenced t o sightings taken at a core hole d r i l l e d through the slab. A d i r e c t measurement of slab thickness was made at the hole. I t was then possible t o determine slab thicknesses from the l e v e l readings. At each hole th a t was d r i l l e d through the roof at a load point, a specially constructed caliper was used t o measure slab thickness. This \ i n i t incorporated a 0.001-in. d i a l gage and was capable of mea- suring depths i n the 8 i n . and 2k i n . t h i c k slab regions. Further d e t a i l s of t h i s caliper are given i n Ref. 9- Average thickness of the panels i n the r i b areas and m the s o l i d areas surrounding the columns are shown m Fig. 27; depth of the edge beam along coliamn Line K i s also included. CONDUCT OF TESTS Afte r a l l equipment and instrumentation was i n place and had been checked f o r proper operation, an I n i t i a l set of "zero applied load" readings was taken. Load i n t e n s i t y of l*-7 psf was then applied t o equal the load f o r the flagstone walks and waterproofing that had 1- 19
elsewhere.(9*13) uvo levels located on the mezzanine were used t o observe deflections from the underside of the roof. These instruments sighted on targets located at the middle of each panel and at midspan of the column l i n e s . A t h i r d l e v e l was used t o record movement of targets mounted on top of the roof at the centers of columns i n the t e s t areas. A l l levels were referenced t o a standard mounted outside the structure. Designation and location of a l l targets used i n the three t e s t s are given i n Fig. 22. Those read during each t e s t are l i s t e d i n Table V I i Most locations f o r measviring deflections were common to a l l three t e s t s ; how- ever, locations remote from the loaded area of a p a r t i c u l a r t e s t were not monitored. E l e c t r i c a l resistance s t r a i n gages were used t o obtain strains i n r e i n - forcement and i n the concrete at selected locations. ManuaJ-ly operated switch boxes and portable s t r a i n indicators were used t o monitor the gages. Table VI also l i s t s the s t r a i n gages monitored during each t e s t . Designation and location of these gages are shown i n Fig. 23 and 2h. A number of gages i n Test I and Test I I areas were read during both t e s t s . Gages i n the Test I I I area were monitored only during Test I I I . A l l gages were waterproofed t o reduce ef f e c t s of moisture on indicated strains. Typical gage i n s t a l l a t i o n s on top reinforcement of the slab are shown i n Fig. 25. Each active gage was provided with a temperature compensating gage mounted on a separate, unstressed piece of steel or concrete. The compensating gages were placed near t h e i r companion active gage dviring the tests. I n addition t o the gages on the slab, one gage was mounted on the concrete on each face of Column Ch. S i m i l a r l y a gage was placed j u s t below the edge beam on the reinforcement of Coltmin E 3 . Hydraulic pressure was used as the primary means of detexmining applied load. For f u r t h e r control six load c e l l s were placed i n the t e s t system t o measure d i r e c t l y the applied load at selected points. Locations of the load c e l l s i n each t e s t are shown i n Fig. 26. They were read by a portable s t r a i n indicator. Ì -Ì ^Ì Pressure was meas\ired i n three independent 1-18
TABLE V ARRANGEMENT OF HYDRAULIC LOADING EQUIFMENT DURING TESTS Test Load Number Panel Area Remarks No. Number of Rams Per Ram, Per Test Ft^ For a l l of Test: 1-T kS 75-68 Ram pressure i n panels 3,4-B,C I 8-19 10k 34.93 and 3,4-C,D controlled by pump 1. 20-28 208 17.49 Ram pressure i n panels 4,5-B,C and 4,5-C,D controlled by pirnip 2. For f i r s t part of t e s t : Ram pressure i n panel 2,3-D,E and one-half panel 3,4-D,E con- t r o l l e d by pump 1. 1-7 36 79-55 Ram pressure i n panel 4,5-D,E I I 8-19 8k 34.09 and one-half panel 3,4-D,E con- t r o l l e d by pump 2. 20-27 k2 34.09 For second part of t e s t : Ram pressure i n panel 4,5-D,E and one-half panel 3,4-D,E con- t r o l l e d by pimp 1. 1-7 Ik 65.27 For a l l of t e s t : I I I 8-20 ko 22.85 Ram pressure i n panel 5Ì 6-C,D 21-38 80 11.42 controlled by p\jmp 1. TABLE V I ACTIVE DEFLECTION LOCATIONS AND STRAIN GAGES Test No. Deflection Location No. Strain Gage No. I 1-6, 9-11, 17-42, 44-47, 59, 60 50-56, 1-13, 20, 21, 30-42, 60-66, 68 I I 8-11, 13-16, 22, 24, 26, 3k-k6, 48-56, 58-60 30-32, 8-11, 14-15, 21, 35-38, 42-45, 61, 62, 64, 66-70 I I I 6, 7, 11, 12, 20, 26-29, 33, 38-43, 46, 47, 54-57, 32, 61 16-19, 46-51, 71-74 1-17
Two e l e c t r i c a l l y driven pumps were used to supply hydraulic pressure. (9) Low load i n t e n s i t i e s f o r each t e s t were applied with only about 20 per- cent of the t o t a l ntanber of rams connected. As higher i n t e n s i t i e s of load were required, additional rams were activated. This procedure was used so t h a t , f o r any load i n t e n s i t y , hydraulic pressure was at a l e v e l high enough t o reduce the importance of ram f r i c t i o n . For each arrange- ment, active rams were evenly spaced over the t e s t area as indicated i n Fig. I T , 18, and 19. Table V l i s t s pertinent data concerning the load- ing arrangement. INSTRUMENTATION The condition of the structure before and a f t e r each t e s t was recorded by photographs. Before each t e s t , cracks v i s i b l e i n the t e s t area on the underside of the waffle slab were marked with white chalk. A com- posite photograph of the t e s t area was then made before t e s t loads were applied. Several composite photographs were also taken during the t e s t s ; however, new, unmarked cracks were not easily discernible. A f t e r each t e s t was completed, cracks were again marked with chalk and a f i n a l composite photograph was taken. Both s t i l l and movie cameras were used t o photograph s t r u c t u r a l damage a f t e r the tests. During the t e s t s , 16 mm and 35 mm time-lapse cameras were focused on c r i t i c a l regions. In Test I , two cameras were trained on the underside of the roof slab at Column Ck where punching occurred. In Test I I , cameras were used t o record behavior of the structure i n the areas about Colvmins E3 and El*-. An o v e r a l l view of the loaded panel i n Test I I I was taken from the top of the roof slab. For these detailed observations, cameras were operated during the time load was being increased and f o r a short period a f t e r the higher load stage was reached. Deflections of the roof slab were measiored using precision levels sighting on targets attached t o the slab. A description of these t a r - gets and of the instruments and techniques used t o read them i s given 1- 16
of the waterproofing membrane. Before any tests were conducted, photo- graphs were taken of a l l walls and columns to record t h e i r condition. I n Test I , computed shear strengths of the roof and foundation slabs were about the same. To provide an adequate margin against a shear f a i l u r e of the foundation slab, a reinforced concrete sheaxhead was cast at the base of Coliann Ch. Details of the shearhead are shown i n Fig. 20. Nominal concrete strength was ItOOO psi at seven days. The sequence of work i n preparation f o r each t e s t was similar. Detailed information describing techniques and special equipment i s given else- where . (9) Holes were d r i l l e d i n t o the foundation and through the roof slab at each of the load points. Holes through the roof were 1-in. dia- meter. Next, the slab thickness was measured, instrumentation i n s t a l l e d , and cracks mapped i n the t e s t area. F i n a l l y , the loading equipment was put i n place. After the t e s t was completed, loading equipment was re- moved and cracks were again mapped i n the t e s t area. Photographs were then taken to record the condition of the structure a f t e r the t e s t . LOADING SYSTEM Hydraulic rams were used to apply load to the structure. To approximate uniform load on the t e s t panels the rams were evenly d i s t r i b u t e d over the t e s t panels. Nominal spacing of the loads f o r each t e s t i s given j n Fig. 17, l8, and I9. Individual rams were moved s l i g h t l y o f f these grids so that holes d r i l l e d through the roof slab would not h i t major reinforce- ment or pass through the r i b s . A schematic drawing of the loading equipment i s shown i n Fig. 21; de- t a i l e d information i s provided elsevrhere.^9) On top of the waffle slab, a hydraulic center-hole ram was placed over a steel load d i s t r i b u t i o n plate centered on a hole d r i l l e d through the roof. One end of a 7-wire prestressing strand was coupled to the ram and passed through the roof slab. The other end was coupled to a rock anchor set m the foundation. As hydraulic pressure was applied, the ram reacted against the upper strand g r i p and pushed down on the roof slab. 1-15
