My house in Milwaukee WI has very cold 6" concrete slab terrazzo floors on the first floor (maybe 50F), and basement below (which needs much insulating work) is about 48 degrees.  I am interested in the possibility of adding radiant heat to some parts of the slab from below (easy access) to make walking around barefoot possible on main level.  Forced-air furnace is not very effective.  The radiant pipes could be insulated with closed-cell foam (I have read they need R-19 at least).  My questions are two:  1) would this heat up the floor effectively and 2) how should pipes be installed under the floor?  Do they need metal plates to spread the heat?  One installer (from Vermont) said I should use 5/8" pipe 10" on center.  Any ideas?
Thanks

Are you saying that you want to add tubing beneath the 6" concrete slab.

Yes, if it can be put in so that it will heat the floor above.  I realize some heat would probably move down into the basement, but it's 48F down there.  Consequently, so is the first floor in many places.

Every proper radiant floor design starts with a heat load analysis, performed by an experienced radiant designer, using dedicated software made for the task. This is even more critical on renovation jobs when driving radiant heat through floors, ceilings, walls or even concrete, is being considered.

After the heat load is established the various radiant panels applications can be assessed for efficacy. In most of my slab-on-grade residential applications the tubing is stapled directly to the sub-slab insulation, yours would not be much different but for the lack of tube contact to the concrete. (No there is no good reason to suspende tube in a residential 4" slab).

Insulation and heat in the basement would take the load off the upper floor and improve comfort considerably.

and there is almost never a reason to use 5/8" pipe in a residential distribution application other than seller profit margin.

5/8" (I.D.) PEX does smell of an online retailer (present company excepted). It is a gimmick,

I am not following, you are going to do a "staple up" with heat transfer plates beneath a 6 inch slab suspended above your basement???

Agreed 1/2 inch pipe, easier to work with, save the 5/8 for the 3,000 sq foot pole building.
Dan

Do to the extreme thermal mass of the concrete slab and the positioning of the tubing, the majority of the radiant heat would tend to migrate downward to the basement. An extreme radiant barrier would need to be installed to move this energy upward through the slab. Baseboard or radiators may be a preferred option. Controlling the temperature in the space would be more responsive. With the amount of concrete and placement of the tubing, it may take days if not many hours to change the temperature in the space.

with good tubing contact it needn't be much worse than if the tubing were in the mass itself, especially if it is properly insulated (no wasting time with radiant barriers).

A low mass emitter in the space may make sense in some cases though for sure.

And I want to know how they supported a suspended 6 inch concrete slab in a residence. I have seen such in apartment buildings, but not any I know of in a single family, though they must exist. Wouldn't that take an extensive amount of steel I beams etc?

-Rosalinda

Unless you really loved those terrazzo floors, I would think a radiant installation above the slab would be less expensive and more responsive that going below a 6 inch slab.

If you want cheap and fairly easy, box in the area below the floor joists with rigid insulation and then run hot air from the furnace through the area (you will need some holes in the floor as duct openings). Works fine even if it isn't code compliant.

Thank you to everyone who responded.
There are 2 concrete-filled I-beams supporting the floor, about 10 inches deep, as well as the outside block walls.  Span is 30 feet (perhaps a bit overbuilt? :).  I was told it was built by a concrete contractor for himself in 1920.  Terrazzo floors are considered a luxury detail here (I even have them in the garage and the basement); however icy terrazzo floors probably would not add value to the property. 
Basement is 47F today.

I have a few questions:
I am not following, you are going to do a "staple up" with heat transfer plates beneath a 6 inch slab suspended above your basement???
Yes, you got it.  "Heat transfer plates?"
Do to the extreme thermal mass of the concrete slab and the positioning of the tubing, the majority of the radiant heat would tend to migrate downward to the basement. An extreme radiant barrier would need to be installed to move this energy upward through the slab.
would 3-4 inches of closed cell foam be "an extreme radiant barrier?"
with good tubing contact it needn't be much worse than if the tubing were in the mass itself, especially if it is properly insulated (no wasting time with radiant barriers).
So does that mean insulation is not a radiant barrier?  And how would I attach heat transfer plates, or would they be necessary?  (It's been a long time since I studied thermodynamics, and I didn't really understand it then) And the insulation would stick to the concrete ceiling, according to an insulation guy I asked, but would it stick to the transfer plates, or tend to pull them down from the ceiling?

