inspection of CJP at moment connection vs prequalified joint 2

[boffintech]

RE: inspection of CJP at moment connection vs prequalified joint

connectegr, I need some of that CWI support for a field welding/bolting issue...question: On a CJP beam to column moment connection, the prequalified joint designation shown is TC-U4a, but have a situation where the column appears to have cut a little short resulting in the bottom of the top beam flange being actually equal to the top of column cap plate. Not exactly the prequalified joint...

What should be the proposed fix on this?

 

[Duwe6]

You have found a 'bad one'. Immediatly draft an RFI - request for info - describing the condition you found and send it to either the Owner or the Engineer-of-Record. Include column #, elevation, a sketch with measurements of the mismatch, and a digital photo. I know that there is nothing that the Owner can do about this problem, but your contract may require to notify him of "significant findings"

There are a couple of simple "fix's", like extending the column. But those are WAY above your level of authority and authorization. You have done your job as the CWI: you found a significant erection problem that renders a Moment Connection useless. Now the Engineer [and probably the Fabricator] have to give the Erector a plan on how to fix this problem.

 

[connectegr]

Is there room, or can the deck be cut to allow a top flange plate? Is the moment taken in the column or is there a beam on the far side? Is there is a moment beam on the far side the flange plate can be narrower than the beam flanges and fillet welded to both flanges. If this is a lateral moment, the plate must be checked for compression across the unsupported width. If a gravity moment only, then the top plate is in tension.

How many sides of the column have moment connections?

Is the cap plate fillet welded to the column? They can arc gouge the fillet and replace the cap plate with a thicker plate. Remember if full moment is required the plate should be grade 50 to match the flange.

Does the moment force require the full thickness of the beam flange? If the bottom of the flange aligns with the top of the cap plate, a fill plate could be welded to the cap plate, with fillet welds in the direction of the load. The flange could then be welding with B-U4a weld. Note that the seam between the cap plate and fill with probable not fuse and show and defect in UT testing. If less than the full thickness is required and effective weld thickness can be given. This is basically a partial penetration weld.

If a plate on top is a problem. Two plates can be welded to the cap plate under the beam flange and on both sides, with CJP welds. The plates can be fillet welded to the toes of the beam flange. Note that these fillet welded have some eccentricity.

If the intent was to weld to the cap plate the weld should have been a B-U4a. The cap plate should have extended beyond the column, which would have allowed easier access for welding. This also allows a fillet weld for the column to cap plate. The bottom flange to the column flange is a TC-U4a.

Hope this helps...

http://www.FerrellEngineering.com

 

[boffintech]

Duwe6: Usually I just tell the welders what to do; I mean if anything, RFIs are WAY over, rated that is. That's why we mostly just ignore them, kind of like the general notes. Just kidding...just kidding...RFI, etc., all that is taken care of. If you guys would answer the RFIs with some witty commentary and whatnot about how you arrived at the answer well I wouldn't bug you here. I'm sure the RFI answer won't be as nearly as interesting or educational as what I read here.

connectegr: B-U4a at top flange, correct.

See attached sketch for other answers. Curiously the beam on the uphill side has correct fit up. Top of column cut at incorrect angle perhaps...

Also, what would you do if one end of a beam has the flanges beveled upside down?

 

[connectegr]

I am assuming you are not the EOR on this case... If you are looking for a maximum strength proposed alternative, I would recommend a plate across both across the flanges. Depending on beam size, you can narrow the width for the flange plate and match the cross-sectional area. The plate should be grade 50 to match the flange tensile strength. (if A36 the thickness should be increased proportionately). The plate length should provide enough lap to develop the tensile strength in shear with fillet welds, on both beam flanges. This is a pretty clean fix, but will increase the TOS. The EOR should approve the repair. Also using the actual design forces may reduce the weld sizes and plate thickness.

I agree that RFI's can be a none productive exercise. And I second your approach of suggesting a repair. In my experience EOR's can be short on imagination. However, the easy answer to your conservative suggestion, is usually YES. They won't suggest anything less. Ultimately this does require EOR involvement whether in a direct conversation, non-conformance report, or RFI.

I know just the right firm, if you need official help. Good luck

http://www.FerrellEngineering.com

 

[connectegr]

Sorry my last remark was not a "plug". But, beyond the input given would require a more professional involvement.

http://www.FerrellEngineering.com

 

[connectegr]

boffintech
I have attached a sketch of two options. I wish I could have drawn these with my iPad or iPhone. A picture is worth a thousand words. Does this satisfy your technically curiosity?

http://www.FerrellEngineering.com

 

[Duwe6]

Nice stuff, connectegr. I like the option on the right, personally. And since either will develop more KIP's than the originally spec'd simple moment conn, I would also dispense with an RFI. I would just send in a "Red-Line" of the as-installde connection.

It is easier [and faster] to get forgivness than to get permission

 

[connectegr]

Duwe6
Thanks

Respectfully, "easier and faster" can be a dangerous approach. I have spent many client's money fixing what they thought was a conservative field repair. Just saying...

Both details can be developed to match the moment capacity of the beam. But, they are incomplete as shown. And there are pros-cons to each. If the moments are given, then a real solution can be developed. However, the erector rarely has enough information.

http://www.FerrellEngineering.com

 

[boffintech]

RFI written...
RFI transmitted...
Pencil sharpened...
Severe and intense analysis of numbers...
Much cogitation...
More severe and intense analysis of numbers...
Conclusion drawn...
Papers shuffling...
RFI returned...

MC = deleted

Dang.

 

[connectegr]

Even if the EOR did not intend on a moment connection, he probably needs one now. The fixed connection on the far-side of the column, will induce a moment in the column. The column probably not designed for the moment. The EOR can choose whether the intend is a moment frame or not. But, somewhere in the middle is not a good choice. If the structure is analyzed as simple beams, the other fixed end connections are a big problem. This is not one of those more connection is better situations.

http://www.FerrellEngineering.com

 

[connectegr]

Without the moment connection on both-sides, the bay is basically a very tall portal frame. Simply releasing the moment on one side does not release the welded fixed end connection on the far side. This creates a large moment in the column and in the column baseplate. For this to work with the removal of the MC connection, the lateral system is redundant or very, very conservative.

When I get "incredible" RFI answers, I usually send the Cover My A--, RFI. Again I am not a big fan of the RFI process, but they do serve a liability purpose.

http://www.FerrellEngineering.com