Any PT EXPERT here?? 1

[shp6]

Hi:

I have some questions for unbonded post-tensioning construction design.

1. For an unbonded post-tensioning strand in a mutiple span beam or slab, can we use the average effective force in strand after all losses or we have to use the effective force on the furthest point from stressing end? Is there any code indicates what effective force shall be use, average or the least one?

2. I am using ADAPT to do PT beam and slab design. Currently, I found the shear strength calculations of beams in ADAPT has a huge "mistake". It is found that the distance " d " (output sheet Section 12) used in ADAPT 6.11 for concrete shear strength calculations varies below 0.8h, which is a lower bound specified in ACI318-99 Section 11.4.2.3. This mistake actually requires more shear reinforcement than other programs which violates ACI code and also makes the design more costly. Do you have this same problem?

Thank you for your attention.

 

[ishvaaag]

Expert I am not but...

PTI manual example 2 that is unbonded tendons T beam definitely takes unto account friction losses directly considering different values for different points, so average would not be permissible for major elements.

For quite thin slabs I have not so firm opinion, but as long not exceptionally long tendons are used I would say use of average would be acceptable. It would be wise to prestress even and odd tendons from opposite ends, even masons of the past did this for wood pole beams, putting thin and thick ends alternately.

Respect using the real mechanical tendon arm when less than 0.8 h I don't see it being an error but when using some closed form accountancy of the strength against shear that specifies to proceed stating such limit. If it uses the formulation of the code, but does not acknowledge something of what therein specified, you may want correct it manually now and suggest the author make a patch revision allowing the 2 assumptions.

 

[Taro]

shp6,
1. In general, losses due to friction, etc. should be considered. They may not make much difference though, depending on the length of the pull and whether it's a one-ended or double-ended stressing. After you've done a few runs, you will be able to make the judgment call about whether it's necessary in your case.

2.I'm not certain about how the software is using the "d" value you refer to, but it might be that it is performing the shear calculation per ACI section 11.4.1. Two different "d" definitions are used in this equation. One is the actual distance to the prestressing tendons and one is the effective distance that is bounded by 0.8h.

ishvaaag,
The field crews are going to be saying all kinds of nasty things about you if you require stressing every other tendon from opposite ends. That's going to create a lot of extra work in inspection and calibrating jack gauges.

 

[shp6]

Thanks...ishvaaag and Taro. These information you provided are very valuable to me.

For the effective force, I have a problem with my job. Due to the layout of the structure (parking garage), I can only do one-ended pull for a length of 240 ft over four spans. The PT system is unbonded greased and wraped single strands. I did the friction calculations and the effective force varies from 22 kips/strand far end) to 27 kips/strand (stressing end) and the average force is over 25 kips. Engineer said I have to use 22 kips/strand to determine the number of strands but I would like to use the average effective force of 25 kips/strand. What do you think about it??? Is there any code prevision indicates what effective shall I use?? Thanks a lot.

 

[Ingenuity]

Dear "shp6",

1. For multi span beams it is usual to use effective prestress forces at each midspan and at each support location, depending if double end stressed, tendon length etc. For slabs, most US engineers use the average effective prestress at mid-length of the tendon. For unbonded slabs this is okay becasue the profiles are small and the unbonded tendon tends to average out the friction losses over its length.

BUT, i prefer to use the estimated effective prestress at each location i am checking. i use software that models the tendons, steps, haunches, etc and calculates effective prestress forces at 1/15 of each span. so when checking serviceability stresses the applicable effective prestress force is used to check tension and compression stresses.

on ADAPT you can use FELT to check for losses and get a value for effective prestress along the tendon length. only problem is that you cannot use this force back in ADAPT 6 to check stresses etc!!!

one problem i see with US PT detailing is that the engineer details an effective prestress force (say 26k per strand) and requires calcs and shop drawings by the PT contractor to prove that you have such a force. BUT where is the 26k per strand measured? at mid length, at one end? it does make a difference - too much or too little prestress can be detrimental to serviceability, but won't really effect strength.

2. Seems like ADAPT have neglected to have an "if" statement in their programming. again, i do NOT use ADAPT, and when i checked my software i don't have your problem. contact ADAPT and see what they say.

i am thinking that this "error" will give you some VERY large shear reinforcement, most especially for web-shear cracking reinforcement, eg at end span of an exterior support, because the tendon will be at the centroid of the gross section but you still have the beam depth that is providing capacity - part of the reason why the 0.8 rule applies.