STRUCTURAL TESTS OUTLINE OF TESTS Three load t e s t s were conducted on separate areas of the waffle slab roof. Locations of loaded panels f o r each of the three te s t s are i n - dicated i n Fig. 17, 18, and I9. The t e s t numbers indicate the chrono- l o g i c a l order of tes t i n g . I n a l l t e s t s , loads were applied through concentrated reactions evenly d i s t r i b u t e d over the panels t o approximate a uniform load. Test I covered the four panels withi n column Lines 3,5-B,D. This t e s t was intended t o determine the shear strength of the slab at Column Ck. A l l four panels were loaded during the t e s t . I n Test I I , the slab- column connections at ex t e r i o r panels were investigated. The loaded area i n Test I I included the three panels along the south edge of the bui l d i n g between column Lines 2 and 5 and column Lines D and E. A l l three panels were loaded during the f i r s t phase of the t e s t . I n the second phase of Test I I , Panel i^,5-D,E and the adjacent h a l f of Panel 3,if-D,E were loaded. I n Test I I I , only the single Panel 5,6-C,D was loaded. This loading was intended t o show the significance of i n - plane forces on a panel f a i l i n g i n flexure. A f t e r i n i t i a l s t r u c t u r a l f a i l u r e occurred i n each t e s t , loading was continued t o determine reserve strength. I n Tests I and I I , the post- f a i l u r e loading was terminated when i t appeared that f u r t h e r loading would damage the structure i n the area of the next t e s t . TEST PREPARATION The structure was stripped t o the concrete both inside and out. Buildings and the flagstone pedestrian walks were cleared from the top of the slab. V/here possible, the decorative facing was removed from the outside walls. Paneling and a l l f i x t u r e s were removed from the i n t e r i o r of the building. On the roof, the t e s t areas were fur t h e r cleaned to remove what remained 1- Ik
Between column Lines 2 and 3, 3 and h, and U,and 5Ì intermediate reinforced concrete columns were cast t o support the a r c h i t e c t u r a l facing along the south wall. Locations of these columns i s shown i n Fig. k. Although they were not intended t o be part of the s t r u c t u r a l system they were cast mono- l i t h i c a l l y with the edge heani, consequently, these coltimns acted to support the roof slab. Since none of these columns between the p r i n c i p a l supports was considered i n the s t r u c t u r a l design, no negative moment reinforcement was added above them i n the edge beams or i n the slab p a r a l l e l t o the beams and cracks developed i n the slab. Before t e s t i n g , the columns were cut away from the edge beams. 1- 13
Cracks were observed throughout the roof area running, i n general, i n a di r e c t i o n p a r a l l e l to the numbered column l i n e s . A f l a t black paint covered the underside of the waffle slab. Consequently, cracks were re a d i l y v i s i b l e . White stains from leaching through the slab accented some of these cracks and indicated t h a t the cracks penetrated through the 8-in. slab thickness above the pan openings. OveraJl crack patterns i n the t e s t area as w e l l as a close- up view of t y p i c a l cracks are shown i n Fig. 11. Cracking i n the i n d i v i d u a l t e s t areas viewed from the underside of the slab i s shown i n Fig. 12, 13, and 14. Although a portion of the observed cracking could be a t t r i b u t e d t o shrinkage of the concrete, both the extent of cracking and the crack patterns suggested another cause. Flexural cracks caused by loads on top of the waffle slab were discounted as the cause since many of the cracks occvirred where the con- crete would be i n compression from such loading. Further investigation indicated that cracking of the waffle slab probably was caused p r i m a r i l y by settlement of the structure. From l e v e l sightings, con- tours of defle c t i o n were drawn as shown i n Fig. 15. A point at the north- center of Panel 4,5-B,C was taken as the zero reference f o r deflections. Assuming that the roof slab was o r i g i n a l l y l e v e l , i t i s seen th a t settlement may have exceeded seven inches at the southwest comer of the structure. Further consequence of t h i s settlement was observed along the south face of the b u ilding. The wall between column Lines 1 and 2 was severely cracked diagonally as shown i n Fig. l6. Similar diagonal cracks were noted i n the edge beam along column Line E between Lines 2 and k. This cracking probably had considerable influence on the results of Test I I , as w i l l be discussed l a t e r . During construction of the b u i l d i n g , a very soft material was encountered under the southwest portion of the structure. I n addition, p i l e s that were used t o support a structure t h a t had previously occupied the s i t e of the "Rathskeller" were found i n the central portion of the foundation excavation. These p i l e s were cut o f f at about the l e v e l of the bottom of the foundation slab. Such foundation conditions should be expected t o produce a s e t t l e - ment pattern s i m i l a r to that indicated i n Fig. 15. 1- 12
The e n t i r e roof was designed f o r a l i v e load of 300 psf as required by the New York City Building Code f o r sidewalks; combined dead and l i v e loads from structures atop the roof did not exceed t h i s value. Dead loads f o r each t e s t panel--computed using a u n i t weight f o r concrete of 150 pcf and mea- surements of overall slab depth--are l i s t e d i n Table IV; the average com- puted dead load was about 220 psf. A note on the plans f o r t h i s . b u i l d i n g indicated i t was designed f o r a t o t a l load, l i v e plus dead, of 513 psf. TABLE IV AVERAGE DEAD LOAD OF PAMELS Panel 3,U-B,C U,5-B,C 3,U-C,D U,5-C,D 2,3-D,E 3,i^-I',E U,5-D,E 5,6-C,D Depth of Area of Average Slab, Panel, Dead Load f t f t ^ psf 1.89 923 207 1.92 923 212 1.96 893 227 1.9^ 893 22U 1.95 969 22U 2.02 9^7 233 1.99 9^7 228 1.95 91U 220 The b u i l d i n g was designed t o meet the provisions of the 1956 ACI Building Code.Ì -'-Ì ^ The "Elastic Analysis" outlined i n Chapter 10 of that Code was used to determine design moments and shears. CONDITION PRIOR TO TESTING In preparation f o r tests of the structure, the frame buildings and the f l a g - stone walks were removed from the roof of the "Rathskeller." Heavy con- struction equipment was used f o r these operations. During the removal of debris, material was heaped over r e l a t i v e l y small areas of the structure. Load i n t e n s i t y could not be estimated and could l i k e l y have been m excess of previous superimposed loads. Thus i t was not surprising t o f i n d p r i o r to t e s t i n g that certain areas of the structijre were severely cracked. An overa l l inspection of the "Rathskeller" was made before the tests were started. In p a r t i c u l a r , cracking of the b u i l d i n g was noted and deviations from s t r u c t u r a l plans were investigated. 1-11
TABLE I I I PROPERTIES OF CONCRETE Specimen Compressive Equivalent Tesnile Specimen Compressive Equivalent Tensile Location Strength 6 X 1 2 -in. Strength,* Location Strength 6 X 1 2 -in. Strength,* of Core, Cylinder p s i of Core, Cylinder p s i p s i Strength, fi p s i Strength, 1 5360 5250 12 1*00 2 52I4O 5390 - 13 â â 390 3 5170 51Ì 20 - Ih 6U9O 5970 450 1* 5020 1Ì 670 - 15 â â 310 5 1Ì 300 1*170 - 16 5190 1*830 - 6 â â U30 17 5420 5850 1*00 7 â â 3*Ì 5 18 5020 5670 1*80 8 6230 6530 370 19 5800 531*0 hio 9 5200 5620 - 20 51*60 5720 - 10 5800 5390 21 5260 5150 - 11 6760 6220 - 22 3620 3620 1*60 23 3260 3520 - * S p l i t Cylinder t e s t
