Concrete Wall Shrinkage Cracks

[mnovicki]

My boss and I have been going back and forth on where a concrete shrinkage crack propagates from. Our concrete wall is 8" thick and 100 ft long and 4 feet high. It is poured on a footing that is assumed to have no further shrinkage... i.e. fixed. The wall is heavily reinforced, but we still have some cracks, which we know are shrinkage cracks. Some start from the bottom, some start from the top.

My theory is that if the wall is restrained on the bottom and the top is free, the crack should propagate from the top and work its way down. My boss' theory is that the crack propagates from the bottom, since that is where the highest stress from resisting shrinkage would occur.

I see both sides but am unable to come up with a definitive answer, at least this late in the day. Anyone ever thought about this question? I think that starting from the top, each succesive layer as you go down, has more restraint against movement and is less likely to crack. On the other hand, if the concrete is trying to shrink, and it is resisted by the base, the stress will increase and a crack would eventually have to occur at the bottom..... Any ideas?

 

[MiketheEngineer]

Both seem reasonable and probably why we see cracks from both top and bottom.

 

[seimike]

The concrete is going to shrink. When the stress in the concrete due to the shrinkage exceeds the modulus of rupture which is low during initial curing, then a crack will occur. Do to the restraint at the bottom of the wall the resistance to movement will induce cracks. This does not mean the top of the wall is without shrinkage stress buildup, but the point at which the top of the wall exceeds the modulus of rupure would not necessarily coincide with the induced cracks at the base. I typically put expansion joints in such walls at between 25 to 30 times the thickness of the wall - including discontinuing all horizontal rebar, smooth dowels at joint and specify chamfer strips to create a weakened plane in the wall.

 

[hokie66]

They generally start at the bottom, for the reasons given by your boss. And they normally occur at about 3 to 4 metre intervals. Added reinforcement near the bottom of the wall will tend to hold the cracks together and make them less noticeable.

 

[sdz]

At the bottom the cracks should be close together and narrow due to the higher restraint. At the top they should be further apart but wider; expect spacing of 1-2 times the height of the wall if lightly reinforced. If heavily reinforced horizontally the steel will have sufficient strength to stop the cracks from opening after the initial crack forms and you can expect more but narrower cracks.

 

[TXStructural]

Agree with sdz.

Shrinkage cracks are caused by restraint. If the wall were free to move, had an appropriate w/c ratio and mix design, and was cured properly, there would be little reason for cracks.

The bottom restraint and shrinkage result in cracking near the base, while the top goes into tension because the bottom cannot move as the top tries, like a curling slab.

 

[concretemasonry]

Initial or initial cracks start at the area of of highest combined stress. After that, they start on the previous remaining uncracked sections depending on the stress at various locations.

The cracking is due to the combined stresses from both curing shrinkage (very high early) and temperature (due to the weather conditions that vary daily or seasonally). The tops of walls may see higher temperature extremes than the lower portions where the temperature of the soil is a moderator. Once a critical stress is met the crack will begin propagating.

It is not impossible for the initial shrinkage to occur due to the curing shrinkage early and then later cracks due to the combined stresses, including temperature due to the exposure later in the life of the wall.

Are there any sections that have the generally alternating or sequential cracking between the primary (assumed) cracks?

Engineer and international traveler interested in construction techniques, problems and proper design.

 

[csd72]

Shrinkage does not cause cracks, restraint causes cracks!

An unrestrained piece of concrete will normally shrink without cracking unless there is a large differential in temperature..

I would tend to agree with your boss that the cracks should start from the bottom.

 

[Hemifun]

Interesting discussion. I guess I would think the cracks would tend to start at the top also. My reason is my experience with unreinforced concrete curbs. Curbs are nothing more than 18" high by 8" thick walls that can be hundreds of feet long and are unrestrained other than the frictional resistance of the soil on which they are set. I have found that they will develop shrinkage cracks about every 10 feet. We now specify that contraction joints be placed every 10 feet and that pretty much works although we still get an occasional crack between the joints.

 

[mnovicki]

Thanks everyone. This website is a great resource. Yes, the restraint is what causes the cracks.

In the middle of the wall, at the bottom, the stress would be highest and I would anticipate finding a crack starting just above where the bottom of the wall meets the base. A crack could be expected to propagate from this point. I would then expect the crack to propagate upwards, following the stress concentration at the tip of the crack until it reaches the top of the wall. When the crack reaches the top, both sides of the crack retreat to their "shrunken" position. This is what confused me. On paper, the crack would appear to be wider at the top, which made me believe that the crack must have propagated from this point.

On the other hand, I think that a weak point at the top of the wall could be supported by a relatively rigid layer below, which could cause a crack to form at the top as well. This crack could have even been instigated by plastic shrinkage in the early stages, causing an extremely weak point.

Definately not an exact science, but interesting to think about. Thanks.

 

[Ron]

Keep in mind also that the mass of the concrete, in and of itself, can be the restraint...thus cracks can occur in linear members without any other apparent source. Further, concrete tends to shrink volumetrically toward its center of mass, so rebar that "gets in the way" of this shrinkage will cause cracking as well as keep the crack tightly closed.