Attaching WT reinforced W21x62 to column web?

[StructureMan44]

An existing W21x62 is being loaded to ~150% of its capacity. As seen in the attachment, we’re attaching a WT10.5x91 to the beam’s lower flange to provide the additional strength. This beam will only see positive moment. Assuming the welded/bolted connection between the W21x62 and W14x398 web is acceptable for shear, is there any reason to provide a connection between the WT10.5x91 and the W14x398 web?

 

[JAE]

No - no reason since the shear would ultimately extend into the web of the beam.

A few other observations:
1. Your weld symbol (WT flange tip to column) is drawn on the top of the line which implies "other side" of the arrow for the weld. Your weld is on the arrow side so the fillet weld symbol should be on the bottom of the weld symbol line.

2. The weld connection from WT to beam is a stitch weld. AISC requires, for cover plates, a continuous weld near the ends of any added section to a beam. I'd recommend adding a comment to your weld to require 24" of solid weld at the ends.



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[JLNJ]

The enormous tee seems like an add choice given you are looking for just a 50% increase.

 

[hokie66]

What is the purpose of the flange to flange welding?

 

[jike]

Seems like the Tee might be kinda large compared to the W21. Can the beam LL be removed before welding?

The stitch welds should be designed for VQ/I and the end welds for MQ/I.

I see no purpose for welding the TEE flanges to the column flanges. Perhaps a seat angle underneath the TEE can be used for erection seat, if needed.

 

[StructureMan44]

Thanks for the comments folks.
JAE: Thank you for pointing out the backwards filet weld symbol and the recommendation on 2ft of continuous end weld.
JLNJ: The WT is increasing the capacity to about 1.7x the original capacity
Hokie66: Originally the thought was the weld would act as a lateral brace and the welder will be in that area anyway but the weld between the flanges is not necessary.

 

[StructureMan44]

Thanks Jike, the beam currently doesn't have any LL (it will after the added WT). I didn't find any specifics in AISC 325, is it recommended for the contractor to remove DL from the beam or attempt to create a zero slope in the beam before adding the WT? Jacking from the bottom will be difficult if welding is done at that same time as the jacks are placed.

 

[jike]

No need to take out DL or jack the beam level. As long as you combine stresses properly.

 

[JLNJ]

If your connection is OK, you can probably stop the WT well short of the support and make it easier to install.

For a given tee depth, there are diminishing returns for making the tee heavier and heavier because it's the (unreinforced) top flange compression which controls. The 91 #/' tee you have specified increases the compression modulus 100%. A WT10x31 gives you a 65% section modulus increase for 1/3 the weight.


You might want to double check your combined section numbers.

 

[KootK]

Originally the thought was the weld would act as a lateral brace

This is a an excellent article on this stuff: Link. Interestingly, in figure 3, they discuss the need for vertical plates in partial reinforcement schemes to stabilize the WT reinforcing which seems to be somewhat in line with your original thinking. I will confess, however, that I don't actually know how to assess the need for such stabilizer plates.

From a constructibility standpoint, I might pull the WT back a foot or two from the column and truly make it partial reinforcement.

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.

 

[StructureMan44]

jike: Thank you for your feedback. Because the original beam will have deflected due to the existing imposed DL, I assume the new composite shape won't have a uniform distribution of stress, the existing beam will have greater stress.

JLNJ: Thank you for your feedback, how did you calculate the 100% increase in compression modulus with the WT10.5x91?
I think your comment refers to a heavier WT not helping to provide resistance to lateral torsional buckling. For calculating Lr (inelastic LTB unbraced length) is there a simple, conservative way to calculate J, the torsional constant, for a composite shape like this?

KootK: Thank you for sharing this article. The vertical plates they suggest would increase the torsional resistance which I may need, if I can figure out how to calculate J.

 

[KootK]

J will be easy. Just the sum(bt^3/3) business. It'll be Cw that's tricky. At the beginning of the article that I linked, they mentioned some other reinforcement strategies that improve LTB resistance (channel, double angle, side plate).

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.

 

[dhengr]

StructureMan44:
Actually, your detail is a fabricator’s and erector’s nightmare. The fabricator doesn’t know how to fit the WT flange to/btwn. the W14x398 flanges; spacing btwn. flgs. and their parallelism or the radius at the W14 flg./web junction, etc. Usually you want the two pieces side by side when you are doing this kind of fitting and grinding, etc. and this effort is very expensive fabrication if it is not needed. Then the erector will invariably not have the right shoehorn to put that WT into place in the field, into and btwn. the columns. I would terminate the WT several feet short of the columns as long as the existing clip angles and W21 still work in shear, for the new reaction. One thing you might consider is a small plate, standing perpendicular to the axis of the WT and welded to the top of and the end of the of the WT web, which is then field welded to the bottom of the W21 flg. This end pl. might be 6" wide by 8" high, welded to the end of the WT web. This is part of the termination weld for the very stiff WT to keep it from unzipping from the W21 flg. I don’t like terminating such a stiff member as the WT with two stopped fillet weld terminations. Those terminations tend to be stress raisers however well made, and these are overhead. This would not be an issue if the doubler/reinforcement were just a bot. cover pl., not so stiff. You could jack on the underside of the W21 top flg. if you wanted to take some of the DL off of it during the application of the WT, and the jacking stands may even act to support the WT during welding.

 

[JLNJ]

The numbers I quoted are the for the elastic section right out of Shapebuilder. If memory serves, it was 127 for the W21 and 258 for the combined section.