TABLE I I CONCRETE CORE DIMENSIONS AND STRENGTH CORRECTION FACTORS Core No. Height, b, inches Diameter, d, inches ^/d Correction Factor for h/d Correction Factor for Steel Core No. Height, h, inches Diameter, d, inches Correction Factor for h/d Correction Factor for St e e l 1 6.hk 3.73 1.73 0.98 â lU 3.kO h.19 3.73 3.73 0.91 1.12 0.92 â 2 5.00 3.73 1.34 0.95 1.08 15 3.56 3.71 0.96 â â 3 5.62 3.73 1.51 0.97 1.08 16 k.ko 3.73 1.18 0.93 â k k. 50 3.73 1.21 0.93 â 17 2.60 7.67 3.73 3-70 0.70 2.08 â 1.08 5 5.60 3.73 1.50 0.97 â 18 3.39 7.31 3.73 3.73 0.91 1.96 â 1.13 6 3.54 3.74 0.95 â â 19 2.87 k.ok 3-73 3-73 0.77 1.08 0.92 â 7 3,47 3.73 0.93 â â 20 5.83 3.73 1.56 0.97 1.08 8 9 3.74 6.01 7.58 3.73 3.73 3.73 1.00 1.61 2.03 0.97 1.08 1.08 21 22 6.38 2.99 7.64 3.75 3.74 3.74 1.70 0.80 2.04 0.98 â 10 3.80 k.hl 3-72 3.73 1.02 1.18 0.93 â 23 7.62 3.75 2.03 â 1.08 11 k.ok 3.72 1.09 0.92 â 12 3.33 3.68 0.91 â â 13 3.60 3-72 0.97 â â
Concrete cores 1, 2, and 3 were taken from the slab before t e s t i n g started. The remainder of the cores were obtained a f t e r the tests were completed. The rough ends of the cylinders were trimmed by a diamond saw before t e s t - ing. Diameter, height, and height-to-diameter r a t i o of the cylinders a f t e r they were trimmed are l i s t e d i n Table I I . Diameters were approximately 3.75 i n . while heights of the cores ranged from about 2.5 t o 7'5 i n . Cy- linders with a height-to-diameter r a t i o less than one were used f o r s p l i t cylinder t e s t s . A l l others were tested i n compression. Table I I also gives correction factors f o r the e f f e c t s of the height-to- depth r a t i o (â¢'â¢Ì^ and the presence of horizontal r e i n f o r c i n g s t e e l , The "Equivalent Cylinder Strength, f ^ " l i s t e d i n Table I I I i s the pro- duct of the correction factors and the measured compressive strength of each core sample. Tensile strengths presented i n Table I I I are based on measured load without correction. Specifications f o r concrete i n the slab called f o r 0.75-in. maximum size aggregate and a compressive strength of 3»500 p s i a f t e r 28 days. The con- crete contained normal weight aggregat^. I t s u n i t weight was about 15O pcf at the time of te s t . Ends trimmed from the cores were subjected t o petrographic analysis at the PCA Laboratories. Aggregate appeared to be chemically i n e r t and physically sound. Coarse aggregate was p r i n c i p a l l y quartz and quartzite plus a small proportion of granite and g r a n i t i c gneiss. Fine aggregate was p r i n c i p a l l y subangular t o rovinded quartz and feldspar. The con- crete was not air-entrained. A i r content was 1 to 1.5 percent. HESIGS During the Fair, the i n t e r i o r of the structure was used as a restaurant. Atop the waffle slab roof were temporary timber structures with pedestrian walks between the buildings. The walks were made of flagstone set i n a mortar bed, which, i n addition t o the membrane waterproofing, exerted a load of 47 psf on the waffle slab. A view of the roof area i s shown i n Fig. 1. 1-8
Schedules of the reinforcement and controlling dimensions for the columns, r i b s , beams and column s t r i p s are l i s t e d i n Appendix B, Tables B-1 to B-5. S i m i l a r l y , designations of the r i b s , beams and column s t r i p s are shown i n F i g . C-1 to C-3 of Appendix C. Closed rectangular s t i r r u p s were used i n the beams l i s t e d i n Table B-3. MATERIALS During the post-test investigation to determine the properties of the concrete and the location and properties of the reinforcement, several samples of the s t e e l and concrete were taken. Coupons of the r e i n f o r c i n g bars with lengths of about 40 i n . were taken at locations where no y i e l d - ing was observed during the t e s t s . Cores with a nominal diameter of 4 i n . were taken from sound concrete i n each t e s t area. Locations of these samples are shown in F i g . 8 and 9. A l l specimens were shipped to the PCA Laboratories where t h e i r physical properties were determined. Reinforcement i n the structure was specified as "intermediate grade b i l - l e t s t e e l " with deformations conforming to ASTM Standard A305-56. Samples of reinforcement taken from the slab a f t e r the t e s t s were completed were tested i n tension after having been saw cut to lengths of 30 i n . Resiilts of a l l tension t e s t s are presented m Appendix B, Table B-6; average v a l - ues are summarized in Table I . A t y p i c a l s t r e s s - s t r a i n curve obtained from the No. 11 bars i s shown i n Fig. 10. This s t e e l met the require- ments of ASTM designation A15, intermediate grade, as assianed i n the design. TABLE I AVERAGE PROPERTIES OF REINFORCING BARS Bar Size Average Yiel d Stress, k s i Average Ultimate Stress, k s i 8 46.1 78.6 9 53.3 89.7 10 42.7 79.8 11 41.9 80.2 1-7
DESCRIPTION OF TEST STRUCTORE DIMENSIONS AMD REINFORCING DETAILS A plan of the "Rathskeller" roof slab i s shown i n Fig. 2. The waffle slab roof area was bounded by column Lines B, E, 1, and 7- Between column Lines A and B, the roof was an 8-in. deep one-way slab continuous over Line A-B. Overall depth of the waffle slab portion of the structiare was 2k i n . Along Line E there was a 30.375-in. by 45-in. wide edge beam. Beams between columns and s o l i d areas surrounding the columns were obtained by omitting the domes. Within each panel, domes were spaced t o provide a 3-ft center-to- center r i b spacing and a 6-in. stem width at the bottom of the r i b . The domes were l6 i n . deep. This provided a nominal slab thickness of 8 i n . over each dome. Typical sections of the b u i l d i n g are shown i n Fig. 3. Inside the b u i l d i n g , a mezzanine ran along three sides between column Lines 1 and 2, 6 and 7, and A and B. The ex t e r i o r walls of the buil d i n g were 12 i n . t h i c k . I n t e r i o r columns i n the t e s t area were 26 x 26 i n . Several large openings were provided i n the wa l l along the south face of the b u i l d - ing. Line E. Column E3 was 12 x 32 i n . and Colimm Ek was 12 x 24 i n . A view of the south w a l l i s shown i n Fig. k. The "Rathskeller" f l o o r was the top STorface of a r a f t foundation that supported the structure. I t consisted of a s o l i d reinforced concrete f l a t plate having a nominal thickness of 28 i n . A plan view of the founda- t i o n i s presented i n Fig. 5« Typical reinforcement arrangements f o r the roof slab and columns are given i n Fig. 6 and 7, respectively. In the slab, a l l reinforcement l y i n g para- l l e l t o column Lines A through E was placed f i r s t . Temperature reinforce- ment i n the slab was No. h bars on 12 i n . centers i n both directions. The temperature reinforcement p a r a l l e l t o column Lines 1 through 7 was placed k i n . below the top of the slab. Ties f o r the columns were No. 3 bars placed at 12 i n . centers. 1-6
