p.t. concrete 1

[ajk1]

Given:
Existing 2 span post-tensioned (p.t.) garage floor beams with corroding unbonded tendons in the old push-thru sheaths, about 35 years old.
Can not reliably replace the tendons because of various reasons (such as the "pinching effect" where transvers p.t. girders cross, criss-crossing of tendons in the beam, etc. We tried over the years!

Required:

Assume all the existing p.t. eventually fails.

Question:

Can externally installed Dywidg straight bars, 2 near the top of each 2-span beam (just below the slab soffit) and 2 near the bottom of each beam (just above the beam soffit) that are then prestressed to apply the same magnitude of exial force as the existing p.t. in the tendons, be acceptable, if calculations show that the new Dywidag bars apply enough force to keep the entire T-beam section (slab and beam) in compression?

 

[rapt]

There is more to PT than applying compression. You have left out a lot of detail on how these bars are going to be connected etc and the deflected condition of the beams when they are added all of which will determine if such a solution will work.

I would prefer external draped Dywidag bars with saddles at the supports at the top and at about 1/3span points in the bottom.

Then you have to allow for fire rating, corrossion protection etc.

 

[ajk1]

Yes I left out the details because if the concept is not valid, then there seems little point in going into the details. I am aware of the details and how to deal with them. My question pertains to the "concept" of whether post-tensioning with axial force only, with sufficient force that assures there will be no tensile stress anywhere in the concrete, is a valid concept. I think that it is probably not a valid concept, although I am not exactly sure why, other than it does not meet Code requirements for minimum rebar etc., would be very unusual, and may lack ductility too, as well as if the prestress force relaxes the whole beam may collapse with no advance warning. That is the answer that I thought someone was going to tell me.

 

[KootK]

I think that it is probably not a valid concept

I wouldn't be surprised if the concept turned out to have legs pending some number crunching. Pre-stressing confers a number of structural benefits, as you know, and the only major one that you'd seem to be missing out on with this approach would be drape. Often, when I see the realities of tendon profiling in thin slabs, I wonder if we might be better off simplifying slab post tensionioning to just straight cables at mid-depth. Granted, the situation is a bit different with beams.

As far as fire-proofing and ductility go, these thing must be surmountable as you certainly wouldn't be the first to go the external post tensioning route.

The exercise is a good deal more complex than just a one to one replacement of prestressing force but, then, I don't think that you were really suggesting that it would be that simple to begin with.

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.

 

[ajk1]

to Kooyk:

Thank you for your thoughtful reply, as usual.

One of the senior principals in our firm many decades ago (he is no longer alive), suggested that for slabs the tendons just be straight. That was in the context of not having low points in the tendon where water could collect and corrode the strand (the "push-thru" type of sheath was used at that time, with annular space around the strand).

As for my current issue, it is interesting that you think that it may have legs. I will have to give it more thought.

If anyone else has a comment, I would be interested. BAReetired used to be good at this basic structural theory sort of stuff, but I have not seen him here for a while. I will try to make the free-body diagram and see where I get with it. Possible running it as a column with lateral load, might also tell me something.

 

[rapt]

Kootk,

If you worked out the cost penalty, you would not suggest it.

ajk1,

You would need to jack the beam back to its original condition when it was on formwork and still need to provide saddles at appropriate locations to make the tendons effective. Plus it would require a LOT more prestress.

Connecting at just the ends for external tendons would be useless.

Hence my suggestion of draped tendon to high and low saddles. This would be much more effective and would work like externally prestressed bridges.

 

[KootK]

If you worked out the cost penalty, you would not suggest it.

Don't be coy then, what's the cost penalty?

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,

I thought you would be able to figure it out yourself using some very basic design logic.

With draped tendons the effective depth of the tendons at critical sections is D - cover, so in a 200mm thick slab probably about 160mm.

In your scenario, it is an average of 100mm in each direction, about 110 in one direction and 90 in the other.

So you need at least 60% more prestress or reinforcement to replace it. Then you could have major ductility problems with such small effective depths and you have not solved the introduced deflection problem and increased crack control problems. If you are using the ACI flat slab logic with banded/distributed tendons you also have the stress limits at service and partial prestress design is not allowed to solve these. More prestress is required again.

 

[Ingenuity]

Can externally installed Dywidg straight bars, 2 near the top of each 2-span beam (just below the slab soffit) and 2 near the bottom of each beam (just above the beam soffit) that are then prestressed to apply the same magnitude of exial force as the existing p.t. in the tendons, be acceptable, if calculations show that the new Dywidag bars apply enough force to keep the entire T-beam section (slab and beam) in compression?

