What weld to specify when face of column is only 1/4" away from edge of base plate? 1

[MJC6125]

I'm curious what type of weld other engineer's would specify in this scenario. I have an HSS column welded to a column base plate. The column is hidden in a stud wall, so the size of the base plate is limited. There's only room to extend the edge of the baseplate by 1/4" past the face of the column per the image below.

What type and size of weld do you think is acceptable, not from a design load perspective, but from a code/weldability perspective?

I think there's two sections of the AISC manual/spec that may control. The first is Minimum Shelf Dimensions for Fillet Welds on page 8-17 of the 14th addition of the AISC manual. It recommends the shelf to be 1/4" larger than the weld size. With only a 1/4" shelf in the base plate example, you couldn't put any weld there. The image, Fig. 8-11, does not depict this scenario exactly so I'm not sure if it applies. The second is the limitation of fillet weld sizes along the edges of material per section J2.2b of the AISC spec. If that section needed to be followed you could only do up to a 3/16" of a weld. I don't think that section applies because you aren't technically welding along the edge of a plate, but I'm not 100% sure on that assumption.

Which rules do you think apply? Would you do a fillet weld of some size or would you change it to a PJP groove weld?

There's already been a thread on this where a PJP weld was recommended: Link

 

[BAretired]

If you need it structurally, use a PJP weld, otherwise use a fillet weld all around.

BA

 

[XR250]

If it is in a stud wall, my suspicion is that your only loads are gravity, a nominal shear load and maybe some uplift. Welding on two sides should be plenty.

 

[skeletron]

Have the plate flush to the column and butt weld it.

 

[KootK]

If it is in a stud wall, my suspicion is that your only loads are gravity, a nominal shear load and maybe some uplift. Welding on two sides should be plenty.

If it's not galvanized, it's this for me.

If you need a weld on the other two sides for some, non-load reason, maybe just spec a seal weld and see what you get.

 

[phamENG]

I agree with XR. Weld the two sides with room. You'll have essentially no moment resistance in the y direction (as drawn), so I can't imagine you'll have any applied shear in that direction. Shear in the other direction can take advantage of the strength increase for fillet welds loaded 90^o to their axis - same with the resultant shear from a moment about the y. If you still need some, then a partial joint penetration or butt weld should do the trick.

 

[hokie66]

I would just specify a 1/4 fillet all around. Best to close the column, and that is a simple instruction to the fabricator.

 

[Ron]

Agree with hokie66 except I would specify a 3/16" fillet to keep the weld profile code compliant.

 

[dik]

A simple fillet weld would suffice... what loading do you have? all compression? heavy or light? From my notes for HSS...

-IF A FILLET WELD CANNOT BE USED ALL AROUND, A V-GROOVE WELD SHALL BE USED WHERE A FILLET WELD IS NOT POSSIBLE. IF WELDING TO A RADIUSED CORNER, FLARE BEVEL WELDS TO BE USED. FLARE BEVEL WELDS TO BE FLUSH WITH FACE OF HSS SIDEWALL.


Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[MJC6125]

Thank you for the responses. I do have some cases when I have uplift on these baseplates, or they see shear load because the column has an HSS header over a storefront connected to it. I'm going to look at those cases closer with using a PJP or butt weld.

FWIW, I don't think the 1/16" less thick fillet weld requirement per section J2.2b of AISC applies actually. Per the image below that's a requirement because the top edge of one plate in a lap joint can melt away causing the weld to look thicker than it actually is. I feel like if the edge of the base plate were to melt away that would be obvious because that portion of the base plate would look thinner than the rest of the plate. I'm not 100% confident on that since I've never actually welded anything in my life, just drawn symbols on a drawing and seen other people do it.

 

[canwesteng]

Why wouldn't you have the exact same problem of the base plate melting away here? Sure you could see that it happened but then do you throw the assembly out? I would go 3/16" because it seems easier to do, although I have never really welded much.

 

[Ron]

MJC...yes, the provision does apply. The weld must have a proper profile as defined in AWS. If you melt the edge off the baseplate, you won't have the proper profile, thus you will have a weld that does not comply with the building code.

 

[MJC6125]

I was thinking they would have to take the plate off and try again with a new plate. Or add more weld material to where they have the proper thickness. They wouldn't leave it as is.

 

[Enable]

I'm sorry but you are now trying to invent problems where the doesn't need to be any (LOL at a fabricator throwing out a burned plate). Call up a PJP for the loads required and be done with this.

Though, as a curiosity only (see PJP directive above) W59 in Canada allows for what you are describing. As long as the weld size is clearly visible it is seemingly allowed.

(b) 2 mm (1/16 in) less than the thickness of material, for material 6 mm (1/4 in) or more in thickness, unless the weld is designated on the detail drawing to be built out to obtain full throat. In the as-welded condition, the distance between the edge of the base metal and the toe of the fillet may be
less than 2 mm (1/16 in), provided that the weld size is clearly verifiable.

 

[canwesteng]

If they don't throw out a burned plate with a bad weld, what are they doing? Grinding the weld out and filling in the plate?