loading equipment and acquisition of both instruments and equipment was carr i e d out over the next two months. Demolition of the temporary structures on top of the "Rathskeller" and removal of non-structural i n t e r i o r f i x t u r e s was completed in March, 1966 and preparation began for the f i r s t t e s t . Waterproofing was cleaned from the roof to expose the concrete, and holes were d r i l l e d in the roof and floor to accommodate the loading equipment. F i n a l preparations began two weeks before the f i r s t t e s t was scheduled. Load equipment was i n - s t a l l e d , and a l l instrumentation was attached and calibrated. The f i r s t of three t e s t s was conducted l a t e in A p r i l 1966. Each t e s t was carried out over a two-day period. On the f i r s t day service loads and moderate overloads were applied. Since a l l s t r a i n s and deflections measured during t h i s phase of loading were expected to be small, the t e s t s were commenced in the l a t e afternoon and were completed during the night to avoid e f f e c t s on the readings of d i r e c t sunlight. Tests to destruction were completed the second day in each case. The t h i r d and f i n a l t e s t was completed ear l y i n May 1966. After t h i s , one week was used to take samples of s t e e l and concrete and to obtain other informa- tion concerning the physical properties of the stmcture. REPORTS The Building Research Advisory Board, administrator for the project, re- quested that highly detailed reports be prepared on the World's F a i r t e s t s . The report on the waffle slab t e s t s i s divided into two separate papers. This f i r s t paper contains a detailed description of the structure and t e s t procedures. Results are analyzed and compared with laboratory investiga- tions and building code requirements; detailed t e s t data are presented in Appendix B. A second paper describes t e s t techniques m considerable de- t a i l , so that investigators conducting future f i e l d t e s t s can benefit from the experiences gained during t h i s work.^9) 1-5
This paper reports tests of the t h i r d structure, a multipanel reinforced concrete waffle slab. During the f a i r , t h i s b u i l d i n g was known as the "Rathskeller"; i t was located beneath a portion of the "Belgian V i l l a g e " shown i n Fig. 1. The approximate outline of the roof of the "Rathskeller" i s indicated on t h i s view by the dashed l i n e s . The t e s t structure was a box-like b u i l d i n g measuring approximately l80 f t by 120 f t i n plan and 20 f t i n height. A 2-ft t h i c k waffle slab supported on columns about 30-ft on center formed the roof of the "Rathskeller". The f l o o r was a 2.5-ft t h i c k reinforced slab that formed a r a f t foundation f o r the structure. Outside walls of the b u i l d i n g were 12 i n . t h i c k . The roof plan and t y p i c a l elevations are shown i n Fig. 2 and 3- OBJECT AND SCOPE In recent decades there has been an I n t e n s i f i c a t i o n of s t r u c t u r a l labor- atory investigation. Results from such work, combined with p r a c t i c a l experience, have formed the basis f o r rapid advances i n s t r u c t u r a l design procedures. Although some observations of ax:tual structures have been made, f i e l d t ests have been hampered by high costs and lack of the con- t r o l s available i n the laboratory. The tests reported i n t h i s paper were carried out (a) t o obtain data from t e s t s t o destruction of a f u l l - s i z e structure, (b) t o correlate t h i s i n - formation with the results of laboratory t e s t s , and (c) to adapt labora- t o r y procedures to f i e l d t e s ts. As o r i g i n a l l y planned, the tests on the "Rathskeller" were intended to investigate: ( l ) shear strength of an i n t e r i o r panel, ( l l ) strenght of column-to-slab connections at edge panels, ( i l l ) "strengthening" of a slab due t o "arch action" when a single panel i s loaded. Behavior a f t e r f i r s t f a i l u r e under each pattern of loading was also studied. PROGRAM OUTLINE Preliminary investigations of possible t e s t buildings and i n i t i a l planning was started i n December 1965. By the end of January I966, i t was decided tha t the "Rathskeller" would be tested. Detailed planning, design of 1-h
More recently t e s t s by J . B. Read^^^ successfully made use of laboratory procedures in f i e l d t e s t s . An accurate determination of both applied loads and response of the structure was made i n t e s t s to destruction of two f u l l - s i z e p o r t a l frames. Although some problems with the use of vibrating wire s t r a i n gages were encountered, mechanical s t r a i n gages were successfully employed. In these t e s t s , a l l quantities were inea- svired with an accuracy comparable to that obtained i n the laboratory- Demolition of buildings a f t e r completion of the 1964-1965 New York World's F a i r presented a unique opportunity f or f i e l d t e s t s of struc- tures. The planned demolition of these buildings permitted s t r u c t u r a l t e s t s to be ca r r i e d to destruction. Although the la c k of adequate time to prepare for the t e s t s was a handicap, laboratory procedures were used to carry out the t e s t s and to gather data. I t was found that, with pro- per attention to d e t a i l s , f i e l d t e s t s can be conducted with precision" associated with laboratory work. CHOICE OF TEST STRUCTURE When i t was learned that some buildings at the s i t e of the 1964-1965 New York World's F a i r might be available for te s t i n g , a preliminary sur- vey of a l l potential t e s t structures was made. Several highly unusual structures u t i l i z i n g free-form s h e l l s and other l i t t l e used s t r u c t u r a l forms were b u i l t for the New York F a i r . Although t e s t s of these struc- tures would have been valuable, the a p p l i c a b i l i t y of the r e s u l t s would have been limited. Consequently, buildings of commonly used s t r u c t u r a l forms were chosen. Three structures were selected for the t e s t s . Two of these were common types of s t e e l frame construction. One of the s t e e l frames was a seven (7) story tower fabricated with standard s t r u c t u r a l shapes; i t was tested to determine i t s dynamic response. The other s t e e l structure was of (8) l i g h t frame and open web s t e e l j o i s t construction. Portions of t h i s building were tested to destruction under both v e r t i c a l and l a t e r a l loads. 1-3
INTRODUCTION BACKGROUND Field t e s t s of buildings have played an important r o l e i n the development of reinforced concrete f l a t slab and f l a t plate construction. One of the e a r l i e s t tests i n North American was reported by W. A. Slater i n 1913.Ì -'-Ì A few years l a t e r , the results of a number of similar tests were simimar- (2) ized and discussed i n a paper by H. M. Westergaard and W. A. Slater. A l l but one of the tests reported were terminated before f a i l u r e of the structure. Despite the lack of sophisticated equipment now available, these early tests were instrimiental i n the evolution of design procedures f o r f l a t slab and f l a t plate structures. Many of the design c r i t e r i a developed i n t h i s way are s t i l l being used more than years l a t e r . Structural tests conducted i n the f i e l d during the l a s t h a l f century have generally been patterned a f t e r the early tests of f l a t slabs. I t has seldom been possible t o carry the tests t o destruction, e i t h e r be- cause of safety r e s t r i c t i o n s or because the structure was due to be placed i n service a f t e r completion of the t e s t s . Few investigators have made any attempt t o canry lavoratory procedvires i n t o the f i e l d . Where attempts t o do t h i s have been made, they have generally met with l i m i t e d success. In the mid-1950's, A. J. Ockleston tested a bu i l d i n g to destruction i n South A f r i c a , ^ ^ ' ^ ' ^ ^ a reinforced concrete hospital scheduled f o r demoli- t i o n . Measurements were made using modified laboratory proced\ires. 1-2
TESTS TO DESTRUCTION OF A MULTIPANEL WAFFLE SLAB STRUCTURE by D. D. Magura and W. G. Corley* SYNOPSIS Three load t e s t s to destruction were made on the waffle slab roof of the Rathskeller Building in the Belgian V i l l a g e exhibit at the 196U-65 New York World's F a i r . Uniform loading was f i r s t applied to four adjacent ' In t e r i o r panels, then to three panels along an edge of the slab, and f i n a l l y to a single i n t e r i o r panel. The behavior of the s t r i c t u r e was in general accord with e x i s t i n g theories, though a shear weakness was observed a t edge columns. How- ever, damage to the structure before t e s t i n g began may have influenced both performance and strength. Performance of the slab under service loads was sa t i s f a c t o r y . Ultimate strength of the slab in a l l three t e s t s \ia.s governed by shear at one or more columns. Ultimate loads were 1.0 dead load plus applied loads ranging from I.9 to 6.h design l i v e load. Only in the single-panel t e s t did f l e x u r a l reinforcement y i e l d before shear punching occurred. The slab possessed some strength a f t e r f i r s t shear f a i l u r e occurred i n each t e s t . Key Words: cracking; deflection; f i e l d t e s t ; f l a t plate; f l e x u r a l strength; reinforced concrete; slab; shear strength; ultimate strength; waffle slab. * Development Engineer and Manager, respectively. Structural Development Section, Portland Cement Association Research and Development Division, Skokie, 111ino1s. 1-1
