Plastic vs elastic section modulus for welds

[Kom3]

For welds in bending, should we use the plastic section modulus or the elastic section modulus? I understand welds are more brittle in nature but per the AISC, we are already applying a 0.6 reduction factor. Does this already account for the brittle/shear nature of welds and hence we can use the plastic section modulus for our calc purpose?

Regards.

 

[skeletron]

For welds in bending, use the elastic section modulus "S".
The 0.6 factor (0.6 x Fexx ?) accounts for the shear strength.
There is an experimental method that I have been exposed to which checks the plastic capacity of a weld in bending. In very general terms, it is a non-linear relationship, was based on tension/shear/geometry not plastic section modulus.

 

[Agent666]

Note you're also typically using the ultimate tensile strength of the weld metal in weld design, it's not some yielding approach that is capable of redistribution. It is a brittle failure mechanism in simplistic terms.

Fillet welds if working to this ultimate strength are not capable of supporting a level of ductility in the welds before failure to ensure that you might be able to develop a substantial plastic distribution of force in the base metal(unless you're designing the welds to be stronger than the base metal being connected). It is for this reason you really must utilise an elastic modulus-based approach. For example, if you had a 20mm plate welded with 4mm FWs, clearly the plate is going to be stronger than the welds attaching it, so once the ultimate capacity of the weld is exceeded at the point of maximum resultant stress the weld starts to fail.

The weld in this scenario is not able to yield and then sustain the joint strength at the weld capacity whilst reaching a plastic distribution of force over a longer length of the weld.

But if you ensure that the weld is stronger than the attached part of the section so the section can effectively reach its plastic capacity whilst having a margin of strength remaining in the weld (i.e. weld stays within the elastic range or below the ultimate strength of the weld) then the part attached can at least reach it's limiting plastic strength. For example, if you had a 6mm plate with 10mm FWs, the weld is stronger than the plate, so the full capacity of the plate in say tension can be realised.

https://engineervsheep.com

 

[Tomfh]

Agree. The ends will start unzipping prior to the whole length yielding. Use Elastic modulus.

 

[271828]

Please clarify "welds in bending."

For eccentric loading, the AISC Manual Part 8 has an elastic method that sometimes would use S, and an ultimate strength method based on rotation about the instantaneous center. Either is OK to use. Elastic is more conservative, often described as overly conservative. Thus, the design aids in the Manual use the instantaneous center method. I'm not sure you're typing about typical eccentric loading, though.