Actual Loading Effect on Welding Length

[hungrydinosaur]

Hi All,

I am trying to figure out the actual stress distribution on a weld length. In traditional calculation, one can just divide the load by the weld area to find the stress. But consider the attached scenario (see attached), the load is on one corner of the beam (towards extreme left). The supporting weld is a long weld length reaching to the extreme right. My question is what will be the realistic welding distance on which the load will be acting? So that I can take that particular weld length for my calculations (not the complete length). All comments will be appreciated. Thanks in advance. Regards, HD

 

[dhengr]

Hungrydinosaur:
Given the proportions of your sketch, the bigger problem might be making sure the get the 3k load up into the thin web of the WF beam over sufficient length of connection. As long as you do that properly, then I’d take a weld length about equal to the depth of the beam and design it to take the 3k load, and make the rest of the double fillet weld maybe 3/16" ea. side. Basically, I’m assuming a 45̊ distribution of the load up through the beam web depth. You do want to protect the left end of your weld from any flaws or defects, or it could just tend to unzip, from the concentrated load, particularly in a dynamic loading situation. Weld terminations are always somewhat susceptible to weld defects. And, it is not a good idea to weld around the end of the red plate, or around a standing plate corner/edge. You almost always end up with a notch in the plate corner/edge, which is a significant stress raiser.

 

[desertfox]

Following dhengr's excellent advice I would add that the weld is in torsion as well as having direct shear placed on it.

 

[hungrydinosaur]

Thanks Dhengr, Desertfox. The load on the beam is a traveling trolley, so it will be acting as a point load. So it won't be possible to get the 3k load to a sufficient length of the beam web.

 

[kingnero]

@ desertfox: I don't understand the torsion, could you please elaborate?
It seems that the load and setup is symmetrical about the longitudinal axis.
I see shear, but no torsion. Am I looking at it the wrong way?

 

[desertfox]

hi kingnero

I was trying to account for the fact that the load is offset from the centre of the weld, so I looked at the right hand view and considered that the centre of gravity of the weld is approx. 300 mm to the right of the load, then I imagined it trying to pivot at the centre of gravity which made me think of torsion but on reflection maybe I'm wrong.