ADAPT have a online discussion forum where you can post errors/bugs like this.

Good luck.

HTH

 

[Ingenuity]

Dear "shp6"

It is NOT recommneded to single end stress 240 feet!!!

You need to think about how this will be constructed. The contractor is unlikely to build it without a joint in this direction. Fromwork and concrete supply.finish will come into play with the contractor and he/she will want joints.

Generally, you have a live end about every 120 feet +/-. Double end stressing is more complicated, and as you said may not work with access, i guess. But an intermediate CJ needs to be incorporated.

Also, check you effective prestress force calculations.... for 4x60' spans with say 30" to 36" deep beams you will only have about 18 kips per strand at the dead end!!!

HTH

 

[Ingenuity]

Taro,

Just curious, why do you feel that alternating stressing will require extra "calibrating of jack gauges"?

I agree that it will be a bit of a pain for the stressing crew and maybe the inspector, but not sure that you have to re-calibrate jack gauges any more than same-end stressing.

 

[shp6]

Hi guys...I really appreciate your input to my question.

The problem now is I really have a 240 ft long four-span strand need to be stressed without CJ. For a 16"x39" beam, the effective force at dead end is 20 kips (using FELT) and 27.2 kips at stressing end, resulting in an average effective force of 25.05 kips per strand. As a PT supplier, we would like to use the average force but the engineer wants us to use the least force for determining the number of strands. Now, I would like to know is there any CODE or Recommendastion indicates that shall we use the "least" force or the "average" force?? By the way, I have to stress both beams and slabs in that direction.

If we have to use the "least" force, we may end up with using an intermediate "stressing pocket" or other method to increase the strand force. The challenge is how to make this stressing pocket to be " fully enculpsolated".

Thanks for your input and welcome more discussion to improve our knowledge in PT design, which is a course hardly can be learned in school.

 

[Taro]

shp6,
If you really cannot pull from both ends or use an intermediate stressing joint, using stressing pockets like you mentioned is a good alternative. It shouldn't affect your encapsulation requirements, simply add a few extra tendons for the end spans that terminate at the stressing pockets.

Ingenuity,
When the ram is moved to stress tendons in a different location, the gauge is required to be recalibrated. This involves the contractor and the inspector independently measuring elongations for several tendons and verifying that the jack is correct. It really slows down the operation, so it's better if you can pull all the tendons in one direction from the same end. Maybe it's not a universal requirement, but it has been an issue for most of the jurisdictions that I work with. Of course, I live on the west coast where building departments and inspectors seem to be much more particular than in many other parts of the U.S.

 

[ishvaaag]

I really can add little to the much posted.

For a 240 feet single strand I would try to use the evaluated valued at every point even if nonbonded in flat plate.

A paliative could be increasing sagittas in the spans where tension is less, but this might not be feasible.

Respect stressing from 2 sides it could mean 1 recalibration more, no? Taro, in any case I am sure many nasty things are being said of me already, and this would be for a good cause.

 

[Ingenuity]

Shp6,

I think FELT uses PT loss calculations from the paper by ZIA et al (1979) which does UNDER-estimate losses. ACI209 data is better for long-term losses, but for your case it would result in an effective pretress force in the end span of 18 kips - probably not want you want to hear. As a PT supplier you are stuck between a rock and a hard place regarding "what force does the engineer want - least or average". Occasionally, i see engineers detail both forces and # of strands (beams), or spacing (slabs), and this makes more sense because the number of strands (or spacing) and required to do untilate strength calcs anyway.

Unfortunately, there appears to be no recognised standard on whether the force is least or average.

Did the engineer show both ends-stressed on the 4x60' span beams for this project?


Taro,

The PTI requires that rams be calibrated with 3 incremental runs, at 3 different ram locations over the ram stroke, for the capacity of the ram. The average of the 3 runs is used to produce the calibration graph.

Under the logic of your US west coast inspector, if you had one slab edge that was not square (at say 30 degree angle) resulting in all slab tendons being of different length, then you would have to recalibrate for each tendon!!!!

I do not think so.....seems like the inspectors need to take a few lessons in stressing, and calibration and understand the "error" that is involved with elongations (calculated vs theroetical).