Moment Splice of Beam

[C Hietbrink]

I am doing a retrofit of a structure that prevents a full length beam from being brought into the building. Therefore, the beam will need to have moment splice connections in one or more locations. I was looking to have the web plate connection as a bolted connection to assist in the erection process and the flange plates be welded. Is there any restriction per AISC that prevents the use of this type of bolted and welded connection at a moment splice? Everything that I have read only discusses welding all of the flange and web plates or bolting all of the web and flange plates. The beam will be installed under an existing composite deck; therefore, through bolting the top flange of the beam is not really an option so a fully bolted connection is not possible.

 

[rb1957]

If you want a fully welded joint, you could install the web plates with temporary fasteners, then weld everything and maybe leave the temporary fasteners in place or remove and plug weld.

another day in paradise, or is paradise one day closer ?

 

[C Hietbrink]

It is not that I necessarily want a fully welded connection because I could always specify that just as you said; however, I am wondering if there is any specification or code that prevents the use of a bolted web plate connection in combination with the welded flange connections. The shear at the splice point is much less than that of the moment so a bolted connection will easily work.

 

[rb1957]

you're splicing the upper cord with straps on the inner face of the flange ? or angles

another day in paradise, or is paradise one day closer ?

 

[C Hietbrink]

I am splicing a W section and because we are installing it under an existing deck, I can't weld a single plate to the top of the top flange. In lieu of that I am welding two plates to the bottom of the top flange along with the typical single plate to the bottom of the bottom flange. I was just curious if I can specify a bolted web plate in combination with the welded flanges or if they all need to be welded?

 

[rb1957]

cool, what I thought.

Personally I don't see a big problem, but if experience shows either all welded or all bolts designs then maybe there's something there. How would you bolt the web plates on ? before final placement ? You could use CSK bolts (not preferred, I understand) ?

Why bolt the web plates, in preference to welding them ?

another day in paradise, or is paradise one day closer ?

 

[C Hietbrink]

The web plates would need to be bolted on during the erection process because this is all happening in a crawl space with limited room to work and bring in the beams. They could be bolted on and then the beam jacked into place, connected at the girders, and then weld the flange plates. The preference of bolting to welding is because some of the beams are not very deep and to make the weld of the plates on the bottom of the top flange and the top of the weld plate may be a little challenging. Not to mention it just may make erection a little easier not to have to weld this plate.

 

[rb1957]

I get bolting the web plates is a reasonable solution, and I'm sorry I can't add more about the mixed design.

Maybe using angles instead of straps (to get around the flange welds impacting the web welds) ?

another day in paradise, or is paradise one day closer ?

 

[KootK]

For special moment frame joints, it is now considered poor practice to use bolted shear connections in combination with welded flange connections. Under cyclical plastic hinge development, the play in the bolted web shear connection tends to shift the resistance of shear to the flange welds which screws with their performance. This is one of the lessons learned from the Northridge earthquake. This may be the code limitation that you're trying to tease out. In your particular application, I know of no code prohibition on the use of the mixed system and would have no qualms about using it myself.





I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[MGaMart]

When dealing with combining bolts and welds in the same plane, you do need to be quite careful. For the same reasons KootK mentioned, unless the bolts in the web are designed as slip critical, the hole tolerance may lead to the flanges taking on a significant portion of the shear load prior to bolt bearing being initiated in the web. This would lead to a large reduction in the flexural capacity of the beam you are trying to splice. Bolts and welds can be used in combination, however doing so comes with a reduction in the shear capacity of the bolt (which results from a comparison of the relative stiffness of the weld and bolt). It's a much safer practice to go all bolts or (in your case) all weld. I can't speak for AISC, but the CISC has provisions for bolts and welds in combination. IMO, combining should only be done as a worst case scenario when you need to squeeze out as much resistance as you can afford (often within confined/restrictive connection conditions)

 

[C Hietbrink]

Thanks for the responses, That is a good point. I have decided to specify a welded plate but I was going to give them a couple of bolt holes to assist with erection.

 

[jike]

How about bolted end plates like is used by the PEMB manufacturers? If it is a simple beam than the top is in compression and only the bottom needs to be extended below the flange to fit all the bolts. The top end plate can be flush with the top of the beam. I would use a minimum of 6 bolts (2 just below the top flange and 2 just above the bottom flange and 2 just below the bottom flange).

 

[SAIL3]

good decision to go with a welded web pl and since you are welding the fla pls anyway the welders will be available and on site.....