Experimental Determination of Eccentricity of Floor I D. Watstein and P. V. Johnson Building Research Division Washington, D.C. Architect TH 7 по 2. Description of the compressive steel reaction strut for measurement of eccentricities, and its calibration. 3. Eccentricity of reaction of an I-beam bedded in gypsum plaster___ 4. Eccentricity of reaction of an I-beam supported on neoprene rubber_ 5. Eccentricity of supporting force at the base of a bearing wall sub- Experimental Determination of Eccentricity of Floor Loads A to a Bearing Wall D. Watstein* and P. V. Johnson* The eccentricity of the loads applied to a specially calibrated compressive strut simulating a brick bearing wall was experimentally determined for a variety of bearing materials and conditions of contact. In one series of tests, an I-beam was bedded in high strength gypsum plaster, bonded and unbonded. For the unbonded plaster bed the eccentricity ratio 1 increased with the applied load to a maximum value of about 0.42, while for the bonded plaster bearing this ratio decreased to an average value of about 0.24 at the maximum load. In the second series of tests the eccentricity was observed for an I-beam supported on neoprene rubber pads, capped and uncapped, of different thicknesses, and of different bearing length. In general the eccentricity ratio increased slightly with the applied load. Lack of intimate contact between the I-beam and the rubber pad % in thick resulted in an eccentricity ratio of about 0.40, or nearly the same as for unbonded plaster bearing. Intimacy of contact produced by plaster capping resulted in a marked reduction in the eccentricity ratio to about 0.29; the confinement of the bearing length of the rubber pad to one-half of that used in previous tests and placing it at the extreme end of the beam, further reduced the eccentricity ratio to about 0.18, and to 0.13 for a rubber pad 0.25 in thick. Key words: Bearing pads, bearing walls, brick masonry, design of bearing walls, eccentricity of applied loads. 1. Introduction Since exterior bearing walls are designed as eccentrically loaded compression members, it is important to know what the eccentricities of the applied floor loads are for different bearing materials and different conditions of contact between the supporting structure and the floor beams. Some recently completed exploratory studies by the Structural Clay Products Institute Research Fellowship at NBS indicated that it is feasible to measure the eccentricities of applied loads using a specially designed stress-sensitive compressive strut calibrated under loads of known eccentricities. The strut was assumed to simulate a load bearing wall of brick masonry ever the boundary conditions and the elastic p of the strut were different from those enc in an actual masonry structure. The exploratory study included an inve of such parameters as thickness and ris bearing materials, the intimacy of con tween the supporting structure and the members, and the effect of bond with the material on the eccentricity of the floor 1 The feasibility of measuring the ecc of the force supporting a masonry wall s to an eccentrically applied load was also e 2. Description of the Compressive Steel Reaction Strut for Measurem Eccentricities, and Its Calibration The compressive steel reaction strut was a rectangular steel tube 4 by 8 inches in cross section having a wall 0.187 in thick. The strut was 18 in high and had a %-in welded steel plate insert at the top providing a closed end. The strut was capped with a 1- by 4- by 8-in cold-rolled steel plate bonded to the top welded plate insert with epoxy cement. The whole assembly was then capped with a solid extruded clay brick which served to receive the load, simulating the bearing conditions at the top of a brick masonry wall. *Present address: Structural Clay Products Institute, 1750 Old Meadow Road, McLean, Virginia 22101. 1 The eccentricity ratio is defined as the ratio of eccentricity to the overall thickness of the strut. The open bottom end was machined no the axis of the strut and was support machined steel plate 4 in thick. The dimensions of the capping brick by 36 by 7% in and hence the brick cover the 1-in steel cap completely. T was set flush with one side of the stru nated as side H) and was centered with to the 4-in dimension of the strut as s figure 1. The strut was instrumented with bond strain gages at two different levels. The of the gages is indicated in figure 1. The theoretical relationship given by (7) and the experimental curve determined in two separate laboratory setups are compared in figure 2. The experimental values of eccentricity e were varied over a range of 3 in on each side of the center line of the strut. It should be added that the strain values given in figure 2 were those obtained from the lower set of strain gages, since they yielded a more consistent relationship than the top gages. The departure of the calibration curve from the straight line predicted by eq (7) may possibly be accounted for by the nonhomogeneous nature of the welded steel tube used in fabricating the strut, and the possible presence of slight irregu 2P Omax+omin A (3) 0.4 SIDE L larities commonly found in thin walled extruded steel shapes. In future work, it is intended to fabricate a strut of greater strain sensitivity and have it relatively free from the deficience might have caused the deviation from the relationship shown in figure 2. 3. Eccentricity of Reaction of an I-Beam Bedded in Gypsum Plaster One series of five tests was carried out with an I-beam 6 in deep and a flange 3%1⁄2 in wide bedded in high strength gypsum plaster. As in all determinations of eccentricity described in this paper, the end of the I-beam extended to the center line of the supporting strut. The arrangement of supports and the load are shown in figure 3, along with the device for measuring the rotation of the beam end at the strut. This measurement was an approximation based on the assumption that the strut did not depart from its initial vertical position, and that the vertical displacement of the point at which the micrometer dial assembly was attached to the I-beam was negligible. In Tests 1 and 2 the I-beam was bedded in unbonded plaster. The bond between the plaster and the bearing surfaces was destroyed by confining the plaster putty between two sheets of polyethylene. The variation of the eccentricity ratio with the rotation of the beam end is illustrated in figure 4. It is noted that at small rotations of the beam supported on the strut simulating a wall, the eccentricity ratio e/t1 was about 0.35, or nearly the value usually assumed in design of masonry walls. The eccentricity ratio increased with the rotation of the beam and tended to reach a constant value at large rotations. For Tests 1 and 2, the maximum values of e/t were 0.43 and 0.40 respectively, with an average of 0.4 maximum center load applied in these te 25 kips and the beam span was 44 inches. In Tests 3, 4, and 5 the I-beam was in bonded plaster and the span lengths wer and 48 inches respectively. It is intere compare in figure 4 the effect of bonded bonded plaster bearings on the behavio I-beam. The unbonded I-beam showed an in the eccentricity ratio with load, while the bonded I-beam showed the opposite. T of e/t for the bonded I-beam was about low loads and decreased with the rotatio beam; the eccentricity ratio tended to constant value as the rotation increas average e/t for the three tests at the m recorded rotation was about 0.24, and re a reduction of 42% as compared with th tricity ratio for unbonded plaster. Although the use of high strength plaster is not practical as a permanent material for floor beams, its effect as a bearing material was investigated as one the broad problem of load transfer to walls. It is possible that some other more nent bedding material can be found whic have the same favorable effect on the ecc ratio as the high strength gypsum plaste |