5.3. Prisms

Test results on eccentrically loaded 2-block high and 3-block high prisms made of 6-in and 8-in block are given in tables 5.5 and 5.6, respectively. These tables note the age, as well as the date of fabrication of prisms. This is noted since the strength of prisms fabricated on certain dates sometimes deviated markedly from the generally observed trend.

6. Interpretation of Results

6.1. Stress-Strain Relationships

Stress-strain relationships, measured on unreinforced as well as reinforced walls, are shown in figure 6.1. Curve A shows a stress-strain curve computed from the longitudinal deformation of an axially loaded 16-ft unreinforced wall (specimen 2). This is

the only case where deflections were measured to the point of ultimate load. In all other cases instru mentation was removed prior to failure. Curves B and C are stress-strain curves for axially loaded 10-ft unreinforced walls, computed from the average of measurements of 4 linear variable differential transformers having a gage length of 24 in, which were specially installed for that purpose. (Curve B is for specimen 14 and curve C for specimen 4.) Curve C was obtained from one of the specimens with exces sive joint thickness and may therefore represent a wall of lower than normal strength. The ultimate compressive strength of the specimens from which curves B and C were derived is also shown in the figure. Stress was computed on the basis of the average net area, determined in accordance with ASTM Standard C140-65T.

There is good agreement between the three In general, the measured stress-strain

curves.

minimum face-shell and web thicknesses, the of the grout and a transformed steel area based o "n" of 29 assuming that the average modulu elasticity of masonry is approximately 1 × 10 and that of steel is 29 × 106 psi. It can be seen f a comparison of curve D with curves A, B and C the area transformation which conservatively sumed equal stiffness for grout and masonry also was based on minimum, rather than aver net area of masonry probably resulted in over mating the stresses in the masonry.

The value of the initial modulus of elasti derived from curve D is 2.8 × 106 psi. A tan modulus at failure could not be obtained since strumentation was removed before the mas developed its ultimate strength. Note that the str strain curve is essentially linear up to a stres about 80 percent of the failure stress. A linear str strain relationship probably approximates the st distribution up to failure reasonably well.

6.2. Cross-Sectional Capacity

In order to study the capacity of slender walls, it is first necessary to consider the strength of short wall sections. Figure 6.3 shows a plot of the failure loads of eccentrically loaded three-block high 6-in prisms. Loads were applied axially and at eccentricities of t/6, t/4 and t/3. Note that even though there is considerable difference between individual test points at the same eccentricity, no trend can be observed for the failure load to decrease with increasing eccentricity. Similar behavior has been observed elsewhere for solid sections of concrete as well as clay-masonry [8]. It is apparent that flexural compressive strength of masonry increases significantly with increasing strain gradients.

The 6-in block of which prisms were tested were used in the construction of the reinforced masonry walls. However, these walls also contained grout and steel, and the section capacity of a reinforced wall will depend on the strength and relative stiffness of all the component materials. The correlation between prism strength and wall strength is discussed below.

The average axial compressive prism strength computed from the 6-in prism tests, based on minimum net area, is 1890 psi. This stress, multiplied by the transformed area shown in figure 6.2 (191 in2), results in a computed axial failure load of 361 kip. This should be compared with the average 336-kip failure load for the 10-ft axially loaded reinforced walls. Thus the 10-ft walls developed approximately the predicted ultimate axial strength. This is a good correlation, considering the difference between individual test points.

It is somewhat more difficult to compare flexural compressive strength, since at each eccentricity the 6-in prisms developed different flexural compressive

strength, with an apparent increase in strength with increasing strain gradients. Average flexural com pressive strengths developed at an eccentricity of t: by the 6-in prisms and by the 10-ft reinforced wal respectively, are compared below. The transformed section in figure 6.2 was used to compute stresses it the reinforced walls. The following stresses were computed:

Average flexural strength of 6-in prisms, at t.3 eccentricity: 4,400 psi; Average flexura strength of 10-ft walls, at t/3 eccentricity: 2,900 psi.

Stresses were computed for a linear stress dis tribution. While the 6-in prisms developed flexura compressive strength which exceeded the strength, under axial loading by as much as 130 percent, the strength increase in the case of the wall was only 50 percent. Thus, there is good correlation betwee wall strength and prism strength under axial loading while under eccentric loading the prisms developed higher flexural compressive strength than the walls The question arises whether the strength of the 10-1 wall was reduced by slenderness effects. The failure mode of these walls indicates that they failed nea the top, where the eccentric load was applied. It wi be explained later that this type of failure is an ind cation that slenderness had no significant effect wall strength. The discrepancy between flexura strength of prisms and walls is probably caused b composite action of the wall rather than by sler derness effects.

Figure 6.4 shows the failure loads of the 10-ft hig) reinforced walls, plotted against applied moment