Why use external PT bar when you could simply use external PT strand, and drape/harp it like has been done many times in the past?

Keep it simple...therefore more competitive...quicker to construct...and therefore less expensive...without compromising durability.

 

[KootK]

I thought you would be able to figure it out yourself using some very basic design logic.

You thought right. I could have figured it out myself if I felt that the problem were as simple as estimating efficiency loss from a material quantity perspective alone. I feel that it's a more nuanced problem including an accurate accounting of the following:

1) A straight tendon arrangement simplifies both design and construction. And those simplifications embody cost savings that need to be weighed against any material quantity inefficiencies. Any answer that doesn't include this is no meaningful answer at all.

2) Most modern slabs are governed by serviceability rather than strength. I haven't worked through the implications in detail but I suspect that a flat tendon layout would be less detrimental to serviceability than it would be to strength. With the right level of prestress, you're still getting to use lots of uncracked slab stiffness which is a big deal.

3) When I perform field reviews on modern, thin PT slabs with irregular column layouts, I can't help but get the sense that heavy reliance on meaningful drape is mostly just an academic issue. Technophiles stuck behind computers all day, fooling themselves into disregarding realistic construction tolerances and thinking that slivers of drape are having a big impact.

I very much suspect that a flat tendon layout would prove economically disadvantageous. That, for the simple fact that PT has been around longer than I have and I can't possibly be the first to think these thoughts. If flat was better, I'm sure that's what we'd be doing.

 

[rapt]

Kootk,

You should do the calculations and the construction sometime to get a feel for it all.

I have been doing it for over 40 years, the first 15 of them without design software (until I had RAPT working) and the first 10 even doing analysis without computers.

1) Wrong on design and not a large effect on construction with a properly trained crew.

2)Just looked at the results for a flat tendon design versus a draped design using the same number of tendons
- untensioned reinforcement increased by over a factor of 2 nearing it to an RC only design
- uncracked section stresses at service increased by over a factor of 2 and steel stress at service increased from about 70MPa to 265MPa, similar to what you would have for an RC design
- deflections doubled to nearly the same value as for an RC design on the same member using the same concrete depth.

It would have taken me over a day to do the above design assessment of 3 models for the same slab to a much lower degree of accuracy 40 years ago. Having the software and the understanding of design and construction allows me to do it all much more quickly and accurately.
Agreed, people using the software should have the understanding and experience first. That is a problem with the education system and the industry, not the software.

I obviously agree entirely with your last sentence.

 

[KootK]

You should do the calculations and the construction sometime to get a feel for it all.

I do PT calculations and field review on a regular basis and do have a feel for it. It's presumptuous of you to assume things about me that you do not know.

Wrong on design

Not wrong on design, different on design. Sometimes folks like to explore new ideas. Folks like me... and Eugene Freyssinet.

not a large effect on construction with a properly trained crew.

If you're unable to quantify this, then your answer does not satisfy my curiosity about the concept in any meaningful way. That a theoretical analysis of a slab with no drape would yield poorer performance than a slab with drape is, of course, a surprise to no one.

 

[rapt]

Koottk

Nothing presumptuous at all. Just replying the information in your points.

 

[KootK]

Nothing presumptuous at all. Just replying the information in your points.

You suggested that I obtain more experience with the calculation and construction of post-tensioned slabs. To say that is to presume that I currently lack such experience, which I do not.

 

[ajk1]

to Ingenuity;

You say external p.t. has been done many times in the past. I don't doubt that is true, perhaps more in Europe, but in my area I have been able so far to identify only once where it was done, and that was 20 years ago for an arched structure (the added external p.t. was straight bars) , which is not my case.

I am curious as to what area of the world you are located? Northern U.S. or Canada?
Approximately how many times has it been done, of which you are aware? <3 times? Or >10 times?
How did it work out?
Did they have to destress some of the existing tendons in order not to have too much prestressing uplift when they added the new external draped p.t.?
Maybe you can suggest someone who has done it, that I can contact. It seems a formidable thing to carry out on site, with the required standoffs from the existing beam soffits, drilling thru the supporting girders to get the tendons thru to the other side to place an end anchor, how to accommodate the tendon curvature where it goes thru the drilled holes on the supporting girders, placing the draped strand inside a duct to be grouted, etc., but perhaps I am overthinking it.
One of the local p.t. contractors suggested we do it with straight p.t. bars, but I don't see how we can do that and still get uplift to reduce deflection.

