punching shear of pile caps

[oneintheeye]

Hi, when checking against punching shear at 1.5d zone (according to UK code), but at whatever distance your code states. How do you deal with overlapping zones, say you have 4 piles and all perimeters overlap?

 

[swearingen]

If they overlap a good bit, I'll draw a straight line between the tangents of the 1.5d zone to make a block around the pile group; it will become sort of a rounded corner square (for 4 piles in a square formation).


If you "heard" it on the internet, it's guilty until proven innocent. - DCS

http://www.eng-tips.com/supportus.cfm

 

[hokie66]

I like to be conservative in designing pile caps, so would check the punching shear based on one pile and the design load of the pile, ignoring all others. Don't forget to check punching shear at the supported column.

 

[oneintheeye]

sorry I dont see how that is conservative if the zones overlap?

 

[hokie66]

Look at it this way. If the zones did not overlap, you would check punching shear as normal based on the specified perimeter. If punching shear is inadequate, you increase the depth. At some point, this increase in depth makes the zones overlap. What then, do you check punching shear based on the external perimeter around all the piles? That doesn't make sense.

It seems to me that if the zones overlap, you don't really have punching shear, but rather beam shear. With 4 piles, you would use the full pile cap width and effective depth. Check punching shear under the column.

 

[asixth]

If a pile is going to punch through the pile cap, it doesn't matter if the pile next to it is within the perimeter, it will still punch through base on it's own critical surface perimeter.

 

[oneintheeye]

but if the zones overlap wont the shear stress at this point be greater i.e. some combination of the stresses imposed by the two piles.

 

[kslee1000]

If cones overlap, you have developed a group pile w/r to the cap (it feels one big pile below). Yes, check single (each) pile capacity as usual, then check the outer perimeter that encloses the pile group.

 

[oneintheeye]

ok lets complicate this further. Piles are not evenly loaded, three piles are in varying compression loads, one is in tension. How does that effect the combined perimeter?

 

[kslee1000]

If you have couple in pile cap, it becames a beam, or slab. One way action tends to govern the design.

The overlapping condition can be eliminated (for coupling piles) by adjust the pile spacing farther apart (note: the two cones are in opposite directions vertically). I wouldn't worry too much about overlap in this case, draw cone diagram with stresses along cone surfaces, you would know why.

 

[oneintheeye]

I understand, my probelem is due to exsisting structures i cannot space piles out further. Ah the joys.

 

[kslee1000]

For compression pile, the wide open mouth is on top. For tension pile, the cone is flipped facing down. Unless the pile heads are set in different elevations, I don't see the cones can overlap. Are those battered piles? If not, I afraid your analysis lacks logic.

 

[oneintheeye]

but three are in compression and all overlap. Do you work out the total area of the 1.5d zones (subtracting areas due to overlap) add up the total load of three piles and divide by the perimeter of the theoretical 'cone' of whatever shape.

 

[kslee1000]

I guess I vaguely understand your question, for certain, please post a section view with direction of pile reactions shown. Also, need to know - do you have rigid metal cap on top of tension pile to develop the required tensile strength, or it is relied on the rebars only?

 

[oneintheeye]

its reliant on rebar only. Its a simple square base with equal spacing of piles in 4 corners.

 

[kslee1000]

Then. shear cone would not develop around the tension pile, as it relies on bond developed around each indivisual tension bar.

However, your design is likely governed by one way action, treating the pile cap as rigid beam, supported by structures above, with concentrated loads from piles, and soil reaction.

Strut-tie method is strongly recommended for this case.