PT elongations - widespread deviations 2

[mmodfr2013]

I've got a project in construction, a concrete podium supporting five levels of timber above. 16" thick flat plate concrete podium, about 190'x60' in plan. The contractor pulled the tendons and basically every one came up to pressure and was under the design elongation, only a handful (out of ~500) were inside of 7% short. All of them were around 15% to 25% under the design elongation, banded lines are 190' and pulled from both ends and distributed tendons are about 60' and single pulls. There are also some shorter ~30' tendons. They have performed liftoffs on 10 randomly selected tendons using a different jack and didn't see more than nominal additional delta.

Has anyone seen anything like this? The PT engineer's response has been that the evidence points to the appropriate PT force being transferred except for the elongations. My leading position at this point is to detension some percentage of tendons and re-pull them to verify the PT force. But I don't think much will come out different unless we alter the initial assumptions for friction/wobble. I can't find much in the way of PTI/ACI code recommendations or requirements, or precedent for a situation when every tendon is out of spec.

 

[IDS]

The PT engineer's response has been that the evidence points to the appropriate PT force being transferred except for the elongations.

Isn't that like saying that the evidence points to the appropriate concrete strength, except for the cylinder test results?

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[rapt]

With unbonded PT, if they are that low, then the appropriate PT force has only been supplied near the anchorages. As you move away from the anchorages, the tendon forces are lower than required by design. At the mid-length of the double end stressed tendons, if the extensions are 25% low, the tendon force available will be 50% of the design value.

Unbonded PT design is much more dependent on the force in the tendons after stressing than bonded PT design. The design needs to be redone using higher friction factors to give elongations equivalent to what has been achieved on site. This will then give the correct PT design forces and a significantly reduced ultimate capacity.

 

[Ingenuity]

Did the PT detailer who placed the calculated elongations on the PT shop drawings simply use 0.08"/ft (i.e. 8" per 100 ft) to arrive at the stated expected elongations?

Calculated elongations should be based upon the actual vertical drapes and horizontal sweeps, with allowance for wobble, using the appropriate friction values.

And extending from what rapt stated, did the PT engineer/designer use the 'average effective force' in his/her design calculations, or use the more accurate variable effective prestress?

Don't go repeating lift-off tests and re-stresses to the same tendon/s - broken wires (or full strand failure) may result.

 

[rapt]

Thanks Ingenuity, I forgot to mention the US predilection for using average prestress force over the full tendon length. Even average in each span should not be used, especially where the friction loss is high or the tendons is very long.

 

[mmodfr2013]

The PT designer used ADAPT to calculate the elongations, and accounted for the tendons drape/sweep/etc, and used the variable tendon forces. My guess would be that the friction/wobble factors were underestimated, and so my effective PT forces are lower than what they should be.

What I was interested in was if anyone had ever had a similar situation come up, to try and find any other possible causes for all of the pulls to come up short.

 

[Ingenuity]

I have had a few (very small percentage of projects) where the tendon elongations for say a full floor have been less than calculated, and less than other previous floors/areas.

I have typically checked the mill certs and check if the coil of strand was form a different heat number, then I 'played' with varying the elastic modulus - sometime it can be a bit higher than any default/typical value. Couple this with additional friction and you can usually get the re-calculated elongation close to the measured elongations. I then go back to my PT calcs and see what effect it has to the design.

Here are two reference for future use:

1. The Stressing Report: by Jim Rodgers from 2011 CI Magazine: Link

2. Elongation Tolerance for Short Tendons in Post-Tensioned Building Structures: by Carol Hayek and Thomas Kang from ACI Structural Journal from July 2017: Link​