Pile cap design Questions

[Berniedog]

We are using RISA Foundation to design a pile cap. The piles are in a radial layout. We want to design the reinforcing in a rectancular grid. We have modeled it and it looks good. We how do we check our results? Any ideas?

 

[Teguci]

Check 2-way punching shear, check 1-way shear, check bending.
Check bearing.
Provide more info?

 

[rowingengineer]

Hand Cals HaHaHaHa.
But seriously, we need more info, because there are over 20 mechanisms of failure that could be apparent in your pile cap, and I don't have all day. If this is a typical pile cap, say 4 piles you could do a strut tie for reo and punching check by hand. but again I want more info.



Arguing with an engineer is like wrestling with a pig in mud. After a while you realize that they like it

 

[DaveAtkins]

I don't think it is funny to check the RISA analysis by hand calcs. I don't know how else to do it. RISA gives you Von Mise stresses, which are meaningless for reinforced concrete design. So you should have your RISA output show the moments and shears, and then check them by hand.

DaveAtkins

 

[Berniedog]

Attached is a layout of the piles. The pile cap is 24" thick and the livie load is 1900 psf.

THe piles are 18" diameter.

 

[rowingengineer]

Dave,
Wasn't meaning to say hand clas were funny, What I was hahaha was the fact that saying hand cals without more info is about as helpful as a hole in the head.

SO Hand cals are required Just to be clear!!! Once we have more info we can be more specific.

Arguing with an engineer is like wrestling with a pig in mud. After a while you realize that they like it

 

[KootK]

That's quite the pile cap. If it's truly uniformly loaded, as I gather from your second post, then you're in luck. Your pile cap should be much easier to check than a typical pile cap supporting concentrated / lateral loads.

Work out two rough continuous beams:

1) A radial beam
2) A circumferential beam

The top and bottom steel that you calculate should roughly match the output from your model.

Also check punching shear and one-way shear based on a tributary area calculation.

 

[JoshPlumSE]

For what it's worth, I believe he is talking about RISAFoundation (not RISA-3D).

If so, then the FEM output for that program is much friendlier for producing concrete design information than just looking at Von Mises stresses.

To hand verify the results, I would take a two or three step approach:
1) Verify the reactions in the piles. A quick and dirty method would be confirming that the sum of reactions = sum of applied loads. Then also confirm that the centroid of applied load = centroid of the reactions.

2) Graphically view the bending moment and shear diagrams in the design strips. Compare this to what you would get from a continous beam model with a similar applied load.


3) After you confirm those numbers are in basic agreement with your hand calc method, then moving from moment and shear demand to reinforcement and code checks should be pretty simple.

The program doesn't do punching shear calculations yet (as it is geared towards soil supported foundations), so you would have to check the punching of the piles through the slab manually.

Josh

 

[hokie66]

I would have set out the internal piles orthogonally, but that makes little difference. Your most critical bending condition is the 4'9" cantilever, so I would like to see that reduced. Then I would just pick a distance between piles, about 8' per your plan, but may need to be increased a bit to cut back the cantilever. Reinforce both ways top and bottom for wl^2/11, keeping the cantilever moment below that value. Check shears. Done.

 

[ishvaaag]

RISA 3D (and many other programs) also give main principal stresses at the faces, and Mx, Mxy etc. These are more useful to provide reinforcements, easy to determine for predominantly flexural situations. Other FEM give the principal stresses in the mass itself of concrete, this can be useful to check the feared compressive strut failure. Once you have the reactions you can also make some checks on what appreciated critical by strut and tie approach but normally this will be inferior in quality to what you can ascertain from stresses derived from an analysis.

 

[ahmedhegazi]

My advise ,try to check one way shear and 2 way shear by hand as Finite Element mesh sometimes provide jump in stress due to meshing.

you can get bending Reinforcement based on finite element stress but again consider that finite element mesh affect the results.

Many times in this problems i do computer model but should study the results and verify it and sometimes i re-distribute the reinforcement as well

 

[steellion]

This pile cap certainly needs to be designed by hand. The CRSI manual has a step-by-step example for designing any pile cap for reinforcement and required thickness.

 

[vandede427]

at only 24" thick with a uniform area load, I would say this needs to be designed more as a two-way slab than a pile cap.

 

[Teguci]

2-way slab design using trapezoidal pieces and column support lines at each circumference of piles.

 

[ATSE]

Mostly agree with JoshPlum.
Think of your problem as an elevated slab, not as a foundation, since you have concentrated pile reaction points (like columns) and not a bed of uniform soil springs.
Use a simplified strip method - coefficients from ACI 318 from direct design method (section 13.6) will usually suffice - to get moments & shears within 15% accuracy. Easier to think in 2-D.
You should do this before you turn on your computer.

 

[abusementpark]

What is that supporting? A liquid tank?

 

[Berniedog]

It supports a municiple water tank.

Thanks all for the input.

 

[vandede427]

if the top of the slab / pile cap will hold the water, use ACI 350.