Double reinforced slab

[UnusedHandle]

I have a 10" thick concrete slab on grade that is reinforced with #5 bars 12" c/c, both directions, top & bottom. The slab supports machinery that imposes moments in both directions and in sign (at any given time it may be a positive or negative moment about different axes). For a given load scenario, is it acceptable to use one layer of steel as the tension reinforcement, and the opposite as compression reinforcement?

If yes, what would I use for the strength of the slab? I haven't touched concrete design since college and there I used Mn = As1*fy*(d-a/2) + A's*f's*(d-d'), which I cannot find in the code, and it may only apply to beams.

Thank you in advance for your help.

 

[fegenbush]

If you do the calculations, you will find that the compression reinforcement contributes almost nothing to the moment capacity. In my experience, it's not worth the work to calculate it out when it is more conservative to decrease the bar spacing (or increase bar diameter) on the tension side.

 

[UnusedHandle]

Thank you, the slab is already existing and we are doing an analysis to determine if it is adequate for the new machinery. Before I got too deep into the calcs I wanted to find out if this would apply.

 

[BAretired]

Fegenbush is correct; neglect compression steel in your calculations.

BA

 

[SlideRuleEra]

I, too, agree with fegenbush. Assuming that a 10" slab (on grade) with two mats of #5 bars has proper rebar cover (3" on the bottom and say 2" on the top) the separation between the rebar mats is of minimal structural importance - IHMO. Therefore, the size of the bars is not important. For practical purposes, calculations will be a paperwork exercise that will give "precise" results which are not "accurate".

For industrial slabs where loads were not predictable, we considered a 12" thick slab with two mats of #4 bars to be the default design. There was no need to use larger bars since even in a 12" slab the separation between the mats was still relatively small.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[IDS]

I disagree that the "compression" steel does not make any difference because (as implied by SlideRuleEra) with the sort of cover normal for an on-grade slab it will be working in tension. With 3 inch and 2 inch cover the top steel increases the positive ultimate moment capacity by over 20%, and using a computer to find the capacity the extra time taken by including both steel layers is precisely zero seconds.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[KootK]

@UnusedHandle:

1) Yes, so long as you've got appropriate laps on both mats of steel, it's valid to use both mats in your flexural resistance calculations.

2) If you're going to use both mats of steel in your flexural calculations, you will need to employ a calculation method that takes proper account of strain compatibility. Depending on the specific geometry of the situation, the extra effort required to include the "compression steel" may not be justified.

3) Your equations are the right ones. The trick is the value of f's. Depending on the locations of the bars, f's may be so small that it adds a trivial amount of compression steel flexural capacity. Or, as IDS has suggested, f's may even be negative, indicating that your compression steel is really just more tension steel.

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

 

[BAretired]

The effective depth of reinforcement in slabs on grade is quite variable because chairs are not seated on a hard surface; bars often get displaced during the concrete pour; to rely on the bottom layer as contributing to the tensile steel is wishful thinking; to achieve significant tensile stress in bars close to the the bottom face requires large curvatures; ignoring the bottom steel in a slab with tension on top is a sensible and prudent measure to take.

BA

 

[IDS]

BA - With similar reasoning, by how much do you reduce the area of a single layer of steel?

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[BAretired]

I never reduce the area of steel. If the bars are chaired on a firm surface such as a steel or plywood form, I do not reduce the effective depth either; if they are chaired on sand or gravel, I make allowances for the possibility of the steel moving during the pour.

There have been cases where chairs used in suspended structural slabs have been insufficiently secured to the form to prevent the steel from moving during the rough treatment they receive during a pour. It is often difficult to take appropriate action while concrete is being placed and in some cases, I have called for remedial measures after the pour because of questionable steel placement during concreting.

To assume the bottom layer of steel in a ten inch slab can be relied upon to carry tensile stress is in my opinion, not a prudent judgment call, particularly when the slab is existing and cannot be inspected.

BA

 

[UnusedHandle]

BAretired, by this line of reasoning, we can't know anything about the slab at all. I worked construction before I got into engineering and I've seen slabs that were poured 2 inches less than called out. Somebody didn't maintain quality and a plain floor ended up being 6 inches (intended) at one end of the building and 4 inches at the other. Without being present for the pour, too many things can be different from what we "know". If we can't assume that the bottom layer is capable of carrying tension, because we don't know where it's at, how can we assume that the top carries anything since we don't know where it's at? What allowances can be made for an unknown depth? Should we just assume a plain slab?

 

[dcarr82775]

You have to design for what you specify. It is prudent to take into account some things that you know will happen in construction. For example, I never specify WWF in a slab because I have never seen it done right, ever. But at some point you have to leave it to the contractor to build what you specify.

To the OP's topic, even if the steel closer to the compression face is technically in the tension zone at ultimate conditions, I wouldn't count on it as contributing to the strength. Strain compatibility and other such things being what they are.

 

[BAretired]

UnusedHandle,
In your initial question, you asked: For a given load scenario, is it acceptable to use one layer of steel as the tension reinforcement, and the opposite as compression reinforcement?

The answer is no; the reason is that the layer of steel on the compression side of the slab is not going to be in compression at ultimate load; alternatively, if it is in compression, it is misplaced.

IDS suggested that with two or three inch cover, the so-called "compression" steel will be in tension and could be used to increase the moment capacity of the slab. I concur with that but I would not rely on it because a) large and unacceptable curvature would be required to develop that steel to yield stress and b) minor deviations in placement make large differences in its effective depth because of its proximity to the compression face.

While it is true that one does not know with certainty whether an existing slab was built precisely in accordance with the drawings, it is reasonable to assume for the purpose of checking its strength, that it conforms reasonably well.

BA

 

[Brad805]

I am curious how you are determining the moments and what you might know about the subgrade?

I agree with BA. Without some NDT tests, cores or some reasonable records from the original pour I would err on the side of caution. Look at the ACI tolerances for guidance on thickness/placement max/min scenarios. The decision to ask for testing depends on the value and importance of the machine that you are renovating for.

 

[IDS]

Designing on the basis of a single layer (i.e. the one closest to the tension face) will always be conservative or make no difference (if the other layer is exactly on the NA), but I would like to make a couple of comments about the statement below:

but I would not rely on it because a) large and unacceptable curvature would be required to develop that steel to yield stress and b) minor deviations in placement make large differences in its effective depth because of its proximity to the compression face.

a) The ultimate curvature is controlled by the specified ultimate concrete compressive strain and the depth of the NA, and it is slightly reduced by including the second steel layer in the calculation. If the steel is close to the NA then it won't be at yield in the calculation. The serviceability check (steel stress or crack width, or whatever the applicable code requires) will have both layers of steel well below yield.

b) If misplaced steel is a concern then I'd check the section capacity assuming some reduced cover from the compression face, with corresponding increased cover to the tension face, but I would still include both layers. It may well not make much difference, but it doesn't take any more time either.

c) With a very large cover to the tension face you are going to get large cracks if the section cracks at all. If this is not going to be acceptable then check the cover with a cover meter.


Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/