Deflection due to creep 2

[ozziz]

Can anyone help me on this question.

Based on monitoring & observation, it was found that when a pre-tensioned beam upon releasing the strands, it will hog due to uneven prestessing force. It will continue to hog upwards due to the sustain stressing acting upon it over a period of time over 60 days. According to my calculations, over the period of time, the hog will drop due to shrinkage and creep of creep and steel strands.

Maybe I am using the wrong formula. Has anyone experience continuous hogging of beam after transfer and which formula to use. Thanks

 

[UcfSE]

What is the source of the question? Who did the monitoring and observation? For what type of application are you checking this? Homework?

 

[ozziz]

I am designing a pretensioned bridge girder and was told by a Senior that on the site that they found that the hog (deflect upwards) of the beam increases at (after) transfer 3 days steam cure as the concrete has yet to be fully mature. They found the hog increases at 5, 7, 21 days concrete strength.

I need to determine the amount of hog as they have implications on the thickness of the insitu concrete slab.

According to my calculation, the hog decreases as compare to the actual site situations. Maybe I use the wrong formula. Any advice? Thanks

 

[jhardy1]

You have several phenomena happening simultaneously:

The initial elastic deflection due to prestressing force is upward; gravity loading is downward; the initial combined elastic effect when the strands are released is for the beam to hog upward.

Shrinkage in the concrete tends to reduce the axial strain in the prestress strands, so will tend to cause the tendon force to drop, and the beam will sag (hog will diminish). Shrinkage rate is a function of age of the beam at transfer, quality and type of curing, ambient temperature and humidity, and section profile (thinner sections with large exposed surface area can shrink faster than thicker sections).

Creep in the concrete is analogous to reducing its elastic modulus over time. If all other effects are unchanged, the net upward load on the unloaded beam due to prestress is greater than the downward load due to gravity, so the net effect of creep considered in isolation is for the beam to hog upward even more over time. Creep rate is primarily a function of age of the beam at transfer, quality and duration of curing, and the stress in the concrete.

Relaxation of the strands tends to cause the prestress force to drop, so the beam will tend to deflect downward (hog will reduce). Relaxation rate is a function of the quality of the strand material employed and the stress in the strand.

All of these effects are occurring simultaneously, at different rates. It is conceivable that your beam could hog upward or downward over time, depending on the interaction of these effects.

 

[ozziz]

Julian,

Thanks for your explanation as we found this happen when the beam is release from the mould. IT CONTINUE to hog UPWARDS. We have to calculate how much hogging UPWARDS. But my calculations show the hogging is reducing (sag)due to creep over time (loss of prestressing force).

Is there any references or method of calculation for computation of creep hogging? Thanks

 

[jhardy1]

ozziz,

Any standard reference book, lecture notes, or code on concrete design will go over the principles of time-dependent deflection. Alternatively, try a Google search for "creep deflection of concrete" - you will turn up literally thousands of hits.

Like I said, I suspect you may be confusing "creep" (of concrete) and "relaxation" (of the strands).

To summarise again:

Initial elastic deflection upon release of the strands is upward(for a typical simply supported beam, with tendons biased toward the bottom).

Time-dependent deflection due to concrete shrinkage is likely to be downward.

Time-dependent deflection due to concrete creep for the unloaded beam is further upward. (When additional dead and live loads are imposed, overcoming the prestress forces, long-term creep deflections could be downward, but upward is more likely in the short-term for the bare beam.)

Time-dependent deflection due to steel relaxation is downward.

The magnitude and rate of all time-dependent deflections depends on several factors. Total time-dependent deflection could be upward or downward, depending on all of these parameters.

 

[ozziz]

Julian,
The pretension beam will be steam cured. So it is correct to assume that relaxation will occurs immediately after transfer?

I did account for creep at different period of time ie at 3 days and 60 days and long term which I assume loss of prestress force in the strands.

Anyway, the topic to look for guidance is time-dependent deflection in creep? This is the tropic which I neglected in my Studies as they are very microscopic in subject and not sure if the authors from America, Europe and Australia have different approach eg calculation of crack width which is not in American Code.

At present I have some material from Gilbert (Australia) - maybe Ghali.