PT Slabs 5

[zineelabedine]

Does ACI allow for dual banded system in two way PT Floors ?
Which books or manuals you would recommend for post tension slabs design , that have a good elaborative in detail explanation of manual calculation steps and design with solved examples?

 

[rapt]

I assume you mean a standard column/middle strip in both directions. Yes, ACI allows this as long as the load balances in all 4 strips are consistent. In fact, it should be the preferred layout as that is how the slab wants to act.

Make sure if you have drop panels that the drop panel effect is only used for the column strips in each direction. There are a lot of designers using it for the full slab width (ACI average moment design) and that is completely wrong.

Any book that does not recommend the ACI/PTI one way averaged banded/distributed tendon layout.

 

[KootK]

Does ACI allow for dual banded system in two way PT Floors ?

Not as of ACI 2011. That document states that, for uniform loads, tendons in at least one direction must be spaced at no more than 5' o/c or eight times the slab thickness.

Double banded systems have been used in the past, however, and some folks are interested in bringing them back.

DISADVANTAGES

-Cumbersome to install the "basket weave" approach of old where there was still some amount of uniform tendons.
-Complex to design at irregular slab areas.
-Reduced drape
-Higher tensile stresses in some areas.
-Higher initial stresses
-More prestress "shadow" rebar

ADVANTAGES

-Better punching shear performance.
-Better capacity to transfer moments at columns.
-Lower placement costs for regular slabs and using no distributed tendons at all.
-Easy post construction coring, at least in some spots.
-Potential for incorporating additional technologies such as voided slabs.
In my estimation, the best books are these, in order of preference:

1) Bijan Aalami (Link)
2) Ken Bondy (Link)
3) Williams and Kahn (Link)

For prestress fundamentals, Michael Collins, T.Y. Lin and Ian Gilbert are my favourites. Neither of their books treat post-tensioned systems is as much detail as the others that I listed 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.

 

[zineelabedine]

rapt,

thanks a lot. The drop panel effect remark is very considered.

 

[zineelabedine]

KootK,

That's a great help actually, and i highly value your interaction and recommendations.
About that ACI clause you mentioned, is it still exist in the latest version?

For now, i'm exploring my options in the tendons layout, i'm not really in favor of any specific configuration, although i think banded/distributed might be convenient, the dual banded came to mind, so i'm investigating it a little more. Thanks very much.

 

[KootK]

You're most welcome. That clause is in fact still in ACI 2014.

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,

18.12.4 does not preclude the use of banded/distributed in both directions. As the commentary states the spacing limitation permits the use of banded in one direction which would otherwise not be acceptable in a 2way slab system as it is essentially a one way solution.

As long as the tendon spacing is not > 8 times slab thickness in one direction, 2 way banded/distributed is allowed as well by this clause. Normally we would modify the spacing limits for 2 way banded/distributed to 6 times slab thickness in the column strips (which are carrying more than the average load) and 10 times slab thickness in the middle strips (where there is a 2 way pattern of tendons allowing a higher spacing than would be allowed for a one way pattern) (refer TR43 and AS3600). But if someone wants to be pedantic they can apply the 8 times limit.

The only advantage for one way banded/distributed is tendon weaving especially with unbonded tendons. It is not as much a problem with bonded systems as the duct is normally laid first in 6m lengths which are relatively easy to weave.

Most of the American text books on this will still encourage design using averaged moments across the whole panel. I would avoid them if trying to get a true understanding on how PT works.

Read TR43 2nd edition British Concrete Society report.

 

[Ingenuity]

The document below (attached) implies that dual-banded (100% column/0% middle strips in both directions) are NOT currently permissible in ACI-318.

 

[KootK]

18.12.4 does not preclude the use of banded/distributed in both directions.

Agreed. But it does preclude the use of 100% banded / 100% banded which is what "dual banded" means to me. Perhaps zineelabedine can clarify what his intended system was.

Read TR43 2nd edition British Concrete Society report.

What the heck? I've been asking you for references on this "way better than ACI" method for years and you never mentioned this document that you helped create? I've been reading 800+ pg Australian pre-stressed concrete books that you recommended when apparently everything I need is in this 133 page pamphlet...

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,

Actually, banded/Banded with no middle strip tendons is normally an accepted method as well. The bands provide the column strips and the slab between the bands is designed as an 2 way RC slab so the reinforcement covers the spacing problem. This was called "support strip" prestressing in the old days and was used in Netherlands in the 1970-80's. I could never understand the advantage myself.

The only layout that should be rejected outright is distributed/distributed as it does not provide a load balance or load path for the prestress, though it can be made to work with FEM analysis and good detailed design (no averaging simplifications etc) allowing appropriately for cracking etc and long term deflections.

