PT Slabs - Restaint to Shortening Cracks and Forces

[RFreund]

A few questions related to shortening related cracks and forces.
Background for reference:
A muliti-story building (say 3-4 stories) w/ a basement for Parking. The construction is PT-Slabs supported by concrete columns and shear walls. To reduce RTS cracking at the first floor but still support the walls for out of plane forces we have released the slab as shown below.

Questions:
1. In the section below what are your thoughts on having issues with the CFS framing moving with the slab and causing issues with the brick? Supposedly the cladding would be installed about 2 months after this floor is constructed. I'm tempted to say this will be ok...

2. Does this chart from ADAPT seem right for the amount of shortening that will occur at 60 days?

3. At the upper levels the columns would be more flexibility and may allow for more uniform cracking. However, do you typically design the columns for shortening forces? Therefore the outer columns would have significantly more reinforced (due to shortening) than the interior columns. Or could you distribute this force more evenly to the columns so how. I feel like I see this ignored sometimes.

4. Getting back #1 - lets say I didn't connect the CFS to the slab and somehow connected it to the fdn wall. The top of the wall would still move inward so I would still get differential movement, but in this case it would be over a 10' span rather than 6". Any better?

Thanks!


EIT

http://www.HowToEngineer.com

 

[KootK]

1) Is there not some mechanism that, at some point, locks the relative displacement between the slab and the foundation wall? Are you delivering shear to the wall through the angle in a way that allows relative out of plane movement to occur?

In theory, soil pressure will cause your wall to move inwards with the slab edge so no relative displacement. That'll be less true at building corners etc however.

2) I don't have any basis for comparison really.

3) My experience has been that designers plan expansion joints and delay strips in order to keep shortening down to a level where it can be ignored in the design of the columns. Is that the way that it should be done? Not sure. Is it done rigorously enough to begin with? Probably not.

4) But this would entirely eliminate differential movement between the CFM and the brick which is your primary concern, right?

For other reasons, I've tended to go with mildly reinforced main floor slabs. I guess that's why I haven't yet encountered this.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[RFreund]

Is there not some mechanism that, at some point, locks the relative displacement between the slab and the foundation wall? Are you delivering shear to the wall through the angle in a way that allows relative out of plane movement to occur?

No, nothing ties them together. The angle is there to resist the horizontal reaction at the top of wall.

In theory, soil pressure will cause your wall to move inwards with the slab edge so no relative displacement. That'll be less true at building corners etc however.

That is a good point. The would construct the wall, construction the slab, backfill the behind the wall. So some shrinkage and elastic shortening will have occurred by then. However, there still would be some further shortening and thus some wall rotation. Maybe this just happens to be in an acceptable range so that everything is moving together.

My experience has been that designers plan expansion joints and delay strips in order to keep shortening down to a level where it can be ignored in the design of the columns. Is that the way that it should be done? Not sure. Is it done rigorously enough to begin with? Probably not.

Hmmm, I see, that makes sense. Any parameters on what level is 'acceptable' and how to get there?

But this would entirely eliminate differential movement between the CFM and the brick which is your primary concern, right?

I think I said this wrong. Lets ignore the foundation moving due to lateral earth pressure. I was thinking that the first floor foundation wall would stay where it is. The second floor would move inward due to shortening, thus pulling the CFS and masonry along with it. So you would have the same displacement but 10' between points of displacement. Now considering that he wall may move due to earth pressure...Maybe all is well, sorta.

EIT

http://www.HowToEngineer.com

 

[KootK]

No, nothing ties them together. The angle is there to resist the horizontal reaction at the top of wall.

So how do you get main floor slab diaphragm shear out into the basement wall?

Any parameters on what level is 'acceptable' and how to get there?

1) Take a look at how stiff your columns are. A 12 x 36 is a lot stiffer one way than the other.

2) Pick a level of shrinkage induced moment that you think will have a relatively negligible impact on your columns or that you're willing to include in the design of your columns.

3) Estimate how long of a continuous slab you can have before shrinkage generate moments larger than #3.

Or, just separate your stiff walls, apply some rules of thumb, and hope for the best. It's the difference between what I've done and what I think you should do. I know, it's terrible.

I think I said this wrong. Lets ignore the foundation moving due to lateral earth pressure. I was thinking that the first floor foundation wall would stay where it is. The second floor would move inward due to shortening, thus pulling the CFS and masonry along with it. So you would have the same displacement but 10' between points of displacement. Now considering that he wall may move due to earth pressure...Maybe all is well, sorta.

Ah, I get it now. I would not expect the system to be that sensitive to that type of movement. Again though, corners are probably the trouble spots where you'up up against perpendicular brick.



