Eccentric Loads on Weld Groups

[chicopee]

While perusing thru my manual of steel construction-ASD- I came across the Ultimate Strength Method showing me exponential functions for R values. Under what condition is this analysis carried outin light of the fact that when I come across such problem I'll use the more conventional way of calculating weld size based on torsional stresses and polar moment of inertias.

 

[DaveAtkins]

Your conventional method is fine--just more conservative than the method in the steel manual.

DaveAtkins

 

[JAE]

I agree with DaveAtkins - look in your AISC Steel Manual (if you have one) and in the connection sections there is an example of this and short discussion stating exactly what DaveAtkins said.

 

[UcfSE]

You can choose either. If you have enough data you can use the ultimate method which based on the instantaneous center of rotation. It is a longer method that better captures the behavior of the welds (and fasteners for that matter) at failure and hence is much better suited to an ultimate load approach to design.

The elastic approach is better suited to hand-written calculations. The assumptions don't really follow the true behavior of the welds but it errs on the conservative side.

If you need to size one weld or a few welds, either method it acceptable. If you have a process or connection that will be replicated many times, you may be able to save the client some money with a less conservative calculation.

 

[yakpol]

Chicopee,
Check out weldgroup spreadsheet on

http://www.yakpol.net

. It does it both ways, and it's a pretty big difference in capacities calculated by LRFD and ASD.

 

[chicopee]

Yakpol- I can not believe the diference in results. The LRFD capacity was at least twice as much as the ASD capacity. Can that be right? It is definetly a well constructed spreadsheet, though.

 

[yakpol]

Well, the instantaneous center method is based on the hundreds of tests and curve fitting, the elastic is based on typical rigid body rotation analogy not allowing any distribution of stresses due to plastic action. ASD and LRFD methods for bolts are closer, from 15 to 35% difference. Amazing that AASHTO code still likes elastic method saying that factor of safety is more consistant there. No way.

 

[miecz]

AASHTO is very concerned with limiting the stress range in bolted splices, which undergo significant stress reversals, and not so concerned with the ultimate strength of the splice.

 

[yakpol]

jmiec, you may be right. Per AASHTO bolt connections are designed for ultimate load (bearing, shear) and for service load (friction) but there's no explicit requirements for fatigue. Some of the state DOT's allow AISC LRFD method to be used on the bridges.