Composite or not? 1

[timothyd]

I have a condition where a contractor would like to encase a steel beam in concrete. The problem is that the steel beam is in a loaded, and overstressed condition. He proposes to reinforce the concrete such that it can carry the entire load and essentially abandon the steel in place. My concern is that he does not want to relieve the load. Therefore, I am having difficulty rationalizing that the existing beam can be abandoned in place in this condition. I first thought that adding shear connectors would be a good idea, so I am I counting on the studs to transfer the load - or am I counting on the beam to act compositely?

 

[msquared48]

In order to pour this he is going to have to shore the beam anyway, so the load to the beam can be relieved to at least some degree.

Steel studs would be a good idea to achieve at least partial composite action. However, would there be a concrete flange too to develop?

I think I would still add extra rebar to take the whole load, and shear/confining seteel to keep the concrete on the beam.

Mike McCann
MMC Engineering

 

[timothyd]

No concrete flange - just a rectangular section. Thanks for the input.

 

[rowingengineer]

Is your situation deflection sensitive?

ANY FOOL CAN DESIGN A STRUCTURE. IT TAKES AN ENGINEER TO DESIGN A CONNECTION.”

 

[timothyd]

For this case, no. Thanks.

 

[RHTPE]

timothyd / Mike

In order to pour this he is going to have to shore the beam anyway, so the load to the beam can be relieved to at least some degree.

Having done many forming/shoring projects with concrete encased steel, I would say it could be done in one of 2 ways:
1. Build a deck to support the concrete and the side forms for the concrete beam. This would not directly contact the existing steel beam and would not help to unload it in any way.
2. Suspend the entire formwork from the existing beam. This would be totally contrary to the end goal, as it will add additional load to the existing steel beam.

Assuming this is a simple-span condition, one could at least consider the concrete thickness on each side of the top flange as the concrete beam's compression flange. Essentially construct a concrete beam on each side, then tie them together with 2-piece "U" stirrups. Each beam could utilize ladder stirrups (common in wide-module one-way ribbed slabs).

None of this serves to reduce the current over-stressed condition however. To accomplish that, one has to somehow shore the beam to unload it to some degree, then construct the concrete beams. Perhaps 2-3 "holes" penetrating the bottom concrete encasement to permit the use of an appropriate shore to unload the steel beam. The shores would have to contact the steel beam's bottom flange. Depending on the required upward force to unload the steel beam, these "holes" could be substantial in size.

Using shear studs welded to the underside of the top flange/web fillet (at a 45 degree angle) will certainly help achieve composite action, would better engage the concrete side "beams", and would negate the need for substantial cover over the top of the steel beam's top flange.

Concrete placement could be difficult depending on the side cover and achieving good consolidation below the steel beam's bottom flange and below the top flange will be difficult. Perhaps a good application for SCC.

Just my humble thoughts on the situation.

Ralph
Structures Consulting
Northeast USA

 

[timothyd]

Thank you very much for your input. Seems to me that trying to relieve some or all of the load brings the system to some standard, rather than letting the original steel remain in an overstressed condition.

 

[csd72]

Have you actually done the numbers to check if this could work? I have tried similar things before and found that the additional concrete was of little benefit.

 

[timothyd]

It's a review of another engineer's design. His design proposes to ignore the state that the existing steel is in now and simply construct a new concrete beam to carry the entire load, dead+live, without relieving any load and abandoning the existing steel in its overstressed condition. There is no calculation or consideration for composite action between the two materials. So the concrete will work fine on its own - but I still have some difficulty rationalizing that the load gets transferred and difficulty concurring that the steel beam can remain in that state. One analogy that I can think of is if permanent shoring was to be installed under a beam that is failing, and then left like that. Looking at it like that - one could say that the concrete beam is a type of permanent shoring. But my instinct is telling me to have him relieve as much load as possible first.

 

[Teguci]

Ok, possible failure mechanisms:
- Bending moment - Concrete will be designed for full capacity but never see it due to the steel beam currently carrying most of the moment. If the steel beam does get overloaded, it will form a hinge (probably at the center span). Any relaxation or hinging of the steel will transfer load to the concrete under load sharing. Don't see a problem here.
- Shear - Concrete will be designed for full shear.
- Servicability/cracking - Concrete beam will be detailed to take the full load. Servicability will be at least as good as the concrete alone.
- Magical unloading of steel beam - Beam curvature will try to "unsmile" putting the top flange in tension and the bottom flange in compression. Of course this would be due to more load being placed on the concrete beam which would couteract the loads from the steel beam.

I'd be tempted to drill some holes in the steel beam web to feed some reinforcement through, but would be OK with it under the loaded condition.

 

[timothyd]

Thank you very much for your input. Good point regarding the tendency for the steel to straighten out.

 

[csd72]

There is a strain compatability issue here.

The steel beam is basically taking all the load at the moment so it is very unlikely to deflect any further. The concrete beam will be installed at this state but will have a tendency to crreep over time.

The result is that the long term stiffness of the steel beam will be much higher than the concrete ensuring that it takes almost all the load up to the point that it yields.Not only that but you have also added aditional weight.

Not a problem if it is just a plastic yield issue but what if the beam fails by buckling? Will this lead to spalling of the concrete?

This actually highlights a new question, how is the beam failing - i.e. by buckling or yield strength. If is is by buckling the the concrete will help provide restraint though it needs to be checked against spalling.

 

[hokie66]

Late to this, but I think Teguci has summarized it well. If the steel beam yields, so what? The reinforced concrete beam will then start to carry some of the load.

 

[csd72]

If it yield yes, if it buckles.....

Not saying that it will necesarily be a problem but it needs to be considered in the design.

 

[hokie66]

As described, I thought buckling was far-fetched, as the steel is to be encased in a concrete beam.

 

[csd72]

hokie,

in how much concrete?, will it have adequate reinforcement to hold it together around the steel if it starts to crack. Will it be adequately compacted to the underside of the top flange? There are a lot of assumptions being made about these things.

If all these items are adequately addressed then I would think it would be okay but not knowing the exact situation or type of loading...

 

[hokie66]

OK, but I can't conceive of a situation where a concrete beam of greater capacity than a steel beam, completely encasing the steel, would not prevent buckling of the steel.

 

[Teguci]

csd72 has a point. If the beam suddenly unloads due to buckling it could kick laterally (this of course will be limited by the deflection). In concrete columns we prevent this by tying hoops around the rebar. For this case I think a similar hoop encasement is justified. If the hoops can't go over the top, they should be fed through holes drilled through near the top of the web.

 

[steellion]

Allow me to pull out Occam's Razor and ask why can't the steel beam be reinforced with steel plates? This concrete solution may or may not work but seems messy compared to just adding more steel to increase the beam capacity.

 

[timothyd]

It certainly can be reinforced with steel plate. In fact, that was my recommended repair design to the owner. The concrete retrofit was recommended by another engineer as an alternate and more cost effective approach. I, in turn, was asked to comment on it. My initial reaction was not positive. But, not wanting to criticize it without due regard, I started to research the concept of such a repair. Thus, here we are. Also, the steel beam will be exposed to a corrosive environment. The original steel is galvanized. The contractor was looking to avoid welding and the owner liked the idea of encasement. So, a lot of related factors leading to the question. The basic question of "does this fix work?" was asked. Ultimately, the answer is yes. But I doubt in the end it will be cost effective.