Contents--continued Discussion of Results ^7 Service Load Behavior 47 Strength Under Overload 0Ì Test I 50 Test I I 52 Otest I I I 53 Post-failure Behavior 53 Conclusions 55 Figures 57 References 117 Appendices 119 A. Frame Analysis B. Tables--Detalled Test Data C. Figures--Detailed Test Data
TEST TO DESTRUCTION OF A MULTIPAWEL WAITLE SLAB STRUCTURE CONTENTS Synopsis 1 Introduction 2 Background 2 Choice of Test Structure 3 Object and Scope k Program Outline h Reports 5 Description of Test Structiore 6 Dimensions and Reinforcing D e t a i l s 6 Materials 7 Design 8 Condition Prior to Testing 11 Structural Tests 1Ì^ Outline of Tests ik Test Preparation ik Loading System 15 Instrumentation l6 Conduct of Tests 19 Test Results 21 Test Data 21 Behavior 21 Test I 21 Test I I 2h Test I I I 26 Analysis of Results 30 Equivalent Frame Analysis 30 Deflections 30 Column Loads 32 Shear Strength of Slab 3^ Comparison of Measured and Computed Shear Strengths ^ Test I kO Tfest I I ho Ttest I I I h2 F l e x u r a l Strength hS
I l l SUBCONTRACTORS REPORT TEST TO DESTRUCTION OF A MULTIPANEL WAFFLE SLAB STRUCTURE and TECHNIQUES FOR FIELD TESTS OF A MULTIPANEL SLAB by D. D. Magura and W. G. Corley 13
application of t e s t loads^ damage t o the structure was v i s i b l e , which co\2ld have reduced shear capacity, p a r t i c u l a r l y at the edge columns. This i s an area which c a l l s f o r further research. With f u r t h e r regard t o shear f a i l u r e , i t was noted that the shear strength of the slab a t the I n t e r i o r column loaded from a l l four sides was greater than Implied by ultimate strength design methods of the 1963 ACI Code; and the shear strength of the slab at edge columns and at an i n t e r i o r column supporting a single loaded panel was less than t h a t implied by design methods of the Commentary on the I963 ACI Code. The Chimes Tower Structiire F a r t i c t i l a r l y impressive was the high degree of damping e3q)erienced and the low degree of p a r t i c i p a t i o n of the foundation i n motion of the structure. S i g n i f i c a n t l y , damping did increase with increasing stress l e v e l and with increasing mode frequency, and description of the damping mechanism re- quired I n t e r f l o o r dashpots; a hlgja. degree of correlation was reeilized be- tween actual and t h e o r e t i c a l response. I t i s also s i g n i f i c a n t t h a t : 1. During one dynamic t e s t , response of the structvire induced about 75?^ of the base shear which would l i k e l y have resulted from groiind motion of I n t e n s i t y equal to t h a t of the l^ko E l Centro strong-motion earthquakfij and 2. the foundation slab rotated p r i m a r i l y about one horizontal axis i n the plane of the bottom of the base, perpendicular t o the d i r e c t i o n of the e x c i t i n g force, but there was also r o t a t i o n about a horizontal eucis p a r a l l e l t o the d i r e c t i o n of the e x c i t i n g force, located approximately 2.5 t o 3.0 feet beneath the bottom of the foundation slab. 11 -
Of the results obtained from l a t e r a l load t e s t i n g of the Bourbon Street structure, i t was s i g n i f i c a n t that the structure resisted lateraJL loads i n excess of the t h e o r e t i c a l f u l l y p l a s t i c capacity of the whole frame. This, combined with the plate diaphragm action of both the roof and se- cond-floor systems, which permitted f u l l p a r t i c i p a t i o n i n l a t e r a l r e s i s t - ance not only of the wind frames but also of a l l pipe columns, led t o a l a t e r a l load capacity considerably i n excess of that required f o r wind design. I t i s s i g n i f i c a n t also that f o r the types of foundation and column base used, the wind frame column base connections acted as i f neither f u l l y f i x e d nor pinned. Concentrated load t e s t i n g of the structure demonstrated that the 0 . 5 -inch diameter horizontal bridging was f u l l y e f f e c t i v e and eidequate i n providing l a t e r a l s t a b i l i t y t o the st e e l bar j o i s t s ; composite action between the concrete deck and j o i s t s did not e x i s t t o a s i g n i f i c a n t degree. The f l o o r v i b r a t i o n tests developed excellent, usable data on the v i b r a t i o n character- i s t i c s of an open-web j o i s t f l o o r ; the most interesting feature of the t e s t was the high degree of damping reaJ.ized. The Rathskeller Structure This structure served adequately as designed, providing s u f f i c i e n t factor of safety. For example, i f i t i s assumed th a t the t o t a l design load was approximately 567 psf (dead load of k7 psf f o r flagstone, mortar bed, and waterproofing, plus 220 psf f o r concrete slab, f o r a t o t a l of 267; plus a design l i v e load of 300 psf) r a t i o s of oatimate t e s t load t o 567 psf wo\ild be approximately 2 .0 , l.k, and 3 .9 , respectively, f o r the tests with four i n t e r i o r panels loaded, three edge panels loaded, and a single i n t e r i o r panel loEided. Despite the ample safety provided, flexursJ. capacity was not reached i n any of the three t e s t s . Rather, ultimate strength was governed by shear at i n t e r i o r and edge columns before a y i e l d i n g mechanism could be devel- oped completely. Cause of such f a i l u r e i s not known; however, p r i o r t o 10
structures and tested In the laboratory t o determine actual strength characteristics; actual dimensions t o which the structure was b u i l t should be obtained; location, type, and q u a l i t y of concrete r e i n - forcement bars determined; number and q u a l i t y of b o l t s used i n previously concealed connections determined; f i x i t y of end connections assessed.) The Bourbon Street Struct\ire This structure, one of the two tested t o destruction, served adeq\iately as designed, providing more than s u f f i c i e n t factor of safety. Under uniform loading, ultimate f a i l u r e of both the roof and second f l o o r occurred at points near the ends of the bar j o i s t s . Mode of f a i l u r e could not be p o s i t i v e l y i d e n t i f i e d ; however, a close examination of the records suggests that f a i l u r e of both the roof and f l o o r systems was I n i t i a t e d by y i e l d i n g of the j o i s t s followed by buckling of diag- onals and top chords i n the area of f a i l u r e . Cause of the f a i l u r e could not be d e f i n i t e l y i d e n t i f i e d e i t h e r , but various hypotheses f o r the cause are given i n the detailed subcontractor report (Volume l ) on the structTire. The spandrel beams experienced torsional d i s t o r t i o n ; t h i s would l i k e l y have contributed s i g n i f i c a n t l y t o i n i t i a t i o n of the f a i l i i r e . The horizontal forces exerted on side walls due t o the vacuum created w i t h i n the structure t o apply the \inlform load also might have had a contributing e f f e c t . Comparison of results from the uniform load tests with those predicted using conventional methods of analysis and design indicated t h a t : 1. Deflections of the s t e e l j o i s t s roof system could be predicted accurately by use of the simple-span beam formula with a m u l t i p l i e r of 1.25, while deflections of the f l o o r j o i s t s ranged between values below those com- puted with the beam formiila without a m u l t i p l i e r and those computed using a m u l t i p l i e r of 1.25; and 2. top and bottom chord a x i a l stress can be closely estimated by beam analysis using only chord area t o determine section properties of the j o i s t .
