I was running a design for a WT vertical brace and found what I believe to be a mistake in the calculation of the shear lag factor, U, for tensile rupture. I have (7) rows of bolts connecting the WT flange to the gusset. From AISC Table D3.1 I would expect my shear lag factor to be the larger of Case 2 (U=1-x_bar/L) or flange portion of Case 7 (U=0.9).
However, it appears that for Case 2, the x_bar being used in the program is not the distance from the faying surface at the flange to the centroid of the member as expected. Rather, it is using the distance from the geometric center (depth/2) of the section to the centroid. Unless I am missing something, the distance from the geometric center to the centroid is arbitrary in this case and the distance from the faying surface to the centroid should be used.
For Case 7 it also appears that it is using the web connected case (U-0.7) instead of the flange connected case (U=0.9).
Is there any guidance on how I should interpret these results? Has anyone else run into this issue?
Note: I am using Connection Standalone Version 10.0.0.177
I do not appear to reproduce these problems in a sample model here. I see that x-bar is taken as the distance from the faying surface to the centroid of the member and see U=0.9 for the flange connected case.
Please generate a Bug Report using the instructions on the web page below and include the model file with report. Then, use the Secure File Upload link at the bottom, left of this web page to share this file and post a message here to let us know it has been uploaded.
http://communities.bentley.com/products/structural/structural_analysis___design/w/structural_analysis_and_design__wiki/10032.creating-a-bug-report
See attached for the bug report with model file. Let me know if you need any additional info.
Thanks for the help!
BugReport.zip
I will log this as a defect and alert the development team. The connection associated with this issue is shown below:
The brace is a WT 7x30.5 and I see the same x_bar is used for the case where the brace is oriented at 90 degrees and connected to the flange (as shown above) and for the case where the brace is oriented at 0 degrees and connected to the web.
The x-bar appears to be calculated based on the angle shape remaining after the WT is split in half. U is calculated assuming the WT is always connected to the web. The 90 degree rotation of the brace does not appear to be recognized by the program.
In my tests, I used a WT brace oriented at 0 degrees and connected to the gusset using claw angles. In this case, the program is using the correct x-bar and U when designing the brace.
Thank you for bringing this to our attention.