post tension design

[sklgleb]

Hi all,

I've been tasked with determining a nominal thickness of a post-tensioned slab - the actual slab design will be done by another firm, but we do need to know the minimum so that the architect will not have to revise elevations later. I am basing my calcs on puncture shear w/ shear from unbalanced moment factored in - we are not using any drop panels/beams/capitals - just plain two way slab on columns. However, there are just some really funky column perimeters that I am running into that I cant get around. One case is where the the slab connects to the (edge) column at one face only, and does not wrap around - I am getting a pretty significant unbalanced moment, but my Jc term, as I see it (please let me know if this is wrong) is reduced to the (1/12)bh^3, b being the column width, and h is 1/2 the depth of slab - critical distance. Naturally Jc is coming out way to small, and my shear stress from moment is on the order of several thousands. Is this a no go, or is there another way to look at this - we will be adding in shearheads later, but at this moment we are questioning the feasibility of just using concrete. In general, for those of you who have done this enough - an off-the-top-of-the-head estimate for slab depth with a typical bay of 36x28? - I got 10" right now.

Thanks,

G.S.

 

[KootK]

Sweet. Another chance to get myself in trouble on a PT thread.

Residential loading?

My guess is that the edge column situation will turn out to be a no go requiring modification. It's not really punching shear any longer. Rather, it's one way shear and flexure right at the connected column face.

Additional thoughts:

1) One of the drivers behind PT use is thinner slabs. I find that, to take advantage of that, I almost always end up using stud rails.

2) The balancing load effects associated with PT will tend to reduce the unbalanced moments at your columns (a portion of the DL unbalanced moment). It's conservative to ignore that for prelim however.

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.

 

[sklgleb]

yes, 40 live 25 dead.

 

[sklgleb]

doh, one way shear and flexure it is - smacks self in head...

 

[Retrograde]

Take a look at this:

http://www.ccaa.com.au/imis_prod/do...es/CCAAGUIDE2003-T36-Long-span floors-TBR.pdf

I would think your slab needs to be around 280mm thick or 11". Possibly deeper for an end span condition.

 

[hokie66]

The 36' span direction, as well as the silly one face only connection, is going to kill a flat plate. The reference which Retrograde gave would have you in the range of 11 or 12" for a flat plate, and close to the not economical range. But that document does not envision connecting a flat plate to one face of a column.

 

[rapt]

sklgleb,

Yes, one way shear and flexure, possibly with torsion as well depending on the transverse column stiffness/moment. But it can only be resisted by flexural reinforcement that develops fully past the slab/column intersection and extends to the back of the column. You will not get much capacity unless it is a really wide column and you can develop a lot of bars into it. You will need a full beam cage with top and bottom reinforcement extending out from the column.
It is unlikely that you will get any assistance from the PT. It is possible you could even get some direct tension stress at the joint caused by the prestress shortening and restraint to this by the column. This would reduce the shear capacity further

I would think 10" slab is definitely too thin, especially if the 36' span is an end span.

Kootk,

Commenting on PT again. The Mafia is back!

Though the effect is the same in this case, it is still only the Secondary (Parasitic) Prestress moment that can be subtracted. For a column moment from a prestressed slab, Mp = Msec (because there is no prestress in the column P.e in the column is 0) so they are the same value.

But technically, you cannot reduce the loading by the balanced load effect, you can reduce the column moment by the effect of prestress in your combinations.

And the Prestress Secondary Moment only has an ultimate factor of 1 (can vary from .8 for prestress effects that reduce external load moments to 1.2 for those that increase them in some design codes (Eurocode does this)) while the DL will have 1.2 to 1.4 depending on the design code, so you cannot reduce your DL by the balancing effect or the prestress effect before you do your combinations of loads. You need to include it separately in the ultimate load combinations with its own multiplying factor.

 

[KootK]

It is possible you could even get some direct tension stress at the joint caused by the prestress shortening and restraint to this by the column.

This is an excellent point that I hadn't considered. If it's an unbonded, ACI 318 design, you'll probably need to run a couple of tendons out to the back of the column which may help with this.

Commenting on PT again. The Mafia is back!

I know, right? As it turns out, I'm oddly intoxicated by the unique combination of humiliation and rapid PT knowledge acquisition. Mmmmm... whatcha' wearin?

Though the effect is the same in this case, it is still only the Secondary (Parasitic) Prestress moment that can be subtracted.

I'm not so sure. Below, I submit the following:

1) In the upper sketch, I show a beam with no secondary moments that is prestressed such that there would be no column moments at all.

2) In the lower sketch, I show a statically equivalent model that, I feel, helps to elucidate the principles involved. Technically, I suppose that the rigid links should be perpendicular to the tendon.

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.

 

[KootK]

1) In the upper sketch, I show a beam with no secondary moments that is prestressed such that there would be no column moments at all.

Actually, I think that I take this back. Not the part about the elimination of the column moments but the part about their being no secondary moments. The system that I sketched wasn't statically determinate so I'm thinking that the moments induced on the columns by the prestressing alone actually are hyperstatic/secondary/parasitic. Is that the right way to think of it? If so, then I suppose that my only objection would be the suggestion that the effect is minor. Code factoring aside, one could conceivably post tension a member to completely elminate the unbalanced column moment resulting from any one set of applied loads.

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]

Kootk,

Yes, there are moments in the columns from prestress in your case.

The moments in the columns are hyperstatic/parasitic as there is no P.e in the columns as I mentioned earlier. Unless someone were to actually prestress the columns as well. Unusual but it has been done before in portal frames, and in that case it would be important to differentiate between Mprest and Msec as only the Msec could be used to reduce the column design moments.