investigations involving both laboratory t e s t i n g and th e o r e t i c a l anal- ysesâwith the aim of obtaining answers t o the questions raisedâmust be pursued i f the p o t e n t i a l benefit of a f u l l - s c a l e t e s t i n g program i s to be realized. Within the time, funds, and resources made available, the Committee could not perform a l l possible analyses of the data collected. Thus, a l l re- corded data have been reduced and are published i n the reports r e s u l t i n g from the program t o allow interested researchers t o investigate areas and aspects of pa r t i c u l a r i n t e r e s t t o them. The Committee encourages t h i s type of follow-up a c t i v i t y and, i n order t o realize the f u l l e s t benefits of the program, urges researchers t o make t h e i r r e sults a v a i l - able through technical journals. (Copies of raw data may be obtained from the National Academy of Sciences by any interested researcher.) Future Programs f o r Testing Full-Scale Structures Much was learned during the course of t h i s investigation regarding con- duct of f u l l - s c a l e structure t e s t i n g which shoiild be of benefit t o the management of any similar program i n the f u t i i r e . Of p a r t i c u l a r import- ance i s the absolute need t o know what i s t o be tested. Thus, provi- sions should be made at the onset f o r investigation of the his t o r y of the structure ( i . e . , when and how i t was b u i l t and how used) as w e l l as f o r : 1. A thorough suirvey t o detennine whether or not the structure has undergone i n t e r n a l stresses due t o unequal settlements, creep, or other influences; and 2. detailed v i s u a l examination of connection, weldments, e c c e n t r i c i t i e s , and the l i k e , t o i d e n t i f y and isolate areas of known weakness. An e f f o r t should then be made t o rate the structure i n l i g h t of t h i s knowledge. As-built drawings ought t o be obtained, and devia- tions from design determined and noted. Perhaps most important, a thorough post-mortem examination of the struc- ture should be performed t o determine deviations from design' specifications (e.g., a s u f f i c i e n t number of material samples sho\ald be taken from
I I REPORT OF THE SPECIAL ADVISORY COMMITTEE The i n d i v i d u a l subcontractor reports emanating from t h i s investigation, reviewed and accepted by the Advisory Committee, contain d e t a i l s of the actual t e s t i n g performed on the three structures Involved, as w e l l as a l l f i n d i n g s , conclusions, and recommendations. Comments from the Committee, on salient aspects of the o v e r a l l program and p a r t i c u l a r l y s i g n i f i c a n t resvats of tests performed on the in d i v i d u a l structures, follow. Value of the Testing Program and Results The t e s t s performed revealed vinsuspected strengths and weaknesses; however, neither design practices nor build i n g codes reasonably could be revised or modified solely on the basis of these few f u l l - s c a l e t e s t s . Nor should judgments as t o the a p p l i c a b i l i t y of t e s t findings t o other similar structures be made without extreme caution u n t i l s u f f i c i e n t data are collected through other f u l l - s c a l e t e s t i n g pro- grams t o permit establishment of findings and conclusions having s t a t i s t i c a l significance. Results of the t e s t s do h i g h l i g h t specific areas i n which f u r t h e r research and laboratory investigation could probably lead t o considerable Improvement i n e x i s t i n g design c r i t e r i a . I n p a r t i c u l a r , no determination co\ild be made of whether the ultimate strengths of the structures tested t o destruction reflected actual factors of safety associated with design techniques or the synergistic strengthening e f f e c t of three-dimensional interplay. However, the program demonstrated the complete f e a s i b i l i t y of t e s t i n g f u l l - s c a l e structures i n the f i e l d with a degree of precision approximating t h a t attainable i n the laboratory. At t h i s stage of development f u l l - s c a l e t e s t i n g of stxnctvires must be viewed p r i n c i p a l l y as a source f o r devel- oping questions rather than answers; consequently, follow-up
the Chimes Tower and Bourbon Street Structure, as wel l as findings, con- clusions and recommendations based on analyses of the collected data; and personnel of the Portland Cement Association prepared and submitted a similetr report on the t e s t i n g of the Rathskeller structure t o the Advisory Committee through Wiss, Janney, Elstner and Associates. Prior t o t h e i r acceptance, a l l reports were c r i t i q u e d by the Advisory Committee meeting repeatedly i n session with the various report authors. Publication of Results Results of the t e s t i n g program are published i n three separate volumes, under the general t i t l e heading FULL-SCALE TESTING OF NEW YORK WORLD'S FAIR STRUCTURES. The three volumes carry the s u b t i t l e s The Bourbon Street Structure (Volvime l ) The Rathskeller Structvire (Volume I I ) The Chimes Tower Structure (Volume I I I ) Each volume i s camprised of three sections: INTRODUCTION; REPORT OP THE SPECIAL ADVISORY COMMITTEE;.and SUBCONTRACTOR REPORT. The INTRODUCTION section i s i d e n t i c a l i n each volume and relates the ov e r a l l purpose of the program, gives a b r i e f description of the three structures tested, and d e t a i l s tbe t e s t s conducted on each of the three structures. The second section, REPORT OF THE SPECIAL ADVISORY COMMITTEE, i s , l i k e the introduction, i d e n t i c a l i n each volume and contains comments from the Committee on salient aspects of the o v e r a l l program and particTilarly s i g n i f i c a n t results of te s t s performed on the ind i v i d u a l structures, the SUBCONTRACTOR REPORT section of each volume i s comprised of the re- port prepared by the t e s t i n g subcontractor on the p a r t i c u l a r structure bearing t h a t volume s u b t i t l e . Presented i n each such SUBCONTRACTOR REPORT section, i s a detailed description of the t e s t i n g program, re- svilts of te s t s and comparative analyses, conclusions and recommendations, and tabulated data.
