Weld Penetration 1

[bigmig]

I have a connection that the contractor was unable to bolt because of a field alignment issue.
It is a moment connection. Their solution was to add on this
external horizontal weld line. I'm not understanding how the weld can penetrate if
there is no bevel on one of the plates (or bottom flange). I see a single pass of 1/4 inch weld....but
I'm just thinking that most of it is just globbed on the side of the flange and column cap plate.

Thoughts would be appreciated.

 

[driftLimiter]

Youve said it all, you cant have a fillet weld without some throat, and you cant just create a throat on the outside of two flush plates. What a mess
Not sure how this is supposed to work for any decent moment. Even if you had two lines of 1/4" throat weld done properly.

The column cap was mostly there to facilitate the bolted connections. I wonder if it would be better off to remove the column cap and just do a direct flange CJP weld to the beam.
I am praying for you that this isn't a seismic moment connection. Godspeed.

I suppose one could try two lines of weld that were properly installed with joint penetration, but the load path from the bottom of the beam into the stiff plates is not clear or simple.

 

[JedClampett]

Grind the weld smooth and weld a plate of same web thickness all around, top and bottom flange and back to the web.

 

[CANPRO]

looks like poor planning and poor execution to me. Looks how far off the beams web stiffeners are from the column flanges. Something was either measured wrong, fabricated wrong, or erected incorrectly.

If understand correctly, I like JedClampett's idea of adding new web material. Jed, is this what you meant:

This will get you some moment transfer, not sure if it is enough. Also not sure how critical your misaligned stiffeners are.

 

[WesternJeb]

Just wanting to add that the weld doesn't even look good for what its supposed to be. I am seeing some porosity on the far left? Aside from the intent of the weld, it would have to be dang near perfect for my to accept it and it isn't.

 

[JedClampett]

If understand correctly, I like JedClampett's idea of adding new web material. Jed, is this what you meant:

Yes. You even put a little more than I envisioned.

 

[JStephen]

On the penetration question- you'll get some penetration regardless. If a specific amount is needed, it would be possible to gouge out first, then weld. If you have full-time weld inspection on the job, that inspector should be aware of what was done.

 

[EdStainless]

If there was no bevel, then the weld is just for looks.
Either grind flush and plate it as has been suggested or grind that weld out deep enough to get a real weld in there.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Flotsam7018]

I would do one of these options:
[ol 1]
[li]If the architectural constraints can tolerate a wider cap plate, cut the column back and replace the plate with a thicker and wider plate to allow for some loss of column length in the cut and grinding process and to allow for a fillet weld all around the beam/plate interface and column section.[/li]
[li]There is a procedure in AWS D1.1 for filling in holes, but it is fairly involved, though this could be done and the correct hole placement field drilled.[/li]
[li]Replace the column entirely and field weld with fillet/PJP/CJP as required.[/li]
[/ol]

 

[EngDM]

I am praying for you that this isn't a seismic moment connection. Godspeed.

Just curious as to why this would be bad? Is it just because welded connections tend to fracture and lose strength during cyclical (earthquake) loading? (NBCC2020 just started to mandate designing for earthquake in my lowlowlowlowlowlow seismic zone).

 

[driftLimiter]

Yes welded moment connections that are not properly detailed to preclude fracture (stress concentrations) do not perform well under cyclical loading. For ordinary moment frame systems such detailing is not required but for enhanced ductility the AISC 341 requires special detailing. Generally when we are in that category we choose from a set of pre-qualified moment connections that have tested performance parameters.

This frame may oray not require that detailing.

 

[TLHS]

Youve said it all, you cant have a fillet weld without some throat, and you cant just create a throat on the outside of two flush plates. What a mess

I think you potentially *could* have a fillet weld style weld in this kind of location. It's just wonky as hell. If you had someone build it up properly, there's a potential throat where the blue line is. It would be really sensitive to prying actions if there weren't balancing welds elsewhere but there is a throat and a load path.

There isn't enough material build up in the picture to give any significant capacity even if this were something that you should ever do.

Never do this, obviously.