About Permissible Stresses in Tendons as per AASHTO LRFD section 5.9.3 1

[saadeh]

Dear fellow bridge engineers,

I'm currently working on a local bridge project which employs precast, post-tensioned girders, and the other day I had an argument with the consultant and my prestressing sub-contractor (Dywidag) about the permissible stresses in stressing strands and it turned out that we might have a problem with about 25 girders regarding the maximum allowable stresses in tendons, my subcontractor field engineer have a poor technical knowledge and so does the consultant !

before jacking, the sub-contractor provides us with his field calculations regarding jacking force, losses, theoretical elongation,...etc.
and for no good reason, I used to forward these calculations to the consultant for approval without reviewing it myself. A wihle ago, we had a problem with one of the girders and due to some site problems, we had to use 14 strands in one of the tendons instead of 15, when we discussed this issue with the consultant, he immediately replied that there's no problem and we can go on and start jacking, later when I asked for a written letter to secure our act of using 14 strands instead of 15 based on our discussion with him, he changed his mind completely and refused to accept the completed girder and told us that the designer checked the girder and the maximum allowable stresses in the tendons are exceeded !!

Now, I'm doing a check on the stresses and going over a redesign of the girder, but I came across section 5.9.3 where it states that the maximum stresses allowed in tendons at anchorage and couplers immediately after anchor set is 0.7Fpu....

My question is, does this mean that the stresses after the instantaneous losses (friction, seating, and elastic shortening)shall not exceed the given limit ? my subcontractor told me that you have only to consider seating losses only when doing this check without taking into account the friction and shortening !! I don't understand why to consider only seating when and at the same time the other two loss types take place !! my local code of practice is a translated version of AASHTO LRFD specifications and the translation is very poor and this check is not clear to me, in the same time the English version states that this check is done (AFTER ANCHOR SET) and my subcontractor understood it literally, but as per my understanding, it should include all instantaneous losses.... I have a copy of Edward Nawy's Prestressed Concrete and it states that this check is done after transfer (which means after all instantaneous losses) so I need another interpretation for this sentence from a prestressed expert and as I said earlier, my subcontractor has poor knowledge..

Another issue, I have three tendons and each tendon is jacked separately, the strands of each tendon are jacked all together, but tendons are jacked separately one after the other, so doesn't this impose elastic shortening losses ?? again, my sub-contractor says no!!

Please Help! and thanks in advance...

 

[rapt]

Yes, the stress check at transfer should be after all immediate losses. Seating loss calculations are dependant on friction anyway so that has to be included. The loss is not assumed to be constant along the length of the tendon, it is affected by friction.

Yes, each tendon that is stressed will be affected by any tendons stressed after it. The first tendon stressed causes no elastic shortening losses min itself as the concrete shortens as it is being stressed. So the first tendon stressed will suffer elastic shortening from all of the later tendons. The last tendon stressed will have no elastic shortening loss. In most cases designers would base this loss on the average of all of the tendons rather than doing a separate calculation for each tendon. It is normally only a small loss compared to other losses.

But the designer is probably worried about the stress in the strands under service loading, not transfer. If there is one less strand than was allowed for in the design, there is about 7% (1/15) less prestress, so the stress at service will be higher and the stress in the strand at service will be higher and may be above the allowed limits for crack control.

 

[saadeh]

Thank you rapt for your explanation, this is what I need to kick some butts

Regarding your last pair of lines, we got around this issue by applying more stress to the 14-strand tendon, so instead of jacking the tendon to 75%Fpu as the design requires, we jacked it to 79.5%Fpu, which reduces the difference you mentioned to about 3% between the applied and the planned stresses. Another point, the Girder is originally over-designed !! When I did a rapid check using Prokon, the girder required only 11 strands per tendon if the original tendon configuration is to be used as is, knowing that the code requirements for allowable stresses of both concrete and strands are not violated !! plus, here in this country where the project is being constructed, the local code implements a special HL-93 live load equivalent to double the loads given in the original AASHTO LRFD Specifications !! for instance, the total truck load for all axles combined is 600kN (lane load is 20kN/m) while originally the loads are 320kN for the truck and 9.3kN/m for lane!! so all in all, the Girder in my opinion is way more than safe...

The designer has very poor technical experience and so does my sub-contractor! The designer told me that the problem is mainly in tendon stresses at jacking because we applied 79.5%Fpu, without doing any checks or mentioning concrete stresses !! imagine that....

Another question if you allow me.... What is the maximum jacking stress that can be applied for post-tensioned members using Low-Lax strands as in this girder?
The same section (5.9.3) allows 0.9Fy = 0.81Fpu (for Low-Lax) but some States DOT's in their local amendments require lower jacking values,,, my local specifications permit 0.9Fy, but so for the sake of knowledge, I would like to hear your point of view on the maximum jacking stress allowed...

Thanks again rapt for your helpful explanation again

 

[rapt]

saadeh

Stressing tom a higher jacking load will probably nor improve your ultimate strength. This is still about 15% below the initial design.

if the design loading required is this HL-93 loading, how can you justify the design on a much lighter loading?

The normal limit would be one you mentioned initially, .7 Fpu after anchoring (after initial losses and draw-in)!

If you are reducing the amount of steel provided and also reducing the required loading, who is taking design responsibility?