 

[rapt]

Kootk

Quote Kootk
Most modern slabs are governed by serviceability rather than strength.

Maybe in North America with unbonded PT and in particular average moment design to ACI. Not in the rest of the world where we think we are also modern and are able to do Partial Prestress as we consider column/middle strip effects. We stopped using the design for service and check ultimate logic in the mid 1970's except for UK code based countries where it took until around 2010.

Quote Kootk
I haven't worked through the implications in detail but I suspect that a flat tendon layout would be less detrimental to serviceability than it would be to strength. With the right level of prestress, you're still getting to use lots of uncracked slab stiffness which is a big deal.


I suggested you work through the implications in detail to get a feel for it.

Quots Kootk
3) When I perform field reviews

A field review is not managing or performing the construction tasks.

 

[KootK]

drilling thru the supporting girders to get the tendons thru to the other side to place an end anchor, how to accommodate the tendon curvature where it goes thru the drilled holes on the supporting girders

Would it be possible to develop a solution that only involves bottom side post-tensioning? If that could be tolerated from a serviceability perspective etc, it would certainly aid constructabilty.

One of the local p.t. contractors suggested we do it with straight p.t. bars, but I don't see how we can do that and still get uplift to reduce deflection.

If you omit the countering, top side post-tensioning, the straight bars will give you uplift. In fact, per unit prestress, I believe that bottom side straight bars anchored at the ends would give you the maximum amount of uplift (think area under the moment diagram associated with the prestress). This may well cause other problems, however, such as flexural overstress at the member ends.

 

[Ingenuity]

You say external p.t. has been done many times in the past. I don't doubt that is true, perhaps more in Europe, but in my area I have been able so far to identify only once where it was done, and that was 20 years ago for an arched structure (the added external p.t. was straight bars) , which is not my case.

I have no experience in Europe, and all of my external PT projects have been North America-based. I would think that Europe would have done a bunch of external PT to bridges, but total guess on my behalf, maybe they do it to building on grand scale too.

Here are some Google images of two California projects (not by me - I know little about them - just some quick Google-fu got these images):

This one is probably going to get partially encased in concrete:

This one has a truss-analogy:

I am curious as to what area of the world you are located? Northern U.S. or Canada?

My experience and area of practice is/has been AU, US and CA.

Approximately how many times has it been done, of which you are aware? <3 times? Or >10 times?
How did it work out?

I have personally executed (design and build) 5 external PT projects. Largest was 8 x 31 @ 1/2" dia strand tendons, each tendon was 1,000 feet long. Smallest was about 50 feet long where we took a 2 span RC beam and removed the center column and used harped tendons to create a clear span. Another project was due to severe unbonded PT corrosion and we installed 300+ feet long tendons, and the 4th project was a fix of a construction error - where the bonded PT to a transfer beam was omitted (long story) and we used external tendons on the side of the beam, BUT we encased it with concrete for fire and aesthetics. 5th project was a podium deck with a PT flat plate with severe corrosion to the existing tendons (hundreds of failed tendons) and we used 2-way external PT to strengthen the slab then apply vermiculite fire proofing to the tendons.

I would estimate that there are many hundreds of external PT projects in North America, dating back to early 80's, with many in the western US (Califormia, Arizona, Nevada areas). Ken Bondy from LA area, was somewhat prolific in "EPT" (external PT) as he termed it.

Specialized companies like Structural Group (VSL licensee), Schwager-Davis, and off course Dywidag and Freyssinet undertake external PT. Dirk Bondy (Ken's son) and his partner, Bryan Allred, of Seneca Structural design/build external PT projects in southern California (not a huge market in this scope), but they probably do 2 or 3 projects a year, mostly to parking structures.

I know that the once Canadian developer, TrizecHahn purchased a Washington DC project in the early 2000's (maybe late 90's) and shortly after purchase discovered severe corrosion damage to existing PT so they used external PT to strengthen the complete ground floor level - I visited the project in 2001 and they used massive tendons in two directions to strengthen the flat slab. The work was done my Structural Group and the engineer was HALSALL of Canada.

Did they have to destress some of the existing tendons in order not to have too much prestressing uplift when they added the new external draped p.t.?