I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[RFreund]

So how do you get main floor slab diaphragm shear out into the basement wall?

Well ideally you dump it into shear walls which are located near the center of the diaphragm. In the real world I am forced to use only as many dowels as needed near the center of the walls (away from the corners of the building). Where the wall is more flexible.

Thanks again for the comments. It is interesting that really all this applies to non-PT slabs (except for Elastic Shortening), but in a mild slab it is assumed that the cracks will be more uniformly distributed in the slab. Although the columns could be the "weak link" where you get larger cracking.


EIT

http://www.HowToEngineer.com

 

[KootK]

Any time RF. It is an interesting subject. I'm hoping that rapt pops in to give us his thoughts. And yeah, I suspect that there are some PT issues that are also less severe, but still under appreciated, CIP issues that we tend to turn a blind eye to.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[rapt]

RFreund,

Have you calculated how much shortening you expect in this slab? In both directions? Remember to allow for temperature change as well as shrinkage. They are the 2 main effects. Creep and elastic shortening will be less. Temperature may be hard to estimate as it depends on the location, time of year and temperature at time of pour. Can your wall ties take that movement?

Edge walls are normally released like this to alleviate restraint effects in the slab parallel to the wall rather than perpendicular. Unless the wall below is very short, it will not provide a lot of restraint to the slab perpendicular to the wall, and soil pressure over time may actually put the slab into compression perpendicular to the wall if there is a connection to the slab.

Tying the CFS wall to the top of the slab is still going to provide restraint in the slab parallel to the wall!

Is your basement wall cantilevering, or does it need to be propped for retaining action? Normally it would be propped by the slab after the slab is poured and connected (possibly using a delay strip around the boundary) where I come from.

My preference has always been to spend my money to provide crack control reinforcement in the slab rather than trying to release it. If you fail with your releases, you are stuffed.

The retaining wall will be cheaper as it is propped. And as I have said many times on this site (to the point where people are probably sick of it) prestressed slabs have really good crack control : until they crack. Then the crack control is terrible. This is especially so with unbounded PT and even worse with banded/distributed tendon systems. Hence my preference to do something we know works which is to provide sufficient bonded reinforcement to provide good crack control, and that requires more than the normal T&S minimum reinforcement as the normal figure in the code is for an unrestrained slab or very nominal (read as none according to some more knowledgeable than me) crack control in a restrained slab.

 

[RFreund]

Rapt, thanks for the response. I few replies below:

Have you calculated how much shortening you expect in this slab? In both directions?

I've just been going by what I've been given as rules of thumb. The plan dims in this case are about 80'x200' (fyi).

Can your wall ties take that movement?

Theoretically, probably not. But I've been told/suggested that it is "the best alternative" and haven't had issues with it.

Is your basement wall cantilevering, or does it need to be propped for retaining action? Normally it would be propped by the slab after the slab is poured and connected (possibly using a delay strip around the boundary) where I come from.

It is propped. We originally had the delay strip detail but it was fought against (of course). I think b/c the GC wasn't used to dealing with these issues. With some explanation they have come around since and we may go back to this detail.

Hence my preference to do something we know works which is to provide sufficient bonded reinforcement to provide good crack control, and that requires more than the normal T&S minimum reinforcement as the normal figure in the code is for an unrestrained slab or very nominal (read as none according to some more knowledgeable than me) crack control in a restrained slab

I like this idea as well. Any recommendations on what you would consider "sufficient bonded reinforcement to provide good crack control"?
I think we are dealing with a PT-design/build contractor that plans to provide no crack control unless we "force/encourage" them. Don't get me wrong they are will to work with us to get a good result.

(read as none according to some more knowledgeable than me)

Nonsense! I know of none.

Thanks again

EIT

http://www.HowToEngineer.com

 

[rapt]

So what is propping the wall? The slab cannot with that detail.

You can calculate the shrinkage shortening from ACI209. It will be around 1 - 1.5" in the 200' direction total but will depend on concrete mix, environment and especially slab thickness.

Temperature will depend on conditions but will be a little less than 1" in that direction probably.

Axial shortening about .1".

Creep .2-.3".

Adds up to a lot! About 2 - 3" total so 1 - 1.5" or so from each end.

The amount of reinforcement you need depends on the degree of restraint and the level of crack control. Australian code would say the absolute minimum for fully restrained is .6%. If you want really good crack control, this would rise to about 1% for fully restrained. If fully free to move when stressed this would reduce depending on the level of P/A.

PS thanks, but in the shrinkage/creep/cracking field there definitely are some that I know and presumably a lot I do not.