slab on grade for rigid frame buildings

huangyhg

slab on grade for rigid frame buildings
when using hairpins to transfer frame lateral loads into the slab, how do you handle slab shrinkage?  i'm guessing there cannot be any control joints in the slab, and the hairpins tie the slab to the foundation walls.
i think this is something that is overlooked often. when the mesh is cut at the joint then there is only the first rectangular piece of slab resisting the thrust. this is made worse when a vapor barrier is directly beneath the slab by  reducing the friction.
i like to use tie beams that span the width of the building and put control joints between and directly on each side of the tie beams.
haynewp-
thanks for the respone. sounds like the joints you describe are running parallel to the tie beam.  i'm really interested in the joints running perpendicular to the tie beam, and i'm really interested in how the slab handles both the thrust and the shrinkage when there is no tie beam.
well, i am suggesting a tie beam so you can run the joints perpendicular to it and not worry about the slab as the tie beams take all the thrust.
if you are set on using hairpins instead of tie beams, then you have to check there is enough slab between the first joint and the edge of slab to resist the thrust. i have done this before when i had a small amount of thrust to resist. i had to balance this requirement with the maximum jointed area of slab for shrinkage.
i am not sure how else to handle this, i always show the mesh cut at the control joints (though this likely doesn't happen) so i can't count on that to grab the remainder of the slab past the first joint.  
i have added bars lapped on to the hairpins to make sure i could activate the required slab area - regardless of where they saw the joints.
i typically use 2" top cover to slab reinforcing and stipulate the sawcuts are 1-1/2" deep; this saves wear and tear on the saw blade, too.
dik
like dik, i detail the mesh at the slab control joint to not be cut.  then you can count on more slab to resist the horizontal force (or even tie all the way across the building).
but then, you may ask, how does the control joint function?  i have been told that there has been research done that shows wire mesh, still intact across a control joint, can elongate enough to allow the control joint to crack and open up.
daveatkins
thanks everyone.  some of the suggestions run contrary to my normal practice (i.e., running the reinforcing through the control joints, using the slab weight to resist the thrust).  but they obviously must work.  after all these years, i still have a lot to learn.
i never use more than one panel to resist thrust, even if the mesh is continuous through the joint and could develop more.  you don't really know where the mesh is in the slab - it could have been cut.  i count on passive pressure against the footing or turndown to provide the rest of the resistance.  there is a practical limit where hairpins are not feasible anymore to resist the thrust - that's where i go to tie rods from column to column.
daveatkins,
i am interested if you can find that research. it wouldn't surprise me if that is the case.
i am pretty sure the control joints don't actually line up with the cuts in the mesh, but i always see it shown this way. the mesh on my projects normally ends up sitting on ground (they try to pull it during pouring) even when i specify it as chaired so i don't know how effective it is one way or the other.
when i have designed buildings with post-tensioned ground slabs, the hairpins were welded to the columns after the slab was given time to shrink.
they used u shaped hairpins with a flat section placed against the side of the column. after the slabs were tensioned they put steel spacers in the gap between the hairpin and the column and welded it to both.
normally we would put joints in a slab even though it had hairpins.
csd