Banded post tensioning system 1

[ajk1]

An old existing unbonded post-tensioned 2-way flat slab system with drop panels and column capitals has uniformly distributed tendons in one direction and banded tendons in the other direction. The banded tendons are in the column strips and there are no tendons in the parallel middle strips. The design loading is for truck (12.0 kPa, or 250 psf ±).
Is that layout of tendons a common practice? Does it meet Code?

 

[dik]

I don't know about current times, but the banded system was common construction in Canada for post-tensioned slabs and was code complying. I've done numerous buildings using this method.

Dik

 

[KootK]

In my experience, the banded / distributed tendon layout is the most commonly used in North America by a wide margin. Distributed / distributed is questionable technically. Banded / banded offers some performance improvements but is not popular with North American contractors.

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.

 

[Ingenuity]

ajk1:

The "old" slab probably meets code with the industry-practice of using banded-distributed tendons, but may not meet the Owner's serviceability expectations.

I am going to hazard a guess and say the slab has wide flexural cracks, not well distributed (like you would achieve in RC) and if it is an exposed environment, probably leaks.

The amount of bonded mild steel reinforcement may not meet the requirements of today's codes (not too sure of CSA codes of today, but I assume they near match ACI pertaining to PT design requirements).

With a 250 psf truck loading I hope there is a decent quantity of well distributed rebar, including the top of slab at midspan due to 'pattern' loading of spans.

 

[rapt]

ajk1,

The tendon layout is accepted in North America for flat plates. The design approach can result in acceptable ultimate strength via some very significant redistribution, the effects of which are normally ignored by designers. The design method usually ignores moment concentrations in column strips and uses an average over the whole width.

Some have extended its use to flat slabs with drop panels. This is very questionable with drop panels as all of the tendons in this design method are assumed to have the effective depth of the drop panels at the supports. This is obviously completely incorrect to anyone who understands engineering and leads to a significant underestimation for the slab capacity in the distributed tendon direction.

For service design, the method assumes an average stress over the width of the slab, ignoring stress concentrations near supports. Even for flat plates, this is obviously underestimating the service stresses and for heavily loaded slabs like this one, normally leads to significant and often unrestrained cracking depending on reinforcement supplied. Partial Prestressed design is not possible as you are not dealing with real stresses and for this level of loading, the slab will normally be partially prestressed. ACI limits the hypothetical average concrete tension stress but this does not stop cracking from occurring. TR43 in UK puts much more stringent limits on tension stress recognising that the calculated stresses are actually a gross underestimate of the real stresses. And does not allow partial prestressing for this type of design approach.

If the slab is cracked based on real stresses and uncracked based on the average hypothetical stresses, then you have no idea how much redistribution has occurred even at service loads so you have no idea of the effects on crack control or deflection, and the calculations will have assumed it is all uncracked!

Also, ACI requires a lot of extra reinforcement for unbounded slabs designed this way. Canadian code does not from what I can tell, so who knows what you really have unless you find out the basis of the design!

Having sais all of that, how can we know if it meets code, if we have no idea of the amount of PT, the amount of reinforcement supplied, the service stresses and everything else?

 

[dik]

rapt:

If you've done a 3D finite element model of a slab with deep drops (I usually use the depth of the slab plus inch or two for the drop depth, you'll be surprised at the effect of the drop in the 'middle strip'.

Dik

 

[rapt]

Dik,

On the applied moments, or the capacity of the tendons?

 

[ajk1]

To Kootk - thank you, you seem to have addressed my question most clearly and directly. As I understand your response, the banded system with no tendons in the middle strip that is parallel to the column strip containing the banded tendons, and no top rebar in that middle strip, is widely used in North America. So that would imply that that arrangement of tendons is in conformance with Building Code requirements for distribution of reinforcement and tendons. Am I interpreting your response correctly?

To dik: you also seem to be saying that the Building Code allows that distribution of tendons and rebar. In your second post, I do not know what you mean by "surprised", nor do I know exactly what you mean by deep drops. In my designs I do not specify drops that have a projection below the slab soffit that exceeds 75% of the slab depth, unless they are required to make the shear work, in which case I do the flexural design based on a fictitious drop that is limited in depth as noted above.

To rapt: you raise interesting points. Re your last sentence - I did not ask if the design met the Code. I asked the specific question of whether the Code permitted an arrangement of tendons and rebar that has no negative reinforcement in the top of the middle strip.

 

[dik]

rapt:
on the stresses through the cross-section... it may be a feature of the deeper drops that I use... tends to increase the negative moment due to the greater stiffness, increases the flexural strength due to the greater depth, and tends to minimise the reinforcing.

Dik

 

[KootK]

Am I interpreting your response correctly?

You are indeed.

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]

dik,

There is a big difference between elastic service stresses and ultimate strength of the slab. And even the stress in the service case are dependant on the selected design strips.

My comment above was about service stresses in flat plates done by average moment versus distributed moments.

But when you get to cracked section analysis and ultimate strength analysis, average moment approach is completely illogical for flat slabs with drop panels. As it is for flat plates at service.

 

[rapt]

ajk1,

The layout might be acceptable only if the reinforcement arrangement and other design aspects are handled correctly. If not, then it may not be to code. You cannot separate the tendon layout from the overall design.

But the Canadian code does not appear to have special limits on stresses and limits on partial prestressing and special minimum reinforcing requirements for average moment design, so it does not cover the design method usually used in North America for slabs with banded/distributed tendons.

There is then the further argument I made above that the design method used in North America for slabs with banded/distributed tendons - averaging the moment over the full width and assuming a T section full width of the panel in the area of the drop panel - should never be used for flat slabs with drop panels, no matter what code you are designing to.