Multiple Web Plates for Built Up Section

[bhengnr]

I have a question regarding a built up section. I am attempting to increase the capacity of a crane from 3 tons to 5 tons. The crane runway beams need to be reinforced. LTB controls, so I am adding web plates on both sides to create a box section.

Existing runway beam:
W21x62 w/ C12x20.7 cap channel
52' Span

Reinforcement:
1/4" web plates each side of web full length

If web plates are only 10' in length(required from fabricator), do I need to provide a CJP weld at the butt joints of the web plates? Is it possible to not have a CJP weld at the butt joints? Would a PJP suffice?

My initial thought is that a CJP would be required at the butt joints to provide the same section properties throughout the beam. Does anyone have an explanation that would justify CJP welds are not required?

 

[KootK]

Is it possible to not have a CJP weld at the butt joints?

It's certainly possible from a mechanics of materials perspective. All those welds need to do is resist whatever vertical shear winds up in the plates as result of the LTB demand and whatever direct torsion you need to deal with. So long as it works for fatigue, which is important, I imagine that just about any weld would work for strength concerns.

Does anyone have an explanation that would justify CJP welds are not required?

Fatigue aside, why do you feel that CJP welds may be required? I may be able to do a better job of responding to your concern if I know, specifically, what your concern is.

 

[DrZoidberWoop]

Is the cap channel oriented "flange toes down"? It may be difficult to install exterior web plates if there is already a nominal weld btw the cap channel and W or S-section.

You can use a cjp at the butted plates, but it will be a more tedious fabrication detail. For instance, B-1a-GF would require internal backing to be abandoned within the built-up section. Unfortunately, the butted connections can't be reached for backgouging, unless you know a welder with 10' arms. If it can handle your loads from first principles, a butted joint w/ a BTC-P4 PJP would be preferable for a fabricator.

We rarely use less than 3/8" plate for anything structural, except shim plates, because they permit the max single-pass fillet welds (5/16") for most processes and resist distortion much better than 1/4" plate.

 

[bhengnr]

KootK - equation F4-5 in the steel manual is for the critical stress. One of the variables that is needed is the torsional constant, J. It makes sense in my head(FYI, EIT here) where the butt joint exists, the J value would stay the same as the existing beam. This would be the weak point. CJP is belts and suspenders; I know it would be more than enough to keep the same J value across the beam.

DrZoidber, cap channel is oriented toes down. The flange plates would be internal - 1.5" from outside edge of the flange. Fabricator said that would be enough room for a weld.

 

[KootK]

It makes sense in my head(FYI, EIT here)...

We love EIT's here. Regardless, thanks for the heads up. It's always good to know who you're talking to.

...the J value would stay the same as the existing beam. This would be the weak point. CJP is belts and suspenders

Your weak point isn't weak at all. Truly, if it were not for fatigue issues, I'm sure that a 1" seal weld in the middle would do the job. That, particularly for the LTB case as that's a second order kind of thing that will produce a very small torsion demand. If your PJP satisfies your fatigue detailing, that sounds like a good solution.

Mechanically, you could even just fillet weld a shear splice plate to join the adjacent web plates. I'm not sure that would be kosher for fatigue though. Perhaps someone with more crane experience, like dik or FLCraneBuilder can comment on that.

 

[DrZoidberWoop]

Is this approximately what you are trying to accomplish?