LIstress
Structural
- Dec 19, 2004
- 26
(My first post of this - from yesterday - has disappeared. I don't know if it is something I did, or if the a moderator deleted it for a legitimate reason. If the latter, please let me know where I went wrong. Anyway, here's my post again.)
I'm a structural analyst who has to work with welds from time to time, and it is not my area of expertise. I'm hoping to get some advice from the welding experts out there.
There seems to be a fairly well developed methodolgy in place for the analysis/design of both groove and fillet welds in steel (as in Shigley's and Blodgett's books), but, aside from tables of allowable strengths (both as welded and post weld heated), I have not been able to find an analogous methodology for analyzing/designing aluminum groove welds with multi-component stress states.
I'm particularly interested in answers to the following questions.
1 - What "measure" or traction does one compute first, for subsequent comparison to an allowable strength? Does one use a square root sum squares value (like Blodgett for steel), or should one be using von Mises or principal stress, or something else?
2- Once one has an appropriate measure in hand, what is the appropriate aluminum "allowable" to compare it to? A percentage of the material's shear stress (as is often done for steel), a reduced 1-d as-welded tensile strength from test data (such as one might find tabulated in a textbook or AWS handbook), or test data allowables specifically for multi-dimensionally loaded welds (are these even out there)?
3- How can one best use computed FEA results (be they from a shell element model or a solid element model) to assist in analyzing/designing a weld? I believe modeling the weld itself and computing stresses is probably not recommended given all the unknowns (residual stress, inclusions etc.), but rather an extraction of appropriate FEA results for for subsequent use with a classical, empirically based methodology might be better. What I'd like to know is what are the appropriate results to extract, and precisely how could I use them to support aluminum groove weld analysis/design?
In general, I haven't had too much trouble locating tensile strengths for aluminum groove welds, but I haven't been able to find anything (especially with example calculations) that show how to use that data for a multidimensionally stressed actual weld, in a manner similar to Blodgett or Shigley.
Any advice, links, references are welcome and encouraged. Thanks.
I'm a structural analyst who has to work with welds from time to time, and it is not my area of expertise. I'm hoping to get some advice from the welding experts out there.
There seems to be a fairly well developed methodolgy in place for the analysis/design of both groove and fillet welds in steel (as in Shigley's and Blodgett's books), but, aside from tables of allowable strengths (both as welded and post weld heated), I have not been able to find an analogous methodology for analyzing/designing aluminum groove welds with multi-component stress states.
I'm particularly interested in answers to the following questions.
1 - What "measure" or traction does one compute first, for subsequent comparison to an allowable strength? Does one use a square root sum squares value (like Blodgett for steel), or should one be using von Mises or principal stress, or something else?
2- Once one has an appropriate measure in hand, what is the appropriate aluminum "allowable" to compare it to? A percentage of the material's shear stress (as is often done for steel), a reduced 1-d as-welded tensile strength from test data (such as one might find tabulated in a textbook or AWS handbook), or test data allowables specifically for multi-dimensionally loaded welds (are these even out there)?
3- How can one best use computed FEA results (be they from a shell element model or a solid element model) to assist in analyzing/designing a weld? I believe modeling the weld itself and computing stresses is probably not recommended given all the unknowns (residual stress, inclusions etc.), but rather an extraction of appropriate FEA results for for subsequent use with a classical, empirically based methodology might be better. What I'd like to know is what are the appropriate results to extract, and precisely how could I use them to support aluminum groove weld analysis/design?
In general, I haven't had too much trouble locating tensile strengths for aluminum groove welds, but I haven't been able to find anything (especially with example calculations) that show how to use that data for a multidimensionally stressed actual weld, in a manner similar to Blodgett or Shigley.
Any advice, links, references are welcome and encouraged. Thanks.
