Cracks in Existing Concrete slab with PT beams

[AK4S]

I was looking at the roof of an existing Building which has Post Tensioned Concrete Beams with 5" thick concrete slab spanning between the beams. I see cracks at the bottom of the slab. The attached sketch shows the crack pattern observed. I am trying to find the cause of the crack before I decide if injecting it is the best repair.
I don't think these are flexural cracks (I ran the slab design and it is stressed to only 70% of its capacity at max. loads). Also the cracks seem to originate at the free edge of the slab and propagate inward.

The slab has only mild steel reinforcement (no post tensioning) and the building was constructed in 1960's.

I don't have much experience working on PT structures and was wondering if these cracks are due to any issues associated with the original Post Tension on the beams. If the sequence of Post Tensioning of the beams or in general the compression at the beams could have led to tension at the free edge of the slab, leading to these cracks.

Any insight is appreciated.

 

[KootK]

I think that I see what you're getting at and I agree, there is a mechanism to support that narrative.

 

[Retrograde]

If the slab is not post-tensioned (only the beams) then in my opinion a 5 inch slab is too thin to span 17 feet, especially given the short backspans. I would expect excessive long-term deflection, which would lead to the observed cracking.

 

[AK4S]

@Retrograde: I agree the slab is thin. But this is the roof slab and the loads are about 80psf, so numbers look ok for flexure. (per existing dwgs, the 5" Slab is reinforced with #4@8" O.C. @ T&B, 1" cl. cover)
@KootK: Yes, your markup is what I was thinking as a potential cause. If this is the case, then these cracks should have been formed during the initial PT and at this stage are no longer moving. So I could inject them.

 

[Ingenuity]

I see cracks at the bottom of the slab.

If cracks are NOT full-depth I would expect the slab cracking to be flexural, however, I would also expect flexural cracking to the top of slab near (parallel with) the PT beams, AND flexural cracking extending further along in longitudinal direction, not just at the edge. As Retrograde stated, with 5" thick slab over 17' span, that is L/D=40.

As this is a roof, were you able to do a good visual review of the top surface? No membrane, coating etc obstructing?

 

[AK4S]

@Ingenuity: The roof is covered in membrane, so I have no idea about condition at the top of the slab.
With the crack pattern observed (wider at the slab edge and dies down away from the edge, multiple cracks - not only at mid span), I am having doubts if these are flexural.

 

[hokie66]

I think what you are observing would be restrained shrinkage cracking. The beams and slabs on either side act as horizontal deep beams, restraining the thin slab across the 17' dimension. The cracks would be full depth.

 

[r13]

Did you check the deflection at the free edge? You need good instrument as the magnitude can be quite small. Take measurements on a few spans may provide clue. I kind of agree with Retrograde that it could be long-term effect (aging), if cracks been there for a long time (since PT at 1960s), how come nobody noticed until now.

 

[rapt]

Yes, the splitting/spalling condition that Kootk has described is true but the stresses should be fairly low and the #4 at 8" T & B should keep any cracks very small assuming the P/A in the beams is not enormous.

And it would be full depth as Ingenuity said so from the outside edge you should see close to constant width cracks full depth.

Then flexure would close the cracks at the top and open them further at the bottom. Plus if the top surface is a lot hotter than the bottom, there would be some tension in the bottom from temperature differential which would add extra tension. It could be a combined effect, worst at the edge.

Normally we would have a cage of bars around the edge so that there would be at least2bars within about 6" from the edge to give slightly better crack control around the edge.

 

[KootK]

Is this a 20' cantilevered roof then? If so, it would not surprise me to find that the beams are quite heavily prestressed.

 

[KootK]

Also, if that's a 20' cantilever, the effective slab span might start to feel more like 24' than 17' as you approach the cantilever tips.

 

[hokie66]

No comments about my theory? I think the slab would have cracked with or without post-tensioning of the beams.

 

[KootK]

I had trouble visualizing your proposal hokie. Sure, the side panels will be laterally stiff but what is it that restrains them laterally? The columns?

 

[hokie66]

No, I theorize that the side panels both act as giant channels which resist the force from the slab as it tries to shorten. These side panels both cantilever laterally off the uncracked slab beyond the edge.

 

[KootK]

Doesn't that make it a differential drying thing with the perimeter drying more slowly than the interior? Wouldn't we expect the reverse?

 

[hokie66]

Not sure about that. The perimeter is probably more exposed, so it would depend somewhat on the atmospheric conditions.

Now, as to differential drying shrinkage, I would expect to see some cracking in the slab perpendicular to the beams, i.e. perpendicular to the cracks which the OP has observed.

 

[Tomfh]

Now, as to differential drying shrinkage, I would expect to see some cracking in the slab perpendicular to the beams, i.e. perpendicular to the cracks which the OP has observed.

It's stressed in that direction.

 

[Ingenuity]

OP - have you been able to observe the crack width/s?

Crack repair using low-viscosity epoxy via plural pressure injection is common, and can repair very fine cracks, however, often it is not necessary if all you are trying to do is seal the cracks.

With this 50+ year old structure, assuming it is still in good condition, maybe you don't need to do anything.

On the flip side, the cracks may be due to corrosion of the rebar in the long direction, precipitating from the slab edge - a case where crack injection without concrete repair is not recommended. However, cracking due to corrosion is usually associated with concrete spalling. Have you seen any spalling?

 

[hokie66]

Tomfh,

The beams are stressed, not the slab.

 

[AK4S]

@ALL: Thank you all for your responses. Sorry I could not get back to responding earlier.

@hokie66: I did not observe cracking in the slab perpendicular to the beams, which would have indicated a potential differential drying shrinkage condition.
@rapt: Not sure about the range of P/A used, the beams do have a 20ft cantilever with 44ft immediate back span, So as KootK said the beams could have been heavily prestressed.
and as you said,flexure could have reduced the crack width at the top and open them further at the bottom. The drawings did not specifically show additional rebar at the slab edge.
@retired13: I did not have access to measure deflection at the free edge. The cracks may have been there for a while and looks like they extend to the full depth of the slab. Probably the roof membrane above started leaking few years back and moisture got in to the cracks. Some of the cracks show efflorescence and rebar corrosion.

@Ingenuity: I was not able to get to the cracks, observed them from the ground 30ft below. Did not see any spalls yet,will however recommend to sound the concrete in the vicinity of these cracks to confirm. The cracks show efflorescence and rebar corrosion due to moisture ingress. A portion of this cantilevered slab projects outside the building envelope and so it is exposed to the elements. So will probably sound the concrete and look at injecting the cracks if no other concrete repairs are needed.