Reinforcing Steel Beam Webs

[atrizzy]

I have a situation where I have a W460 steel transfer beam that has undergone significant corrosion, to the point where the web is too slender to function (thickness testing is currently ongoing). Flange flexural capacity is still OK. Replacing this beam is almost impossible as it supports structure above, as well as beams framing into it. I've had the following thoughts on fixing this:

1. Add web stiffeners at the appropriate spacing to mimic a compact web. The issue here is that the web may well be too thin to weld to.

2. (more out of the box) Ignore the web entirely, treat the flanges as truss chords, and provide diagonal and vertical stiffeners (angles even) capable of resisting the forces in a truss analysis. This will require a bit of fancy detailing around the beams framing into... but I think it's doable.

Any thoughts about my ideas above, or other thoughts?

 

[BridgeSmith]

Add transverse stiffeners at the spacing required to meet the required shear capacity. If the web is too thin to weld, you can do angles on both sides and bolt them together through the web.

If substantial portions of the web are too thin to provide the shear capacity, even with the stiffeners, you may have to do option #2. For ease of construction, I'd recommend bolting back-to-back angle pairs through the web. Fillet welding stiffeners to the flanges will reduce the fatigue stress limit for the beam. If the cyclic load stresses are high, you may have to avoid connecting to the flange, use a bolted connection to the flange, or use full-penetration groove weld.

 

[JAE]

You could add "new" web plates along the length where the existing web is deteriorated...perhaps extending the new plates 2 to 3 times the depth of the beam beyond the point of deterioration ending.

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

 

[atrizzy]

HotRod, I don't think I can bolt through the web if it's quite so thin. At least I don't want to assume that I can. I'd likely need to weld the vertical stiffeners to the flanges. I like your idea about bolting the diagonals to the flanges.

JAE, the problem with your suggestion is that I have a few beams framing into the one in question.

 

[ukbridge]

I would have suggested JAE's idea as well but actually :

1. I'm not sure that'll work either as ensuring a true "fit" between the flanges will be extremely difficult.
2. In-service this will be particularly difficult to do if the beam has seen any significant deflection.

Interesting problem gl.

 

[BridgeSmith]

I was suggesting to put an angle on each side of the web and bolt the angles to each other. This will stiffen the web so that it doesn't buckle. If the web is too thin and could tear in tension due to the shear, then you will have to reinforce it another way, such as diagonals or supplementary plates.

 

[BridgeSmith]

If you opt for the truss option, welding gusset plates longitudinally to the flanges to attach diagonals could work. Welds parallel to the direction of the applied stress carry a higher fatigue stress limit (Category B - same as drilled holes) than transverse welds (Category C or D).

 

[KootK]

1) I'm not so hot on the trussing idea. 18" isn't all that deep. You'll have a zillion webs and I'd think that things would be congested and awkward and the joints.

2) How much shallower are the infill beams relative to the transfer beam? I'm assuming there's some space there if you're able to truss across the infill beams without invoking vierendeel panels etc.

3) Is there any available space to work with below the transfer beam?

4) Any chance the infill beams are offset on the two sides of the transfer beam such that the web will always be continuously accessible on at least one side?

5) Is it at all feasible to jack the transfer beam to an unloaded position prior to doing the reinforcing?

6) Will you reinforce all the way to the ends of the beam or terminate the reinforcing partial span? If it's partial span and you do plating/trussing outboard of the existing web, keep in mind that you'll need to move the shear in the reinforcing back to the beam web laterally where the reinforcing terminates.

7) Is the transfer beam web not corroded at the infill beam connections? If work would need to be done at those locations regardless, that might change the calculus regarding the optimal strategy here.

 

[atrizzy]

Hi Kootk,

1) Agreed, after drawing it out it seems a little congested.
2) The infill beams are W310s.
3) There's a little bit of space, but the client wants to keep headroom if possible. Even if I were to put a beam underneath, I would still have that weak web relied upon in the solution, which I want to avoid. Though this is looking more and more like something to propose.
4) Nada
5) Yes, this is a possibility.
6) I'm waiting for the NDT data currently, but it appears that I'll be able to terminate prior to the end connections, which is great.
7) It appears that it is, though not much more than the remainder of the beam. My plan, if the bolted connections are insufficient due to inadequate web thickness, is to attach stiffeners to the beam connection brackets to spread the load to more of the web.

Currently I'm examining the bolted web stiffener option.

 

[atrizzy]

To anyone reading this in the future, the NDT suggested an low-end web thickness that I could work with. I ended up adding bolted transverse stiffeners as per HotRod10's suggestion. Thanks for the responses everyone.

 

[SKJ25POL]

atrizzy (Structural)(OP),
1- Could you please post couple of pictures/photos of the beam.
(Please at least one pic showing the beam full length and couple that pointing the problem areas?)

2- Also, I didn't catch it, was is this beam's function - is a building beam? Is as bridge beam?

3- What part of country you are located? High Seismic? High Snow? High wind or Hurricane? Is beam seeing any vibration?

Skj
Please even you have fixed it let me know on above questions.
Thanks