2IO Journal of Agricultural Research voi. vi, no. 6 



After readings over the entire slab had been taken, check readings were 

 made at various points; and invariably it was found that these check 

 readings showed an increased deformation in the concrete even though its 

 temperature remained constant. Moreover, upon releasing the load en- 

 tirely it was found that considerable permanent deformation remained in 

 the concrete. This phenomenon can be attributed only to the "flow" or 

 gradual change in length of the concrete even when under small stresses 

 and is significant, for it shows the importance of the time effect on the 

 relation of stresses and strains in concrete. If the strain readings on 

 the top of the slab, loaded for five or six hours, be used to estimate the 

 stresses in the concrete, based on the initial modulus of elasticity of the 

 concrete, this estimated stress will be greatly in excess of the true stress 

 conditions. 



In view of the fact that the deformations which take place in the 

 concrete under a sustained load are continually increasing and remain 

 partially permanent, and that the only deformations of value are those 

 indicative of the stress, all of the final calculations and deductions are 

 based upon results obtained by taking zero deformation readings just 

 before applying the load. Deformations thus obtained by taking the 

 difference between the strain-gauge readings at the zero load and the test- 

 ing load (all within an hour or so), represent more accurately the elastic 

 deformations and are a better indication of the stress existing in the 

 concrete than those obtained from any initial or previous zero readings. 



GRAPHICAL REPRESENTATION OF DATA AND RESULTS 



A great amount of numerical data has been taken during the tests of 

 these three concrete slabs. Some of these data were preliminary and 

 served only to indicate methods and limits. Those data which have a 

 direct bearing upon the problem are shown graphically in the accom- 

 panying curves (fig. 2-28). 



Figures 2,3, and 4. — The layout of the points in the concrete and the 

 steel over which the strain-gauge readings were taken are shown in figures 

 2, 3, and 4. In a few cases readings were m.ade between all points, but 

 in general only the readings along a center line (5-6) parallel to the 

 supports were taken, as this gives sufficient data for determining the 

 effective width. In all mention of strain-gauge or deformation readings 

 it should be understood that they are measured between points on a line 

 perpendicular to the supports, unless expressly stated to be otherwise. 



Figure 5. — Figure 5 shows the variation of the concrete deformations 

 for different concentrated center loads, along the center line of the slab. 

 The ordinates of these curves are influenced slightly by the time factor 

 or "flow" in the concrete; hence, the values for the effective width b are 

 somewhat erratic in their relation to the load. 



