I am doing a 2" topping on a precast concrete plank with 3/4" pex tubing. We need to waterproof the plank first. Then should we put rewire above or below the tubing? I was thinking rewire above would be stronger, does it matter?

usually it is installed below the tubing to allow for easier installation and the ability to tie to the wire. however I was concerned where you have the insulation installed, there is no mention of this. if you do not isolate the thin slab from the spancrete your radiant floor will become a huge heat sink. All the heat you throw at it will absorbed by the spancrete and will become a dissapointment to you since it will not work as you expect it to. I am also curious as to the size of the tubing, choosing 3/4" tube is great for long runs but normally you will see 1/2" tube for two reasons it's profile is shorter esp. in your thin application and it is cheaper to install typically you should be able to use the smaller pex on this depending on the sq. footage of course. Obviously I am not aware of all you have but these are just some typical points

I was wrong, it is 1/2" pex.

I would not think heating the concrete planks would be an issue, since it is living space below as well. And there will be a waterproof membrane to provide a thermal break.

Brad
Most planks are 12" thick and the issue is not weather it is in the space but rather how large is the mass of the floor and where do you want the heat that is radianting to go. It will take an excessive amount energy to heat a 14" slab vs a 2" thick one with proper thermobreak (by the way a piece of vapor barrier is not a thermobreak, you will need something with an R value of at least 5.) If you are trying to radiate up and down then that is whole different story.
We recently started up a system in the beggining of a severe cold snap and although it had 2" of foam below and on the edges, it took a long time to bring the slab from 50* and be able to start radiating heat to the space. the slab is a shop and is only 5" thick and the radiation direction is mostly upward. I would as rule of thumb and with much experience doing this put a proper thermobreak below and on the sides where you can. As an ICF installer you understand the need to keep the heat where it belongs, energy is only going to get more costly.

What if the underside of the plank, the ceiling below, was insulated?

Brad, what groundloop is saying is that it is the total mass of the plank that will make it take forever to get warm and start radiating.  Laying insulation underneath  all that mass does nothing to improve that.  The mass is still there.  Think about it.  You were planning on using 2.5" of concrete.  If you don't thermally break that from the 12" plank, you now have 6 times the mass to heat up.  By January it might be warm!  (a little exageration!), but hopefully you get the point.  If it would normally take a day to heat up the 2.5" now you're a week.  You only want to heat the 2.5".  The thermal break is really necessary.

The doors have been set, the ceiling height determined, we have just two inches for floor thickness. No room for insulation or thermal break. It matters little how long it takes to warm the floor, it is a house and will be heated all winter long. That is why I asked if it would help to insulate the ceiling.

What about the wire?

How, then, are you going to determine how much heat to put into the floor?

heat until the air reaches the desired temperature.

How about the wire?

Could brad glue down 1/4 foam board below the thin slab? You would lose little mass, have a small thermal break which would be better than nothing. michael

I thought the thermal mass inside the building envelope was a good thing, it just takes longer to heat up, which is not a problem. This floor is not 12" solid plank, it is 8" hollow core plank, like most. I have placed concrete on Insulated Floor Forms that had hydronic heat in it, never heard anything bad about it.

I did not place the plank or determine the thickness of the floor, the GC did. And it is too late to change.

How about the wire?

sounds like it will work fine given the core is hollow, is there conditions space under the plank? if not maybe consider a spray foam insulation then fiberglass? aim for R38.
Were I doing this I would use flat wire mat sheets, 1/2 pipe on a 9 inch pattern, bag ties pipe to the wire, perhaps a slab sensor and air but the air sensor alone will work, keep the fiberglass strand out acid wash the concrete and you have a nice finished floor.
Dan

I don't know if there is room on the ceiling below for any insulation.

If the wire is under the tubing it will be laying on the plank and not doing much to add strength. If the wire is on the tubing, it will be more difficult to get the tubing right. What would you do, put the wire above or below the tubing?

you need to insulate below or you'll have a radiant ceiling to the floor below.. uncontrolled. best case scenario is that the 1st floor will work harder than necessary, raising your water temperature requirements, whatever you have on the lower floor will stay off, and it won't be horribly uncomfortable. More likely is the water temps will raise, the basement emitters wasted, and the basement will be warmer than you want it.

Wire is not usually used for strength in 2" pours. Wire doesn't really provide any strength anyway, it just holds chunks of concrete together after they crack. If you want strength, my understanding is that you need rebar.

I have no idea if you do or not except that in the more typical 1.5" overpours my clients do, rebar is not a typical detail.

and 3/4" tubing is a waste of money here. for a thin pour you want a max 9" o.c. install, as Dan notes, and smaller pipe is fine.

Brad,
lay the wire mat down, then tie the tube to the mat. There is no way to attach the tube otherwise. Is the 2 inch slab finished? if so you will not want fiber mesh, or if you do you will have small bumps on the surface.
Your structure is obtained by the plank, any cracks are aesthetic, won't make a difference. If you are doing a stone or tile add glass,
As it is a conditioned space below I suspect in this case insulation will not matter, With the pipe in the top 2 inches most of the heat loss will be upward. likely you will have a sheet rock lid with air gap on the under side of the planks, perhaps some bubble foil bubble would be a good choice in the air space, though I can not see much gain in stopping the residual downward heat loss in a conditioned space..
Dan

without insulation you will have a downward radiant ceiling controlled only by a thermostat on a different floor. he is going to heat this whole mass up, and when he does, downward R-value is negligible without insulation.

stopping downward loss allows you to actually control the temperature of the floor below, and it keeps your water temperatures low for the floor above rather than making it run as if for a much higher load.

There are a lot of 80 degree basements out there because people neglected to insulate under their radiant. It can easily happen in finish space too.

I doubt that you need anything for stiffness or strength if the 2" of concrete is poured directly on top of 12" of concrete. I'm not a structural engineer, but rebar in concrete is used for stiffness, to prevent the concrete from bending and breaking (which is a kind of strength that concrete does not have), not to prevent shrinkage. If you want to prevent plastic cracks or shrinkage cracks, you should use plastic fibers and tool appropriate joints.

On the thermostat, you will not be able to use air temperature to control the slab - the slab will have way too much mass to control over a 2 or 8 or 24 hour period. You'll be able to alter the temperature over a 2 or 3 day period, but not more quickly than that. You just won't be able to pump heat into the slab quickly enough, and the rate at which it radiates its heat (about 7% / hour) will not allow you to turn it off quickly either.

Jeff

If the concert plank is cored, air space, there is going to be a reduction in heat transfer to the downward side, I suspect that the downward heat loss is not going to be the same as a 12 inch mass of cement. remember the pipe is on the top 2- 1 1/2 inch if the surface.
I agree that a simple barrier insulation could help mitigate the downward heat loss. The idea of an air and slab sensor combo would be a good solution.
Dan

a few airspaces in a cast concrete medium means nothing. Cinder blocks aren't very good insulation either. it's better than pure concrete, but it is nothing at all like insulation.

I am not advocating for a "barrier" insulation. I am advocating for real R value. R10 between heated floors, R20 over tempered space, R30 over cold. rough numbers of course.

Sure it is- cinder blocks have nearly 50% more R-value than concrete! :-)  (lessee  150% of near-zero is...  lemme find a calculator...)

And I'm not sure when/how people started thinking of a waterproof membrane as a thermal break either- that's why you never need actual insulation under membrane roofs, right?  Got an ASTM C518 test on that membrane?  It may be a sufficient thermal break to allow you to pick up a cup of boiling water, but probably not enough to hold it in a firm grip for more than 20 seconds, and it sure isn't gonna be more thermal break than an asphalt tile on a concrete floor, and you know how well THAT keeps your bare feet feeling warm on a 55F slab, eh?

Of COURSE you need some real insulation between the heated slab and the zone below it- this isn't rocket science! 

Or is it? Didn't someone post a picture of a heated slab poured on corrugated steel on steel trusses with no insulation below it here last year, wondering why it wasn't keeping up? IIRC There was more exposed surface and less R-value to that exposed zinc galvanizing than the finish-floor materials in the space they were trying to heat.  (I can't imagine the mental models of heat transfer that some are carrying around in their heads...  What were they thinking?)

Insulating paints, anyone?