Existing beam strengthening with top and bottom cover plates

[bvass]

Hello All,

I wanted to get some opinions on a current design issue I'm having with adding top and bottom cover plates to an existing W16x31 shape. The building was constructed in the late 70's, and we've found that most all of the structural steel is 36 ksi. We are adding a significant amount of floor load to the W16, and so it needs to be strengthened in order to resist the new loading. I would like to reinforce the beam with top and bottom cover plates to get me to my required flexural strength. I've already built a mathcad that uses AISC 360-05 Sec F-5 to develop my new flexural capacity with the addition of cover plates (I used F5 instead of F4 because it's less intensive).

I'd like to use 50 ksi plates (in design specs) for reinforcement. However, I'm not sure as to how the two different grades of steel interact? I know how to design a built-up section with different materials using a transformed shape, but since E is the same with 36 ksi and 50 ksi steel I'm not sure how to go forward with design. Is there some sort of hybrid yield stress I can develop based on each steel area? If anyone know of any references I would greatly appreciate it.

I've tried using 36 ksi steel as well for the plates, they just get very thick in a hurry. I also know I'll get more capacity using section F4 of the SCM, however I'd like to shy away from that because of the amount of time it will take to develop and the project deadlines. Thanks for you help!

Brad

 

[KootK]

Do you have access to the top of the beams to cover plate the top flange?

I don't think that you need to sweat this too much. Since E is the same for both materials, your assumptions about relative stiffness and where the load goes should remain valid.

The 36 ksi material will yield at a lower strain than the 50 ksi material but that's okay. The 36 ksi should be able to deform plastically until the 50 ksi kicks in.

If you're doing any buckling checks, you'll probably need to give some consideration to Fy and yield history there. Some simplifying, conservative assumption will probably be the way to go.



The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[LonnieP]

If you have the depth available it might be better and easier to simply weld another beam to the original one. Don't forget to check the end connection for the larger reaction.

LonnieP

 

[jvvse]

An existing beam you usually don't have access to the top plate. Maybe consider a thicker bottom plate only or even welding a WT to the bottom of the existing beam if you have room.

As KootK said, a simplifying, conservative assumption like assuming your plates are 36 ksi was good. Your plates may get thick but the alternative is to spend more time analyzing the specific yield points and you'd come up with a 1-1/2" plate instead of a 2" plate which is hardly worth a few hours of your time.

 

[bvass]

Thanks for the responses.

I don't have enough headroom below to weld another beam to the existing, good suggestion though.

I'll assume my plates are 36 ksi and call it good, you are right it's probably not worth the trouble to sharpen my pencil in this case. I do have access to the top of the beam in this particular case, so welding a plate to the top is feasible.

Thanks again guys.

 

[SAIL3]

and I doubt if 50ksi pl would be readily available....

 

[paddingtongreen]

This would not be my first shot at a solution, if the top cover plate is wider than the beam you have a lot of overhand welding, something only office bound engineers, those who haven't been cursed by a welder for using them can love.

I would look at getting additional 16" beams in between the existing beams, it takes a lot less field welding to get tab plates welded to the girders. I presume you have sufficient access since you can get big cover plates in there. The advantage is: less field work, more shop fabrication.

Another possibility is to add W14s between the W16s, near the ends and place W12s from the W14s, between and parallel to the W16s. Again, mostly shop work. It works if the beams can handle the added shear and moment near the end.

Michael.
"Science adjusts its views based on what's observed. Faith is the denial of observation so that belief can be preserved." ~ Tim Minchin

 

[racookpe1978]

Mentally, I'm picturing somebody trying to weld the flanges of 2x W8's or W10's on each side of the web of a W16 up in the overhead.....

 

[JStephen]

Some ideas:
Possibly use a heavy channel with flanges turned in instead of a plate or bar.
Consider stitch-welding instead of continuous welding the member to the beam; the channel should allow greater distance between welds on the compression flange.
If you do a bunch of welding down the length of a straight beam, it won't be straight when you get done. In some applications, that won't matter, but be forewarned.
See attached sketch for couple of ideas of how to weld the top piece from the top side only.
Wonder if you could put holes in the bar/plate and plug weld it to the beam from the top side?

 

[JedClampett]

Welding plates to reinforce beams is inefficient. It just takes a lot of plate to make much headway. Plus all the dead load stresses are locked in the beam, unless you shore it up.
How much overstress do you have? Most 36 ksi steel made in the 1970's had a yield stress approaching 50 ksi. If the beam is adequately braced, you could use that. Or if it's not braced, maybe additional bracing can get you there.
Another approach that seems like cheating, but it's not, is to the 14th Edition of AISC. If the beam is adequately laterally braced (yeah, I know, I'm a broken record), you can use the Plastic Section Modulus, which buys you about 15% no matter what your yield stress is.

 

[hokie66]

If you can get at both flanges to weld....could you also just take it out and replace it with a heavier beam? Perhaps that would be more economical.

 

[bvass]

Thanks for all the responses guys.

I've sized the top plate to be less width than the top flange, so that they can simply lay it down and stitch weld it along its length. The bottom plate is sized to be wider than the bottom flange, I figure they can clamp it to the bottom flange and again stitch weld from the top down.

Lateral bracing isn't continuous, so I've added a few more members and some horizontal bracing to kick back to the building main LFRS. Even with the new bracing and my capacity based on yielding I'm still overstressed.

Removing and replacing isn't really an option, on one end of its span it's supporting a floor currently. I most likely could add some framing perpendicular at the ends and span a new beam parallel to the existing to pick up the new floor loads, I'll look into that.

I've actually come up with some reasonable plate sizes using 36 ksi steel, so that may be a good option too.

Appreciate the help.

 

[LECT12]

Does the layout/beam sizes of the framing work out in such a way that you can just make moment connections between adjacent beams and form a continuous beam?

Are you pouring a new slab-on-deck? You can gain a substantial amount of strength just from minimum composite action...