Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

PT Slab Pre-Compression

Status
Not open for further replies.

strguy11

Structural
Nov 29, 2005
232
When designing a 2-way PT flat slab, I am having a small issue with the average pre-compression stresses at an area that has been thickened.

I am providing about on average 140 to 160 psi of pre-compression in the slab. (8" thick slab) However, in 2 of the design sections that are within an area of a thickened slab rib (14" thick), the pre-compression falls below the min 125 psi of ACI 318 due to the increase in area. My question is if i am still ok, or if i need to add more strands. It seems like since this is localized to a thickened region that i woulc be ok, and the design sections should be based on the areas outside of the thickened slab. If not, since the depth has almost doubled from 8 to 14", I would need to almost 40% more tendons just for this small area.

I can say that since i am relatievly new to PT design, i have been looking at other engineer's designs and replicating them in my software (ADAPT) to make sure i understand the principles of design and the software. Everything checks out except this small little design, which make me believe that it is not an issue, but wanted to know other opinions.......

Thank
 
Replies continue below

Recommended for you

You should be fine without upping your average precompression.

Much of the original testing done on PT slabs was done at a minimum of 125 psi and that level of precompresion was assumed in the punching shear requirements of the day. Nowadays, ACI punching shear provisions account for the beneficial effects of precompression directly so the 125 psi no longer serves that purpose.

Currently, ACI uses the 125 psi as a means of ensuring a minimum level of reinforcement is provided in each design section. That reinforcement can be any combination of prestressing steel or passive steel. Most international codes present this in what I consider to be a more rational form: simply specifying the minimum quantity of steel required (accounting for differences in Fy). I'd be pretty surprised if the 125 psi doesn't go the way of the dodo bird within a few more code cycles.

To be ACI code compliant in your situation, all you need to do is supply enough passive reinforcement in the thickened area that the effective total, including the prestressed reinforcing exceeds the standard temperature and shrinkage value of 0.0018. This is the approach that I'd recommend.

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.
 
Actually, 125psi is well below the "minimum reinforcement" requirement. If minimum is based on the usual requirement of ensuring that there is sufficient tension force available on cracking of the concrete to transfer the tension required to crack the concrete to tension reinforcement, normally rationalised to an ultimate capacity of about 1.2 * Mcr, the minimum prestress would be about 190-200psi (about 1.25-1.3MPa). But this assumes bonded prestress and reinforcement as discussed in the ACI318 commentary.

Later versions of ACI have changed the minimum reinforcement for unbounded prestress to simply a limit on bonded reinforcement ratio.

There is technically no minimum axial prestress, as long as the combination of reinforcement and prestress is sufficient to provide the minimum above and require capacity.

With less than 125psi and unbonded prestress, I would think that you should be providing minimum reinforcement equivalent to that required for an RC section without any PT.
 
The minimum reinforcmrmt that I mentioned refered to the 0.0018 requirement in ACI. For comparison, 60 ksi at 0.0018 is equivalent to 108 psi.

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.
 
The problem is when it is based on cracking moments, which the .0018 is for RC sections, the cracking moment for a PT section is increased by the prestress and the minimum is closer to 200psi!

Also, the .0018 should be dependant on concrete strength, as it is finally in the latest ACI318.
 
strguy11,

TR43 "RECOMMENDS" minimum P/A of 0.7 MPa (101 psi). In EC2 and BS8110, I don't think there is a requirement for P/A. I am not sure how TR43 derived this figure 0.7. But my interpretation is this if you have lesser P/A, your PT member will behave more like conventional RC (eg. Formwork dismantling guideline).

I don't know which part of the world you are from. Here in Singapore, for bonded PT structure, It is common to dismantle the formwork 3 days after concrete casting and then prop back continue upper storey construction. So if P/A is very minimum, it might not behave properly after 3 days and might crack severely.

You can count & rely on RAPT and KootK replies. They are very experienced and knowledgeable. :)




 
Thanks for the replies... One last question....

I have been looking at this entire slab as a 2-way slab, however i have seen posts that indicate the thickened slab area (3'-6" wide 14" deep) should be designed as a T-beam section rather than as a 2-way. I have changed my design sections to require this and everything checks out.

My question is in the other direction, should these be set to a one-way slab? In this direction, there area essentially 3 spans, however at one of the "supports", there is no slab thickening, so it would seem to be that the 2-way would be more applicable here... But, if i have the design criterial for the thickened slab area as a T-Beam, how should the spans in the other direction be modeled, particularly if non uniform.

See my sketch....
 
 http://files.engineering.com/getfile.aspx?folder=fb1ce780-6f81-4bb1-8d98-826c83133e5e&file=Capture.JPG
1) my gut feel is that your behavior will still be fundamentally two way.

2) your governing concrete code may have advice to offer. Mine (Canada) considers similarly proportioned systems, termed "slab bands" to be fundamentally two way systems.

3) You can test the one-way / two way character of the system yourself by cutting a design section in front of the thickening to view the lateral distribution of moment there. If that moment is fairly uniform, one way. If not, two way.

4) Many of the code rules for two way slab design are aimed at achieving a lateral distribution of moment capacity that reflects a lateral distribution of moment demand based on elastic plate behavior. One of the nice things about FEM, when done properly, is that you end up with a moment demand distribution that's probably better than that specified by concrete codes. Reinforce your slab reasonably for that moment distribution and it shouldn't much matter how you classify it.

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.
 
Normally a full slab with band beams in one direction is treated as one way with some extra reinforcement added over the columns in the slab direction (by those who think it through logically anyway. Punching shear is still checked as the band beam is wide enough to allow a punching failure mode unlike a normal beam. I would normally concentrate the slab direction tendons slightly at the column also to have 1 tendon immediately either side of the column.

In your situation however, with only a single band beam in one span, the slab will be basically 2way but with more moment carried in the column strip where the band is effective than would be the case for a flat plate without the band. So this area will be somewhere between one way and two way.

Kootk is correct in that this can be fully analysed in all of its complexity with FEM software (except that after cracking the stiffnesses will be different and moment distributions different but we normally forget about that). Where many designers miss the benefits of doing this is that they then use the FEMs default design routines which based on ACI/PTI practice would design for a design bent that is a full bay width assuming the moments are averaged over this width. This completely negates the logic and benefit of using FEM for the analysis as you are effectively designing it as one way if you do this. And get very misleading results compared to real life stresses etc. Also, with the band beam included this should not be done anyway as you cannot assume the full bay width for the effective flange on the band beam. (though I have seen this done often by designers who do not understand what they are doing and have listened to someone else who has the same problem).

You have to make sure that you select design strips that account for the different intensities of moment in different areas, similar to the column/middle strip logic normally used for flat slab design. If there are randomly placed concentrated load effects, you then need to define even more strips to pick up the more complex moment patterns.
Where the beam exists, it will still be a T-beam shape including some slab, but not the full panel slab width you would get as a one way slab action.
 
@rapt: we could use you over here: Link

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.
 
It looks like you are using Ram Concept for this. On the point of whether to design it as a one way slab or a two way slab I often recommend that users review the deflection or moment diagram contours to make an educated decision about that (same as KootK was suggesting).
As to the original problem, if the original problem was for the strips normal to the band beam, then you probably want to adjust those design strips' "support width" so that the first section is at the face of the beam (or trim that strip). For strips parallel to the beam I think the others here already addressed things.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor