vapor barrier needed on mat foundation

huangyhg

vapor barrier needed on mat foundation?
i've got a small building addition - which is really just a two story set of stairs and an elevator/lobby.  we set the whole thing on an 18" mat foundation over the entire footprint.
i understand about slab curling and generally have the vapor barrier issues figured out.
but with an 18" thick foundation mat, that is also to serve as the lower floor slab, the question has come up as to whether a vapor barrier under the 18" mat is required.
normally, with a 4" to 8" slab, i would not use a vb if the slab was to serve as an exposed concrete floor as curling, cracking, etc. is helped a bit by not having a vb underneath.
and where there is carpet, tile, etc. i would use a vb for helping in long term control of moisture from below.
but this 18" mat will have vinyld tile applied to the top surface so i'm sensitive to vapor pressure coming up from below.  
would it make sense to not use a vb underneath (thinking that it would impair good curing through the mat thickness) and then notify the architect that special care, or tests, would be required prior to placing any tile on the concrete?  the tests would check for vapor pressure.  might require a longer wait prior to placing tile but that's all i can think of.

i am relunctant to not put vb in a building with finish floors. not having one will often void the warranty on the finish material, even though there is a retro way to solve that problem. even checking the spec on the material does guarantee that they may not change their mind on what they want after the concrete is poured.
incidently, it is not longer called "vapor barrier"! we call it "vapor retarder" since no product can claim 100% that it is a barrier to moisture.
jae...the moisture won't likely come from below, through 18 inches of concrete unless the water table is at the bottom.  the primary source of water that will affect your floor coverings will come from excess mix water (above that required for hydration).  even a low w-c ratio will still yield a lot of extra water.
the vapor barrier will probably be worthless excess for those warranty and code issues as noted.   
this is kind of location dependent.  here in new england, things get pretty damp. so i would be reluctant to eliminate the vapor barrier with vinyl tile. i would use the vapor barrier with 8" of 3/8" crushed stone over it.  you'd have to keep the crushed stone layer dry.
concrete construction magazine had two good articles related to this, june 1996, and july 1998. i can upload those if you'd like.
jae...on other point.  if you put a vapor barrier in, it will not solve the moisture vapor transmission problem. be sure that there is a good time frame between placement of the concrete and placement of the floor covering, specify a solvent based, not water based, adhesive, and do two types of testing before the floor is installed...1) a calcium chloride vapor emission test, and 2) in-situ relative humidity of the concrete, preferably with gradient depth testing.
ron, that's what i was aiming for originally, suggesting to the contractor (via the architect's specs) to test for vapor pressure pior to tile installation.
thanks to everyone for the responses.
jae
when you said:
"normally, with a 4" to 8" slab, i would not use a vb if the slab was to serve as an exposed concrete floor as curling, cracking, etc. is helped a bit by not having a vb underneath."
would you still not put the barrier under a gravel layer? we had a warehouse where the vapor barrier was left and they complained about the exposed slab being moist. i know this topic has been discussed ad nauseum but just curious when you said that.

we've not put a vapor retarder (hat tip to jike) - under warehouse type slabs and haven't had any problems with moisture.  
now if the warehouse was in a "low" area and the water table high, perhaps i would.  
i've never put the vr below the granular base...some suggest that is a better way to go.  i've always had trouble with that for two reasons - one, the granular base probably would puncture the heck out of it and two, with the vr below the granular base all in place, waiting for the concrete, it might rain.  then you have semi-perched water - a saturated granular base to deal with.  in tilt construction this is even more critical as the openings in the slab at the interior columns allows water in, but with a vr you can't get it out.

a concrete construction worker told me the other day that they always intentionally punch holes with a rod when the vapor retarder is placed on top of the gravel. he said this helps with finishing the slab. i thought that was kind of interesting and wondered how many other crews are out intentionally punching holes in the vr.

i guess most of them (concrete finisher) do, as it (the holes) helps to reduce the amount of water risen to the surface.  
that was my understanding from the conversation.