Note I have no joists.  Area from I-beam to wall and wall to wall is roughly 10X18 ft.

Every proper radiant floor design starts with a heat load analysis, performed by an experienced radiant designer, using dedicated software made for the task.
Where would I find such a person, and/or such software?

Thanks to all for the recommendation for 1/2" pipe.  Still 10" in center would be good?  And I'd really like advice on those heat transfer plates.  If someone has a link where I can find the answers to my less competent questions I'd be grateful.
MES

I heated my 'ice' cold kitchen floor from below using 1/2" pex tubing 8" oc (or 2 runs in each 16"oc floor joist cavity) stapled to the above 3/4" subfloor. I stapled  reflectix insulation ( which blocks 97% of the radiant heat) 2" below the tubing to create a 2" hot air dead space for the heated tubing.

Below the reflectix insulation I used 1 1/2" extruded polystyrene tightly fitted to the 10" joists so most of the heat rose into the subfloor which then heated the tiled floor. The temp of this floor runs about 2 degrees cooler than the adjacent sunroom floor where I installed the pex during a floor demo project in 2" of gypcrete (best method).

 I simply used a 50 gal nat. gas heater set on #2 which cycles the water heater temp from 98 - 83 to heat the 4 'zones' of flooring. Return water enters the bottom drain opening and heated water exits the 'old' hot water port.  I keep the floor at 78-80 and run the 3 speed grundfos pump on slow speed thru an air separator 24/7 during heating season.

 My house stat does the main temp control so I chose not to cycle the pump or use any other temp control other than the water heater control. If you turn heat off to a large mass it takes too long to reheat. This constant heat in a large sunroom full of windows works well.
Simple and effective. My dogs come in from the cold outside and head right for the hard, but heated floors!

at your water temperatures, your floor temperature differential is not surprising. just putting that out there, at very low outputs the difference between methods is smaller. Just in case someone reads that and thinks that staple up tubing is as good as gypcrete, it's not. for floor warming only you don't need the best, of course.


Though obviously something is up since if you really had an 80 degree floor with 98 degree water you'd have 20 bTUs/sq ft of output which would cook you out of the kitchen during the shoulder seasons and would break the laws of physics. so I would guess your sensor is in the joist bay, and that's a joist temperature, not a floor surface temp? otherwise, something is wrong.

to NRT:

The water heater has a 15 degree dt, so the max temp is 98 and min temp is 83 on #2 setting.
I start out the htg season on 'vacation setting' on the water heater temp control and slowly raise the temp over a few days until I reach my ideal floor temp in my sun room which we like at 78-80. I reverse this in the spring.
I also ran the colder return tubing near the outside walls so I didn't lose as much heat transfer as there would be if the htg supply tubing were 8" from the outside perimeter of the space.
I hope this answers your concern. Thanks

To NRT:

An additional note. The 98 - 83 temp is the lwt of the water heater supply. The 78-80 is the floor temp I measure shooting it with a temp gun. I don't measure the joist cavity air temp, just the beginning water supply temp and floor temp.

Smosky

Ok, then something is off. If you have an 80 degree average floor surface temp you wouldn't be able to stand the kitchen most of the time. it would be way too hot.

unless you have a thermostat with an air sensor too that cuts it out?

To NRT:

I just 'shot' the floor temp in my sunroom and it's 70 near the outside walls, 75 one foot in, 78-80 on the majority of the floor and 81 where the pex supply enters the gypcrete. Remember the 98 is the lwt and max temp when the water turns off and 83 when it restarts. So the avg water supply temp is 90.
The sunroom and kitchen ( floor to ceiling windows on 3 sides) are open to the whole house and during these 'milder' days of 30 outside, the radiant floor is the main source of heat in the house. That is why I run this system 'wild', to offset the heat loss from the double pane windows. I do not have other controls other than the water heater settings ( vac - warm - hot) and the pump running at slow speed. When we get to warm I turn the WH down a notch, which hopefully will be soon in SD.
Sorry if I am causing confusion.