No. Actually, that is the beauty of external PT. You can monitor and adjust your prestress forces in the field to 'react' to how the structure is behaving from a deflection perspective. Typical the P/A is low enough that axial compression is not an issue.

On the PT project where we had corroded tendons and we used 300+ feet long external tendons, I left 3' stressing tails at each end so that we could go back and re-stress, if required, at a later date. We used a long galvanized steel tube (grease-filled) over the strand tails to provide corrosion protection.

Usually the challenges on these projects are the end anchorages - making them sufficiently robust, yet simple to install and enable ease of stressing. And the next challenge is inexpensive fire protection.

Maybe you can suggest someone who has done it, that I can contact. It seems a formidable thing to carry out on site, with the required standoffs from the existing beam soffits, drilling thru the supporting girders to get the tendons thru to the other side to place an end anchor, how to accommodate the tendon curvature where it goes thru the drilled holes on the supporting girders, placing the draped strand inside a duct to be grouted, etc., but perhaps I am overthinking it.

Contact any of the major specialized PT subs - VSL, Dywidag, Freyssinet. Or for a consulting engineer-to-consulting engineer conversation maybe reach out to Dirk Bondy or Bryan Allred of Seneca Structural in So Cal.

One of the local p.t. contractors suggested we do it with straight p.t. bars, but I don't see how we can do that and still get uplift to reduce deflection.

Unless they can demonstrate the benefits (both technical and cost/schedule) I would dismiss it. They won't be able to get any uplift calcs to work, because there won't be any uplift, with equal T & B PT bar tendons.



When a project like this comes along with such a challenge you usually try and select 3 or 4 options (as you have no doubt considered) and the first option I would consider for a project like this would be external PT using draped/harped strand tendons. I have personally cored the holes, installed the tendons and deviators, stressed the tendons, broken tendons, screwed-up my fingers etc - and whilst it is specialized work it is not difficult with the right experienced crew, and it if has many beams to strengthen, the repetition makes it very attractive.

I can probably dig through my library and get some industry articles about external PT, like case studies of past projects etc. Let me know.

 

[Ingenuity]

...drilling thru the supporting girders to get the tendons thru to the other side to place an end anchor, how to accommodate the tendon curvature where it goes thru the drilled holes on the supporting girders,

Sometimes - subject to calculations - you do not need to go all the way to be back/end of the girder, and can terminate the PT anchorage just on the inside of the girder/column face, with due allowance for stressing access.

The end-anchorage is usually a straight-shot from low point/s to end-anchorage, so just core a hole through the girder at the required angle of the tendon, resulting in no local effects on the girder or on the tendon.

For the internal tendon high-point (over the interior support) you have to accommodate the total tendon angle change from the left and right spans and their next adjacent low points.

After successful GPR scan for any PT/rebar to the subject beam and girder areas - when the drapes/curvatures are small (shallow beams, shorter spans, for example) I have cored holes in the supporting girder and installed a straight steel tube where the front inside edge is machined (over a few inches) to 'transition' the tendon curvature so it does not point-load the edge of the girder. In deep'ish beams, where the drape is more acute, (and the girder is wide) a slightly oversized core hole and a pre-bent (curved) steel tube (to match the desired angle change) can be inserted into the cored holes and the annular space epoxy grouted/injected. I have also cored wider girders, over a half-width, from both sides, at a small angle to horizontal, and then insert the tendon group and upon stressing there will be some concrete crushing at the pinch-point on the inside of the girder at half-width. This latter detail does NOT work well for greased/sheathed strands because it rips (and can destroy) the sheath at the pinch point.

What is your beam and girder widths and depths? The span is 56 ft right?

 

[Ingenuity]

Some Google-fu of external PT experiences/case studies in North America:

Reference 1: From 2009

Parking Structure Damage Scott M. Adan and René W. Luft Concrete International ==> Link​

Reference 2: From 1992

Strengthening Existing Structures With Post-Tensioning Miroslav F. Vejvoda Concrete International ==> Link
​

Reference 3: From 1991

Repair and Retrofit Using External Post-Tensioning Barchas, Karen Concrete Repair Digest ==> Link
​

Reference 4: From 2005

Externally Applied Post-Tensioning Systems Bondy, Dirk Structure Magazine ==> Link​

Reference 5: From 2012

Seismic Retrofitting of Externally Applied Post-Tensioning Tendons Bondy, Ken, ACI Spring Convention 2012 ==> Link​

==> With a BONUS free webinar video presentation here by the author: Link​