Direct Weld Beam Flanges to Column Flanges

[YoungGunner]

I'm working out of RISAConnection to design a connection for a client who said it would be easier to directly weld the flanges of a beam to the flanges of the column than do a BFP (beam flange plate). The loads were incredibly small for our OMF so the BFP had double 1/4" fillet welds for its plate-to-column attachment and I'm guessing the client expected the same with the flange-to-flange connection. However, RISAConnection does not have an option to do a fillet weld for the moment connection of the beam flanges to column flanges - it only allows for PJP and CJP welds. I'm tempted to just allow a fillet direct weld to work despite Risa but I was curious if someone had insight as to why RISA doesn't have that option in the first place before I proceed?
Column size is W10x60, beam is W14x61, moment is 129.24 kip-ft.

 

[DoubleStud]

Probably because you have 0 tolerance?

 

[skeletron]

Have you calculated the required fillet to develop the flanges?

 

[GC_Hopi]

The direct weld has its place. I have used this quite a bit for small shop built frames or one off connections in the field where you have a cantilever or a kickers supporting the other opposite end.

Sounds like you need to run the numbers by hand.

 

[WinelandV]

I would venture that RISA Connection only allows PJP and CJP welds for the flange because it's meeting a requirement from the AISC 341 for directly welded flange moment connections.

Please note that is a "v" (as in Violin) not a "y".

 

[canwesteng]

Why are you mixing flange welds with a shear tab? Might as well just weld the web to the column, I'd be marginally concerned the flange welds end up stiffer and attract some shear too otherwise. No issues otherwise with fillet welds T+B, but you should consider erectability.

 

[YoungGunner]

Double stud - I didn't know tolerance was a thing in RISAConnection - am I missing a toggle?

Skeletron - this used to be a BFP where the flange plates were welded to the column flange with double 1/4" fillets. I would assume that same weld (with minor adjustments) would be adequate for just directly welding.

canwesteng - I'm just showing you the direct weld beam/column option from RISAConnection. I would've thought though that the shear tab would be nice to hold the beam in place before they weld the flanges.

 

[YoungGunner]

winelandv - Yeah, I have found that RISAConnection strictly follows those provisions for connections based on special moment frames and hasn't been a great tool for OMF connections. It gets me most of the way there but it doesn't quite finish it up nicely. So with that in mind, I could probably justify directly welding right since I don't need to adhere to special moment frame connection guidelines?

 

[dik]

Would not a bolted moment end connection be less costly? Use a shim, if necessary.

So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[YoungGunner]

dik - are you saying a bolted moment end connection be less costly than the BFP or the direct weld? I thought the BFP would've been cheaper than the direct weld at least. We can't do a bolted moment end connection though due to geometric restraints in this scenario.

 

[phamENG]

YoungGunner - I'm pretty sure DoubleStud isn't talking about the software. A fillet weld has a max root of something like 3/16" while a square groove weld or single V-groove weld has a fit-up tolerance of +1/4" and can have a detailed root of up to 1/2" (or thicker for thicker flanges). So if you have fillet welds at both ends, the TOTAL cumulative erection tolerance (column plumb, column straightness, beam length, etc.) for both ends of that beam is 3/8". With a single V-groove, it's 1-1/2".

If this is small and being built on the floor of a shop, sure...direct fillets aren't bad. But if this is a large structure, give the erector a break and give them a little more room to play with.

Direct fillets are fine on your flange bolted connection because those are shop welds, not field welds.

 

[canwesteng]

pham - Depends on the frame - giving essentially 0" of tolerance might be fine, they will just make it up by tilting the columns. That assumes the only other members the frame supports are simple beams and they can be detailed with slotted connections.

Re: the shear tab, I had assume this would be a shop weld detail. Is you client the fabricator or the erector? Because the detail is cheaper for the fabricator for sure if it's a field weld, but I don't see how it is cheaper for the erector. Do you have a bunch of welding in the air elsewhere on the job?

 

[JoshPlumSE]

I was the project manager for the RISAConnection project some years ago. Back bolted end plate moment connections were being added / expanding, and when vertical braces were being added as well.

I handed this off to another engineer when the program got into seismic moment connections. Not because I wasn't qualified for managing that project, but because management wanted to give some of the less experienced engineers a chance to grow into more responsibility. Therefore, I not familiar with the reason for this restriction. But, I can guess.

a) I suspect that the Direct Weld connection was added specifically as a variation of the WUF-W connection for IMF / SMF frames.
b) I suspect that even for OMF frames you might have a requirement for PJP or CJP welds for those flanges. But, I didn't look this up before hand. This with be in AISC 341, not 358.
c) For an R = 3 connection which doesn't rely on AISC 341 detailing requirements then I would think that double fillet welds should be allowed. And, R=3 should be the assumption whenever the seismic detailing is set to NONE rather than something more specific.
d) This should be a relatively simple "problem" to fix for RISA. I genuinely think it's an oversight based on the the order in which the development was done... leaving in an AISC 341 restriction for an R = 3 connection.

All the development for RISAConnection was done by the RISA development group in the Ukraine when I was there. The project manager was always a registered engineer in the California office. But, none of the US based developers worked on the program at all....

Now that the Ukraine is embroiled in this terrible war, my guess is that the developers are not able to work. Maybe they've fled the country or something. I don't know. But, this may have affected RISA's ability to do any real programming work on RISAConnection.

Normally, I'd say to ask RISA directly. But, my guess is that management will be less than forthcoming with their customers on any issues related to their reliance on the Ukraine group. Heck, management wasn't particularly forthcoming with their employees once Numetcheck started running things.

To be honest, I'd have to admit that I never officially worked under the new Numetcheck regime. However, I was there until the very end of the old RISA. So, I saw all the changes they were planning, how they presented that to the employees and how they instructed employees to interact with customers.