Comparison of Predicted and Tested Shear Wall Capacity

Comparisons between the shear wall capacity measured during the laboratory testing and that predicted using the analytical model were performed. The intent was for the model to predict the shear wall capacity Sy, Waii at the level of the yield shear strength, Sy. At the same time, in order to verify that the elastic model by Kallsner & Lam provided the most reasonable solution to predict the shear wall capacity, other models that were based on different assumptions and which have been applied in the prediction of wood frame shear walls, were also contained in the comparisons. These models, which were presented by Chen (2004), include Kallsner’s & Lam’s lower and upper plastic models, as well as models by Easley and McCutcheon. In this paper, only the test-to-predicted shear capacity results for the elastic Kallsner & Lam
model have been listed (Table 5). Each ratio and the associated statistical information represent the 16 wall configurations and a total of 103 individual shear wall test specimens.

Table 5. Full-scale shear wall test-to-predicted shear capacity (Chen, 2004).

Monotonic Loading Cases

Kallsner & Lam Elastic Model

Ratio

SD

COV

EEEP 12.5

1.270

0.181

0.143

EEEP 25

1.050

0.122

0.116

Max. Load 12.5

1.093

0.162

0.148

Max. Load 25

0.918

0.117

0.128

Cyclic Loading Cases

Kallsner & Lam Elastic Model

Ratio

SD

COV

EEEP 25

1.012

0.129

0.127

Max. Load 25

0.885

0.122

0.137

In order to specify the connection test data that would best predict the shear wall capacity accurately, four cases were considered for the model for monotonic loading, i. e. EEEP 12.5, EEEP 25, Max. Load 12.5 and Max. Load 25; and two for cyclic loading, i. e. EEEP 25 and Max. Load 25. The shear capacity per connection, Syconn, was represented by the EEEP yield capacity or the maximum shear load; at the same time, two edge distances were considered, namely 12.5 mm and 25 mm. These cases correspond to the connection strength data listed in Tables 2 & 4. Based on the monotonic test predictions, the cases with an edge distance of 12.5 mm were not included in the predictions of cyclic tests. All of these cases were considered because it was not known which one would best represent the behaviour of the connection in the prediction of the performance of a full-size shear wall. The prediction using each model under each case was then compared with the average shear capacity, Sywaii, of the tested full-scale walls with the matching sheathing configuration. An average shear capacity was obtained from the three or more tests that were performed for each wall configuration. Due to the large amount of the data, only the combined test-to – predicted ratios are listed in Table 5. As can be seen, the predicted shear capacity obtained using the test data from connections with a 25 mm edge distance and where the EEEP method was relied on to obtain the connection strength are the most accurate. A more detailed discussion of the comparison is provided by Chen (2004).