Strengthening existing concrete beam

[ajk1]

Is epoxy bonding a steel plate to the soffit of existing steel beams a proper way to strengthen them or would the epoxy fail when the steel plate starts to yield? I suspect that is what would happen, so that would not be an acceptable way to strengthen the beams. Looking for doubling the strength, so carbon fibre seems not to be an option.

 

[Zambo]

I typo I assume, the title is concrete beams but the text "existing steel beams"

 

[jayrod12]

No I'm betting he means concrete beams based on his previous posts. His bread and butter seems to be in concrete restoration.

I can't see a scenario where you'll get double the capacity, unless you shorten the span. Perhaps that's an option.

I also dont believe epoxying a steel plate has industry accepted standards the same way FRP does.

 

[ajk1]

Yes was a typo. I meant existing concrete beams.
I tried many years ago, epoxying steel plates to a concrete slab soffit, but it did not work. The epoxy did not bond sufficiently, despite surface preparation (sand blasting as far as I can remember), of the steel and the concrete surfaces. The plates could be relatively easily removed, so I am not inclined to do it again. Thanks for the opinion that it does not have industry acceptance the way FRP does.

 

[JoelTXCive]

I took an FRP course in graduate school. I don't think you can get double capacity out of any method.

Regardless of what method you attempt, don't forget you need to unload the beam while installing the repair. That way the new FRP (or Steel Plates) can become engaged and take load.

 

[EngineeringEric]

You can bolt a steel plate to concrete. This is just a shear flow exercise.

Fire loads vs service loads are both considerations for FRP as well, although i've only done a small handful myself, maybe not even full....

 

[BridgeSmith]

External post-tensioning with strands or bars would be the only way I know to significantly increase the capacity. In order to double it though, the beam would have to have very light tension reinforcement to start with.

 

[E720]

Engineering Eric - If you bolt a steel plate to the concrete do you have to worry about drilling into existing reinforcement? How would you handle any damaged rebar? It's an interesting idea which I have never considered.

 

[Agent666]

Bolting alone is not that good because the slip in the holes (standard bolt diameter + 2mm) invalidates the strain compatibility when they inevitably slip to take up the tolerance in the bolt holes. Obviously any bolting needs to be designed/located to avoid compromising the existing reinforcement, scanning the beam is relatively cheap theses days and a good investment to avoid issues. Regarding how would you handle damaged reinforcing, avoid creating situations where reinforcement can be damaged as you are just creating a headache for yourself (i.e. practically very hard to repair).

As you initially noted bonding the steel plate or using FRP is the better option. You can bolt the plate to hold it in place, but structurally the epoxy is doing the work.

As others have noted getting double the capacity is usually not possible, the actual capacity you get is highly dependent on the initial strain in the un-strengthened member as you are usually governed by the original reinforcement after strengthening being the limiting factor (or the bond strength of the epoxy). You would never be able to take the full capacity of the plate provided for this reason. You usually cannot reinforce the beam so much that you get a brittle compression failure governing as well (i.e. exceed balanced strain conditions).

I've done jobs with FRP where we cranked out load on a beam using flat jacks and a reaction frame between columns under the beam before installation of FRP, and we still only managed 30-40% maximum extra capacity.

Other options might include
- Installing steel or concrete beams under the existing beam to share the loads if headroom exists.
- Installing steel channels on either side of the beam with bolts sufficient to take the full load, though close analysis of the deflections after the concrete beam capacity has been exceeded would need to be investigated. Uaually only works for smaller spans.
- Consider other ways of spreading the new loads, I once slung beams above the slab between adjacent columns with hangers through the slab to support new steel beams orthogonal to a new infill slab to support a higher loaded plant area (beam was ultimately hidden in a wall above the slab).

 

[ajk1]

I should have said that these were post-tensioned existing beams that had the old "push-thru" type of sheath system which is very subject to water ingress and corrosion failure, and we are trying to allow for the possibility that all the strands fail due to corrosion. That is why I said doubling the capacity, as a simplified way of describing the reason and objective.

To design for movement at the bolts if the plate is bolted on, I normally specify a square plate washer for each bolt, with a hole in the washer that fits tight to the bolt, then weld the plate washer to the plate reinforcing. But that is a lot of expensive field welding, so I don't think that a bolted plate is very economical.

Steel beams each side of the existing concrete, just under the slab, might work. I would preload the steel beams with flat jacks, to some percentage of the load that the steel beam is to carry, to limit deflection of the steel beam if all the existing tendons fail, but not so much preload that the combined effect of preload and all the unfailed tendons creates too much uplift on the beam; if it does, then some of the existing tendons would need to be detensioned, which I don't want to do. If it helps, then surface prep to the top of the new steel beam and the slab soffit and then grout the 2" space between the slab and beam, might achieve some composite action, but am not sure that the cost of this composite action is worth it.

Preloading is mainly to address deflection, not ultimate capacity. Generally the ultimate capacity is not dependent on the load history of the assembly, since at ultimate the materials yield, although when epoxy bonded that may not be true if the epoxy can't yield sufficiently. But anyway, I am not going to epoxy a steel plate because of lack of success many years ago, and the comments from others on this post string of comments.