post embedmen

huangyhg

post embedment
i have a question on post embedment in a concrete drilled shaft/caisson.
i am doing a soundwall design where the soundwall posts are prestressed concrete in the shape of an "i".  the panels are also prestressed concrete.
we are trying to calculate how far to embed the soundwall posts into the drilled shaft/caisson foundations.  the foundations are 42" in diameter.  to me, as the post bends, it will try to split the caisson in half somewhat along the axis of the wall.  the reinforcing hoops that we put in the caisson will try to prevent that splitting.
i feel like i have a handle on the mechanism of failure, but am having a hard time putting numbers to the length of embedment required or the size and spacing of the steel hoops required.  right now we have 4' embedment length, which seems reasonable.
any thoughts?  has anyone else tackled this problem?
check out our whitepaper library.
i have handled this situation as follows:  factor the loads and design using ultimate strength.  the post will load the shaft to a compressive stress of .85 f'c on opposite sides of the post to form a couple that resists the applied forces on the post.  using this method you can determine the embedment that will provide the required resistance.  with a flanged   
broekie 鈥?i agree with the method graybeach has described.  i have used this same method for determining pile embedment in to footings.
thanks for the input.  let me make sure i understand you correctly.  the moment in my column is 5300 in-kips.  to be a little more conservative, i am going to use, from aashto, the allowable bearing in concrete of .85*.7*f'c from aashto 8.16.7.1.  if i have 4' of embedment and a 16" flange width on the post, my equation to solve becomes:
5300 = (.85 * .7 * 3 ksi * 16" * a) * (48" - a/2 - a/2)
a=4.24" so my embedment is ok.  is this what the two of you were trying to convey?
yes....your actual a=0.85x24"=20.4" which is much greater than the required a=4.24" you calculated.  i am not aware of any way to calculate the size and spacing of the hoops for this kind of detail though.
here are a few standard details from ohio dot that show the hoop details in pile supported footings and pier caps.  i think you鈥檙e a bridgie, broekie, so this should be right up your alley.
broekie,
you also need to sum the forces in the direction of the moment and add the pressure resisting that to the embedment.
the way i understood the greybeach post, the effective flange width is not 16".  did i miss something here?
the force that is transferred to the shaft comes from web of the post.  the effective width that presses on the shaft concrete depends on the web and flange thicknesses.  16" would be ok for a 6" thick web and 5" thick flanges.
the web of the post is 8" wide and flanges are 7" thick.  the flange width is 16" so i used that value since it is less than tw + 2*tf = 22".
oh, yes, you did say prestressed i beam, didn't you?