TR43 is really a document for PT slabs initially to BS8110 and Eurocode in its second edition. 1st edition was based purely on the ACI method. 2nd addition we also gave a real design method so it includes both. I did not think it would relate to Canada and USA.

I would still recommend the text books for an in depth understanding of PT design, Gilbert, Guyon, Leonhardt, Collins and Mitchell and others.

Like all documents, this one is not complete either. For example, it does not cover the drop panel comment I made earlier, much to my disgust. Not that they do not agree with me, their comment was that no one would be stupid enough to design a flat slab that way so they would not say anything about it. Tell that to the PTI/PCI ((can't remember which one now and cannot find my copy)design example doing exactly this in an older version in the 1970/80,s. I do not know if it is still there.) And a large number of designers in SE Asia, India, Middle East and even Adapt's own design group were/are doing it.

 

[rapt]

Ingenuity,

I was assuming he meant banded/distributed in both directions! That is column/middle strip in both directions. As I understand it that is allowed specifically in the latest ACI (don't have a copy but I have been told 3rd hand).

 

[Trenno]

I suppose it should be mentioned that for irregular geometry, you should have a look into the 'lines of zero shear' method for determining appropriate strip widths. This method is usually employed when running 3D FEM models.

I've found the following document to be a good reference. Also I highly recommend TR43, very good read.

 

[zineelabedine]

rapt, Kootk

I actually meant banded/banded without a middle strip tendons, sorry for causing any confusion.

 

[sticksandtriangles]

Trenno, I was interested in downloading this document. It does not seem to be downloading from the the website unfortunately . I will shoot an email to the company and see what the hold up is.

Thanks for the link!

 

[KootK]

Trenno, I was interested in downloading this document. It does not seem to be downloading from the the website unfortunately mad

I had the same issue and resolved it just a moment ago. If you come at it through "Downloads" it locks up. Instead, go to "Orders" and click "View" beside the order.

@Trenno: it's a nice doc, much obliged.

I did not think it would relate to Canada and USA.

Canada Schmanada. When the Batman of PT says there's a better way, you look into it. I don't care if it's in Farsi.

I would still recommend the text books for an in depth understanding of PT design, Gilbert, Guyon, Leonhardt, Collins and Mitchell and others.

I've been trying to acquire Leonhardt for years, since you first recommended it to me. "No longer in print"; "$854"; "Not available"; "$1219". This morning, I picked it up on Amazon for $39. Yay me! If any young engineers would like to have the book bequeathed to them pending my timely demise, feel free to trow your HP's in the ring.

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.

 

[sticksandtriangles]

Thanks for update Koot

 

[rapt]

zineelabedine

Why would you want to do it?

What is going to provide load carrying capacity and crack control in the remainder of the slab between the bands? As I mentioned above, this can be done by reinforcing the remainder of the slab as =RC two way slab spanning to the bands. But then you are doubling up on reinforcement.

The ACI logic (lack of logic) that says that the locations of the tendons across the width of a panel does not affect the slab capacity is very dangerous. It requires a proper load balance distribution of the prestress and banded/banded is not a load balance distribution. And the ACI logic can require a lot of unaccounted for redistribution to make it all work. It has also led to the "average moment" design logic for service stresses which results in the significant underestimation of service stress and cracking and deflection calculation. Ductility problems also as you are not considering the concentrated stresses near the columns. And then people do it with drop panels and significantly underestimate the strength of PT slabs (15-20% under!!). It also does not allow the use of partial prestress design as you are not dealing with real stress and strains so cannot calculate crack widths or deflections properly.

In the hands of people who do not really understand what they are doing, it is dangerous.

Column/middle strip in both directions is a load balance pattern and it actually matches how the slab wants to work so it is always my preferred solution, and as I hate unbonded PT then I do not have to worry about the laying problems. The tendons are basically where the moments in the slab are.

One way banded/distributed is also but it is a one way solution to a two way problem and requires an enormous amount of redistribution to give the final ultimate capacity and also does not provide crack control in a lot of places where it is required, so to detail properly a lot of extra (unnecessary for strength) reinforcement is required). Also, if the stresses are allowed to get above the cracking strength, any estimate of deflections will be low as redistribution caused by the unreinforced cracking will increase the deflections above the normal calculated values. In the early days this method was only used where stresses were kept very low and it was not really a problem. Unfortunately with time, designers forgot about the limitations and started allowing much higher stresses. In fact for a very brief period in the mid 2000's ACI allowed the average tensile stress in the negative moment region to be .6root fc and called it uncracked. So the real stress was about 1.5 times this and it was really cracked.

Any combination of column/middle strips between there 2 extremes where the middle strip tendons are consistent with the column strip tendons in the other direction is also a load balance layout and is technically acceptable.