Stress Limits for PT Transfer Girders

[dengebre]

We are designing a concrete high-rise building that includes post-tensioned transfer girders supporting multiple levels. Our standard is to design PT beams with a maximum precompression of 350 psi and as Class U (ft <= 7.5sqrt[f'c]). Is this too conservative for a transfer girder? In particular, we are researching the following as it pertains to transfer girders:

1) Is it reasonable to increase the maximum precompression beyond 350 psi but less than 500 psi? We are concerned about restraint cracking at the columns.
2) Does it make sense to design transfer girders as Class T in order to reduce the beam depth?

Thank you in advance for your help and expertise.

 

[Ingenuity]

It is always interesting that as soon as you place 1 kip of prestress to a beam you have all these extra "service stress checks" to undertake pertaining to flexural tension. For a RC beam we don't check flexural tension stresses. What ever happened to partial prestressing and a unified code? I digress!

I have exceeded 500 psi of P/A and 7.5√f'[sub]c[/sub] for flexural tension on several PT transfer beams, however, we used BONDED tendons to the beam, and took special care and detailing of axial restraint. The stressing was staged too, based upon the incremental loadings.

I assume the project is in the US and therefore prefer to use UNbonded PT to the transfer beam - and if so, IMO, you have to look at your details carefully, especially the quantity of mild-steel bonded reinforcement to the beam for both crack-width control and maintaining a reserve of flexural moment capacity, and also your mild-steel reinforcement to adjacent elements to control restraint-effects.

FYI - some codes (like AU code, but not ACI-318) do NOT have prescriptive limits on P/A or flexural tension service stresses, but do have crack control checks for flexure etc.

 

[slickdeals]

There is no reason to design it as a Class U member. The code doesn't restrict you to Class U, unless you are a two way slab. We have designed many transfer girders using Class T and not had any issues.

I echo Mark's sentiments.

 

[dengebre]

Ingenuity, you are correct that this is a US building project using unbonded tendons. Many thanks for the replies.