Conduct of the Program The concept of f u l l - s c a l e t e s t i n g of selected New York World's Fair struc- tures originated i n 1965 with the American Society of C i v i l Engineers. A f t e r I n i t i a l and favorable discussions with o f f i c i a l s of the New York World's Fair Corporation and following encouraging recommendations from the f i r m of Wlss, Janney, Elstner and Associates regarding f e a s i b i l i t y of conducting the t e s t s , a Special Advisory Committee was appointed by the Building Research Advisory Board at the request of ASCE t o administer the e n t i r e program. Responsibilities assigned to the Advisory Committee Included: 1. Determination of those structvires t h a t would be load tested on the basis of: Analysis of s t r u c t u r a l and construction drawings and on-site inspection; information obtained from designers, contractors, owners, and the World's Fair Authority; and the amount of funds available f o r planning and conducting the t e s t i n g program. 2. Determination of the kinds of data that would be obtained from the t e s t s . 3. Determination of the loading techniques, instrumentation, and procedures f o r recording data. k. Guidance t o the t e s t i n g organization. 5. Review and analysis of data and preparation of reports f o r publication. Under the guidance of the Special Advisory Committee, the engineering f i r m of Wlss, Janney, Elstner and Associates was retained under contract by the National Academy of Sciences, t o supervise general planning f o r the entire t e s t i n g program and t o carry out or supervise a l l f i e l d t e s t i n g ; i n addi- t i o n the f i n n selected and/or designed a l l methods of t e s t and planned the tests performed on both the Chimes Tower and the Bourbon Street structure. For the Rathskeller structvire, detailed plans f o r t e s t i n g were developed \mder guidance of the Committee by personnel of the Structural Development Section of the Portland Cement Association. Subsequently, Wlss, Janney, Elstner and Associates prepared and submitted t o the Advisory Committee highly detailed reports r e l a t i n g t o the actvial t e s t i n g performed on both
k. v i b r a t i o n loading t o determine a. natural frequencies of several of the open-web Joist f l o o r s b. magnitude of the dynamic deflection and the degree of damp- ing under impact. The Rathskeller S t r u c t u r e â This was a one-story, box-like structure meas\irlng approximately l80 by 120 feet i n plan and 20 feet i n height. A 2-foot-thick waffle slab supported on columns, about 30 feet on cen- t e r s , formed the roof. The f l o o r was a 2.5-foot-thick reinforced slab t h a t formed a r a f t foundation f o r the structure. Outside walls of the b u i l d i n g were 12 inches t h i c k . Uniform loading tests were conducted on t h i s structure t o Investigate: 1. Shear strength of an i n t e r i o r panel. 2. Strength of coltmin-to-slab connections at edg@ panels. 3. Strengthening of a slab due t o arch action when a single panel i s loaded. k. Residual strength a f t e r i n i t i a l f a i l u r e under each of the above patterns of loading. The Chimes Tower S t r u c t u r e â . This was a seven-story structxire construc- ted of r o l l e d steel sections bolted and/or rive t e d together. The struc- ture was 9 f e e t - 3-3/8 inches square and 86 feet-8 inches t a l l . A l l f l o o r s had a uniform height of 12 feet except the top story, which was Ik feet-8 inches high. Dynamic tests were conducted on t h i s structure t o determine: 1. Nattiral frequencies and mode shapes, and t o compare r e s t i l t s with values predicted t h e o r e t i c a l l y . 2. Degree of damping. 3. P a r t i c i p a t i o n of the foundation i n motion of the tower. Static t e s t s were aJ.so conducted on the structure t o : 1. Allow comparisons of f l o o r deflections and foundation motion with those observed during the dynamic te s t s .
columns which were part of the wind framing provided at in t e r v a l s of approximately 50 feet throughout the length of the structure. The columns were supported by reinforced concrete grade-beam footings. Roof construction was similar t o that of the second f l o o r , except th a t a standard metal deck supported r i g i d insulation and roofing material. The following t e s t s were conducted on t h i s structure: 1. Lateral loading of the frames t o observe and evaluate a. diaphragm action of the roof and second f l o o r b. s t r u c t u r a l behavior of wind frame and pipe colvmins c. mode of f a i l u r e . 2. Uniform loading of the f l o o r and roof t o a. determine mode of f a i l u r e and maximum load capacity of the two s t r u c t u r a l systems ( f l o o r and roof) and compare re s u l t s w i t h what would be predicted b. investigate those conditions, usually not considered i n simple designs, which may have a strength-enhancing or strength-di- minishing e f f e c t on the structvire ( i . e . , unintended composite action, s t i f f e n i n g e f f e c t s of unintentional continuity, fotin- datlon settlements, supporting-beam l a t e r a l rotations and translations due t o unsymmetrlcal loading, and ef f e c t s of attached cvirtaln walls) c. determine the loads (stresses) Induced i n the various struc- t u r a l elements (chord and web members) comprising a stee l j o i s t and compare re s u l t s with values predicted from beam or truss analysis d. determine deflection behavior of the v£u:lous s t r u c t \ i r a l elements under load and compare results with values pre- dicted from t h e o r e t i c a l analysis e. determine, i f possible, areas where additional laboratory research i s required t o obtain a better understanding of s t r u c t u r a l elements connected i n t o a three-dimensional framework f . evaluate loading and Instrvmientatlon techniques used i n t e s t i n g the b u i l d i n g t o f a i l u r e . 3. Concentrated load tests t o evaluate a b i l i t y of second f l o o r t o sustain such loading.
I INTRODUCTION Purpose The fundamental purpose of the program was t o t e s t f u l l - s c a l e structures selected from the complex of buildings erected f o r the 196^-65 New York World's Fair i n order t o ascertain t o the extent possible the degree of corr e l a t i o n between performance of actual s t r u c t u r a l systems and t h a t predicted from laboratory t e s t i n g and design theories. Scope ajid Limitations While the program was coordinated with the New York World's Fair Corpora- t i o n which gave i t s f u l l support, previously established demolition schedules had t o be maintained, necessitating expeditious planning and execution of the t e s t i n g phase of the program. Structural drawings were obtained and studied t o make a determination of which structures were best suited f o r t e s t ; of 15 i n i t i a l l y \mder consideration, three, repre- sentative of contemporary construction, were selected f o r f u l l - s c a l e t e s t i n g : 1. A l l ^ t - f r a m e , open-web steel-Joist and steel-pipe colimm type known as the Bourbon Street structure 2. A one-story multlpanel relnforced-concrete waffle- slab type known as the Rathskeller structure 3« A seven-story tower fabricated with standard s t e e l structxiral members known as the Chimes Tower. The Bourbon Street S t r u c t t t r e â . This was a two-story structure, approxi- mately 750 feet long and 50 feet wide. The ground f l o o r was concrete slab-on-grade; the second f l o o r consisted of open-web stee l Joists with a lightweight concrete deck supported by a corrugated st e e l decking, spot welded t o the top cord of the Joists. The Joists were carried by wide- flange st e e l sections framing i n t o s t e e l pipe columns, or wide-flange
CONTENTS Page Foreword v l i Section I . INTRODUCTION 1 Purpose Scope and Limitations Publication of Results I I . REPORT OF THE SPECIAL ADVISORY COMMITTEE 7 I I I . SUBCONTRACTOR REPORT l 3 TEST TO DESTRUCTION OF A MULTIPANEL WAFFLE SLAB STRUCTURE and TECHNIQUES FOR FIELD TESTS OF A MULTIPANEL SLAB by D. D. Magura and W. G. Corley i x
