Post-tension slab fold design checks

[mmodfr2013]

I am reviewing my company's slab fold details for unbonded post-tensioned elevated slabs, and in doing document research I haven't found much of a consensus on how slab folds should be treated. I attached a clip from a magazine article that gives the kind of information I've been seeing, which is fairly general in nature. I was wondering if anyone could share their input on design checks or calculations that are relevant. I've look at the balanced load calc (8PA/12L^2) but it doesn't seem to apply very well in these local areas. I've seen other engineers assumed a harped tendon path and just work up an FBD to get a normal force to the slab surface at the fold, but that doesn't seem quite right either.

 

[Boiler106]

Im also interested in this. Is there a separate design check outside of the design software that needs to be performed?

 

[KootK]

Preliminary: I'm going to read between the lines here and speculate that the root concern is the tendency of post-tensioned slabs to "unfold" as the tensioning attempts to straighten out across a slab step or fold. For concentric prestressing, there actually is no such tendency. If you fine fellows find that to be counterintuitive, rest assured that you're in good company. This is the thread in which rapt, Ingenuity, IDS, and others patiently taught me this same lesson: Link.

 

[KootK]

Diving deeper into the realities of this phenomenon, I feel that the following are some of the salient issues regarding post-tensioned slab folds:

1) Slab folds will generally compromise the concentricity of the prestress locally, as we go from skinny cross sections, to fat cross sections, and back to skinny cross sections again. And, obviously, the gal on site cannot profile the tendon perfectly, especially at the areas of complexity. So I feel that it is generally prudent to have some element running parallel to the fold that possesses meaningful stiffness and strength such that it can resist the inevitable transverse load demand that results from an imperfectly profiled tendon attempting to straighten.

2) Where tendons are vertically curved, they create a local, transvers load demand in the slab (blowout). For this reason, it is preferable to keep the plan locations of such areas of tendon curvature located where the slab is locally thick rather than where it is thin adjacent to the local thickenings. This is especially important as a construction QC issue as it's a pretty easy thing for field personal to inadvertently shift the high curvature locations over a foot or two in plan.

 

[KootK]

Hypothesizing why we don't see more problems than we do at slab folds...

 

[rapt]

In terms of analysis, If it gets really complicated, then you can break it down to 2 components

- every change in angle of a tendon results in a force. If the tendon has a parabolic profile the force will be the one you showed (I assume the 12 is a conversion from inch to ft). This is independent of concrete shape, purely looking at the tendon shape.

- every change in direction of the concrete centroid will induce a force. In Koot's second (b) above, there are 2 changes in centroid at each step location, so it becomes a force couple or concentrated moment, but if you think in terms of shear lag, there cannot be a sudden change in force like that so it becomes a force couple allowing for a logical "slope" for the change to represent the way the axial force will spread through the concrete as a result of the depth change normally assuming about 30 degree slope rather than a sudden step.

In Koot's first 2 diagrams, the force from the change in angle of the tendon at every point exactly equals and opposes the force from the change in centroid of the concrete.

Do not assume that all software does this properly. It will depend on the knowledge of the programmer. Not sure about current versions, but I know that earlier versions of Adapt PT grossly approximated these effects and could give very misleading results in more extreme cases.

It is hard to comment further without seeing the actual situation you are looking at.

 

[mmodfr2013]

rapt and KootK,

Thank you both for your responses.

I don't have a specific situation I'm looking at, this is part of a periodic review of our internal design and detailing methods for PT slabs. Mostly, looking out for instances where we do it this way because that's the way we've always done it, and then verifying the engineering behind those details, meeting updated code requirements, etc. It would be nice if I could work up a typical detail for a slab fold, but with all the variables involved I don't think that's going to happen.

 

[KootK]

Can you share the title of the article that you posted?

 

[mmodfr2013]

KootK,

The article is posted here:

https://static1.squarespace.com/sta...gineering+Tips+For+Post+Tensioning+Part+2.pdf

There is a part 1, but for whatever reason it hasn't been made freely available.

 

[KootK]

Thanks mmodfr2013. Kline outputs a lot good PT material.

I've been pondering another wrinkle.

In the first sketches that I posted, from TY Lin, the magic of "no unfolding effect" is a product of perfect balance between the tendon balancing load effect and the anchorage loads applied to the ends of the memeber. However, that depends on both of those components actually being present. Folds and ramps seem to happen disproportionately at grade and below grade where the presence of structural walls will result in the prestress getting syphoned out of the slab and into the walls to a degree. Obviously, when you diminish one slab action but retain the other, things will start to drift out of balance. For this reason, among others, basing "fold beam" design on a conservative assumption like the harped profiling that you mentioned earlier may well be prudent.

 

[rapt]

I have not read the whole paper in detail, but I would definitely ignore his advice on Drop Panel sizing for PT slabs!

Not sure why you think the harped profile would be conservative Kootk. Could be very un-conservative depending on the situation.

 

[Celt83]

Kline are DC folks like myself, so we don't tend to deal with too much seismic and standard of care is a little relaxed.

I agree with Rapt their recommendation on drop panels isn't great and doesn't touch on the impact to the neutral axis depth change at all which can be significant, this is kind of par for the course in my area. To be frank the Kline article is very simplistic and I'd guess a lot of the recommendations break down when you get into slabs with irregular column layouts or seismic design category C or above. For folds they really only address the path of the tendon and making sure it is contained in the concrete. To really check a slab fold I'm of the opinion that you'd need to do a strut and tie analysis of the fold region which doesn't exactly align with creating a typical fold detail, maybe look at a parametric study of various conditions and see what is required to satisfy strut-tie criteria.

My Personal Open Source Structural Applications:

https://github.com/buddyd16/Structural-Engineering

Open Source Structural GitHub Group:

https://github.com/open-struct-engineer

 

[KootK]

Not sure why you think the harped profile would be conservative Kootk.

Let's see...

1) Because my faculties for reason and intuition lead me to that conclusion and;

2) Thus far, no one's tabled any logic or counterexamples to convince me otherwise.

3) My intent was to simply express my opinion that:

a) It may be prudent, if conservative, to provide the capacity to resist some degree of slab unfolding tendency even if it does not exist theoretically.

b) In many cases, the OP's suggestion of a "draped" model would serve such a purpose. I've provided one example below.

It was not my intent to suggest that a draped model would be the universally conservative answer to be applied in all possible situations without further thought.