Steel Self-Stressing Form - Analysis & Design

[WpgKarl]

I am looking at an existing form set-up for a steel self-stressing form in a precast operation. They want to stress 4-0.5" diameter strands, which will react against the steel form at an eccentricity from the centroid of the form.

Would appreciate any help from the steel design experts out there (see attached sketch).

PS - I can check the section as a beam column, not sure that lateral torsional buckling could occur, since the bottom of the box-section is continuously welded to a 4 ft wide steel plate - can't see the compression flange at the top buckling laterally in my case?

Also, I think the ends of the form should be tied down - want to determine the req'd hold-down force. Is it possible to analyze the section assuming the ends are not tied down?

PS - They already stressed strands and cast several times, now we are engineering it before they can continue using it for obvious safety reasons....

 

[InDepth]

This sounds like a very interesting problem. You may want to call the 800 number at the precast-prestressed institute to see if they have any resources available.

You are on the right track with your diagrams and methods. I think you need to tie down the ends in order to limit a loss in prestress in the cable due to rotation and deflection of the ends of the form (or make sure the form stiffness is sufficient to require no tie downs)....but once you pour the concrete the weight may "restress" the cable (another load case). You assumed a single pin near the center but you should also look at the I beams below as vertical load supports (for concrete weight).

Looking at the plate sizes and the number of strands (load) the form looks in working order.

I would also make sure the I beams below the form are tied down/anchored so they down't roll over when the form is moved or "bumped".

In general I think you need a stiff form that mainatins the tolerances specified in the project.

If you are concerened about the previously cast beams, you may want to load test them before sending them out as a QA/QC procedure.

For safety reasons you may also want to assume an appropriate factor of safety that accounts for repetitive loading (cycling) if you plan on cast hundreds...and sizing the welds and detailing the joints to prevent crack initiation and propagation.

 

[desertfox]

Hi WpgKarl

Firstly its a very interesting problem but I would like to ask why if it as been used several times for stressing and subsquent casting without problems what is your concern with regard to the frame? The reason I ask is that if anything is going to fail I would imagine it would be the 1/2" stressing rods that would be the first to go, long before anything on the frame or at least thats how it appears from your diagrams and photo's.
I can't read all the figures on your hand sketch so I am not sure how much tension your putting on the rods or how the tension is applied to the rods, I think your model of an offset loaded column is probably about right,if all the load from the stressing bars is transferred to the box sections at each end, I think I would ignore the central plate and analyse the box sections as you have indicated in your post if this turns out to be safe then the design is conservative as the central plate will stiffen the frame up when added.
The four supports that the box section rests on will mean you have a statically indeterminate beam when you come to analyse the bending moments and deflections as there are more unknowns than equations of equilibrium but its not unsolvable by any means.
If you need a accurate analysis then I suggest you go and use FE.

 

[WpgKarl]

Thx all for the replies....I agree, to do it properly you should model the box beams with multiple supports and moments applied at the ends, plus axial load.

I guess I was looking at trying to keep things simple (statically determinate) so I know what the forces are and where they are going....

PS - we want to engineer this thing to be safe, I always get the "we've never had a problem before" statement, to which I reply - do you want to go home alive today? The FOS could be 1.01 or it could be 1.5 or 2 - that's the difference between a safe design and living by the seat of your pants...

 

[desertfox]

Hi WpgKarl

If the thing as been used several times, if it was going to go wrong due to inept design it would have gone by now surely.
I am not saying its wrong to analyse it but the weakest part to me are the stressing rods.

regards

desertfox

 

[WpgKarl]

Thx desertfox - I agree, I'm just trying to make this thing work, with minimal changes...

 

[desertfox]

Hi WpgKarl

So whats wrong with it, why does it not work as it should? I assumed it worked okay and you were just number crunching.
The one thing I did think was you may have trouble getting the correct tension in your device given the diameter and length of the stressing bars, but once you said it was working okay I dismissed it.

regards

desertfox