The complex of structures b u i l t f o r the 196U-65 New York World's Fair presented an opportunity almost without precedent f o r conducting such f u l l - s c a l e s t r u c t u r a l t e s t i n g . Most of the structures already were scheduled f o r demolition, and, though t h i s obviously imposed a time l i m i t a t i o n on the t e s t i n g program, t h e i r a v a i l a b i l i t y precluded the large f i n a n c i a l commitment which would otherwise have been required t o b u i l d the structures solely f o r t e s t i n g . Also, although r e l a t i v e l y new, the structures had been subjected t o occupancy conditions. Thus, the complex of structures constituted a unique resource. Many of the structures constructed f o r the f a i r represented innovations i n design, but most were representative of contemporary, conventional construction. I t was decided t o select a representative sample of the l a t t e r f o r t e s t i n g i n the b e l i e f that r e s u l t i n g information woiild be of greater universal value since such structures would most l i k e l y r e f l e c t current design c r i t e r i a and construction practices. Obviously, these structiores presented one major l i m i t a t i o n â t h e y were not designed and b u i l t s p e c i f i c a l l y f o r t e s t i n g . As a r e s u l t , one could expect i n - consistencies i n both application of design c r i t e r i a and actual f i e l d performance during construction. A Special Advisory Committee, comprising knowledgeable individuals recognized f o r t h e i r technical expertise i n one or more of the p r i n c i - p al areas of Investigation, was appointed by the Building Research Advisory Board t o conduct the study. Results of the t e s t i n g program, which constitute the culmination of a unique and rewarding research e f f o r t i n the f i e l d of b u i l d i n g science, are reported i n t h i s and two other volumes prepared f o r the Board by the Special Advisory Committee. Each volume has been reviewed, accepted, and approved f o r t r a n s m i t t a l t o the program sponsors by the Executive Committee of the Board, acting on behalf of the" Board. ROBINSON NEWCOMB, Chairman Building Research Advisory Board
FOREWORD H i s t o r i c a l l y , b u i l d i n g science has been hampered by the dearth of informa- t i o n on the performance of f u l l - s c a l e structures. For the obvious reason of cost, t o t a l b u i l d i n g structtires t r a d i t i o n a l l y have not been erected f o r the sole purpose of t e s t i n g t o or near to d e s t r u c t i o n â y e t , doing so i s an obvious necessity i f one i s t o be able to t r u l y validate design c r i t e r i a . I n the absence of such t e s t i n g , the research and design communities have had t o r e l y t o a great extent upon assumed ultimate structure performance extrapolated from scale-model and component te s t s . As a consequence there has always been the nagging doubt as t o whether designed structures actu- a l l y perform as predicted; that i s , whether structures are i n some respects overdesigned and i n others, underdesigned. Engineering l i t e r a t u r e throughout the world does contain l i m i t e d references t o t e s t i n g of f x i l l - s c a l e structvires erected f o r normal use. From the few tests reported, much of what was learned has been of value t o researchers and others concerned with b u i l d i n g performance. Perhaps the most s i g n i f i - cant consequence of f t i l l - s c a l e t e s t i n g has been the indication that the ultimate load-carrying c a p a b i l i t y of three-dimensional structures may be considerably I n excess of that which would be predicted from accepted de- sign c r i t e r i a ; that i s , t h a t factors of safety may be f a r greater than needed or intended. Thus, i t has long been f e l t that i f s u f f i c i e n t data could be obtained through t e s t i n g of a series of f u l l - s c a l e buildings, a better understanding of actual performance would be gained, more rea- l i s t i c performance requirements established, and improved design c r i t e r i a and techniques developed--the net r e s u l t being improved b u i l d i n g e f f i - ciency and economy. v i i
and James R. Bryson of the National Bureau of Standards; Messrs. William Palmertree and Norman G. Hansen of the U. S. Army Corps of Engineers; t o Mr. Albert Peter of the General Services Administration; Mr. Howard Johnson of AAA Photographers; and Mr. Eugene L. Herzman, the structviral designer of the Chimes Tower. ⢠A note of special recognition i s due Mr. Donald C. Taylor, Research Manager, American Society of C i v i l Engineers, whose e f f o r t s were i n s t r u - mental i n making the program a r e a l i t y . Appreciation and acknowledgment also must be escpressed f o r the support provided by a group of sponsors from both industry and government; namely, American Iron and Steel I n s t i t u t e National Aeronautics and Space Administration American Society of C i v i l Engineers (Services) National Bureau of Standards Army Corps of Engineers National Science Foundation Concrete Reinforcing Steel I n s t i t u t e Office of C i v i l Defense Engineering Foundation Portland Cement Association (Services) Ford Foundation General Services Administration Reinforced Concrete Research Council ^li^f^^^^^^ Engineering ^^^^ j ^ . ^ ^ I n s t i t u t e F i n a l l y , the contributions t o the program of the Chairman and members of the Special Advisory Committee are g r a t e f u l l y acknowledged. v l
ACKNOWLEDGMENTS The program of f u l l - s c a l e t e s t i n g of selected structures of the Hew York World's Fair was conceived by the American Society of C i v i l Engineers. The f i r m of Wiss^ Janney, Elstner and Associates supeirvised general plan- ning f o r the t e s t i n g program and also the actual f i e l d t e s t i n g . In addi- t i o n t o designing methods of t e s t and tests performed on both the Chimes Tower and Bourbon Street structures; f o r the Chimes Tower, Professor V. J. McDonald of the University of I l l i n o i s was consultant on instrumenta- t i o n . For the Rathskeller structure, Professor M. A. Sozen of the University of I l l i n o i s I n i t i a l l y suggested the t e s t plan and served as a consultant; detailed plans f o r t e s t i n g of t h i s structvire were devel- oped by personnel of the Structural Development Section of the Portland Cement Association. Field t e s t i n g was carried out by personnel of Wiss, Janney, Elstner and Associates, Simms Engineering, the Corbetta Construc- t i o n Company, and the Portland Cement Association. The v i b r a t i o n generator and a u x i l i a r y equipment used In the dynamic t e s t - ing of the Chimes Tower structure were supplied by the Department of Engineering of the University of C a l i f o r n i a , Los Angeles; much i n s t r u - mentation used i n the program was provided by the U. S. Army Engineering Waterways Esqerlment Station at Vicksburg, Mississippi, and by the U. S. Army Aberdeen Proving Grounds at Aberdeen, Maryland. Grateful acknowledgaent i s due the many individuals who gave f r e e l y of assistance and suggestions which contributed s i g n i f i c a n t l y t o the suc- cessful completion of t h i s project. In t h i s regard, p a r t i c u l a r recog- n i t i o n must be given t o the s t a f f of the New York World's Fair 19614-65 Corporation--to Messrs. John T. O'Neill, Joseph Myers, S. A. Potter, Pazel G. Jackson Jr . , and William McCarthy; t o Messrs. Robert G. Mathey
SPECIAL ADVISORY COMMITTEE on FULL-SCALE TESTING OF NEW YORK WORLD'S FAIR STRUCTURES CHAIRMAN ROBERT B. TAYLOR, Mapleton Development Inc., Minerva, Ohio MEMBERS WILLIAM J. BOBISCH, Engineering Division, Naval F a c i l i t i e s Engineering Command, Washington, D. C. CHARLES D. MORRISSEY, Praeger-Kavanagh-Waterbury, New York, N. Y. RAYMOND C. REESE, Raymond C. Reese Associates, Toledo, Ohio MICHAEL N. SALGO, F a c i l i t i e s Engineering, Col\anbia Broetdcasting System Inc., New York, N. Y. CHESTER P. SIESS, Department of C i v i l Engineering, University of I l l i n o i s , Urbana, I l l i n o i s I . M. VIEST, Bethlehem Steel Corporation, Bethlehem, Pennsylvania TECHNICAL LIAISON DONALD C. TAYLOR, American Society of C i v i l Engineers, New York, N. Y. BRAB STAFF ROBERT W. SPANGLER, Assistant Director-Program Planning (Former) WILLIAM A. COSBY, Staff Engineer JAMES R. SMITH, Assistant Director-Technical Operations i i i
The study which resulted in this report was supported by grants and services from eight private industries organizations and by support through contracts from seven agencies of the Federal Government. Reproduction in whole or in part i s permitted for any purpose of the United States Giovemment. Full-scale l e s ' l i i j i'^' F a i r stiLn.tii'e>« i f J Inquiries concerning this publication should be addressed to: Tbe Executive Director, Building Research Advisory Board, Division of EngineeringâNational Research Council, 2101 Constitution Avenue, N. W., Washington, D. C. 204l8. i i
