Everyone tells me geo and staple up radiant aren't a good fit like it's a mantra. Why not? Radiant underfloor is not the preferred method, but sometimes in retros it's what is available. Like any heating it is simply a matter of how many btu's you need to accomplish your goal. More BTU's with low temp is simply more pipe right? Have any of you put together a successful low temp underfloor? how did you do it?

Well, below-floor typically requires a relatively high supply temp (as compared to slab-on-grade, thin-slab, and above-floor) in order to produce the upward heat gain required by typical buildings. So in a way, you sort of answered your question yourself...it is a matter of how many BTUs you can generate to achieve the goal. We have hydronic radiant floor heating design software on our website if you want to see first hand how different hydronic floor assemblies significantly affect required supply temp.

I would say "below floor requires more btus" higher temps is only one way to achieve that

Right, the other way would be higher flow rates.

....or more pipe

Our typical staple up is designed at 135.
Heat transfer plates, pattern is 8" OC, 1/2 " pex 300' runs.
increasing water flow will only go so far, better to look for ways to increase the temp a few degrees, or as you said double up the pipe.
Geo should be able to hit 125 degrees easy, perhaps what is more likley at lower temp asuming you can make heat loss is the system pumps will run longer, saturating the sub floor with more heat.
Another important factor in staple up is downard heat loss, insulation preferably a foil face or something that will also reflect upward.
We sell a producut called ULTRA NO TEAR (NT)-RADIANT BARRIER rFOIL, this is a tightly roled product 4'x125' that resembels a foil faced blue tarp,
I have used this as a heat deflector as well in combination with heat plates, While in log form it can be cut on a chop saw into 16" roles and stapeled to the joist about 1 inch away from the sub floor. I sugest the perferated version, if there is a leak it will track through the r foil.
Dan

If the underfloor cavity temp approaches the entering water temp, neither an infinite number of pipes nor an infinite flow rate will make any significant difference. It's a case of diminishing returns. On the other hand, if the cavity is 100F and the water temp is 125F, more pipe/plates would definitely move more btus. If the water is entering at 125F and leaving at 100F, more flow would definitely help. Consider outdoor reset so you don't kill the geo efficiency when you don't have to.

with heavy plates you can reduce water temps compared to other joist methods. They are pricey, however, and we are doing a lot more suspended tube/radiator combination systems to keep water temps low, condition floors, and hit higher heat loads with lower temp water.

do not waste your money on foil. use real insulation.

Heavy plates in a STANDARD subfloor situation can perform similarly to many commercially available overfloor products. Most of which are very poorly designed or designed for benefits that are not thermodynamic. I do not count high performance panels like roth, warmboard, raupanel in that assessment.

with geo it's just about water temps. you have to know what water temp you need and what output you'll get. there is no magic beyond that.

Dan: Take the Physics 101 version of thermodynamics, and be sure to wake up when you hear the name Stefan-Boltzman. Be sure to bring a calculator that can do fourth-power exponents so you can be sure to do the math. The effectiveness of radiant barriers at these low absolute temperatures and narrow deltas between the radiating & emitting surfaces are just pathetic compared to the benefit when the temperature difference between ~270 Kelvin earth temperatures and 6000 Kelvin solar surface temperatures. What's essential for spacecraft survival has little bearing on what should go under your floor, and many performance claims from RB vendors borders on FTC-prosecutable labeling fraud (which they verbally wiggle & squirm around trying to stay on the legal side of things, not always with success.)

Foil-facers & radiant barriers are especially useless when using aluminum heat transfer plates, since the emissivity of the aluminum is already so low that the foil has nothing much to reflect. With the one low-E surface called an aluminum heat transfer plate, the heat transfer downward is dominated by conduction & convection.

When the staple up does NOT have transfer plates the water & tubing temps are higher, and both wood & PEX are very emissive. With a 1" gap the radiant barrier does at least SOME good, but not in sufficient proportion to it's cost to be cost-effective. Snugging up fiber insulation to the tubing & subfloor is money better spent, and more effective overall.

Oh boy here we go again,
book smarts and street smarts.
To quote what I said"Another important factor in staple up is downward heat loss, insulation preferably a foil face or something that will also reflect upward"
Perhaps I was not clear, if you have only craft or bare insulation insulation you might " consider" using the Radiant barrier combined with itchy, it will help. The captured air pocket heat plates, foil reflection and fiberglass are all good combined choices to drive the heat where you want it.
We are always at this abyss here.
I am not suggesting the foil has much value for anything but in this application it is well placed for the modest cost.
But to each his/her own.

And yet you called out ULTRA NO TEAR (NT)-RADIANT BARRIER by name...

...which is even at it's alleged "modest cost" far more expensive than could be rationalized by it's performance in any staple up application.

yes. Dan, it's a waste of money. pure and simple. if someone gives it to you for free, go ahead and use it. but never buy it.

.25 sq ft., the product has a name, thus calling it out by name seems to make sense. It is a different beast than the bubble product, looks more like a pressed log and is perforated . Cuts with a skill saw in the role. But, like I said earlier couple it with insulation or don't use it at all.
I like it, but.. what do I know, having placed it on several jobs over the years I never regretted the choice.

the point we're making is that reflectivity is not worth much in residential radiant systems. people like it because it has "radiant" in its name. not because it's worth using. it does something. it gets dusty. then it does less. real insulation doesn't have that problem. so why use it? insulate properly and downward loss is not a problem.

Snugging up fiber insulation to the tubing & subfloor is money better spent, and more effective overall.

I expect that an air gap below the floor (and above the insulation) is beneficial. No sense in eliminating convective heat transfer from the pipe/plate to other parts of the floor (ie, the parts not covered by the plate). And add something to stop air flow if fiberglass is being used.

air gaps in plated systems are very minimal in benefit.

125* is definately not a great geo design, 85* would be but we can live with 100-110F.

Air gaps are a thermal bypass route for air currents in less-than-perfectly seal joist bays. (But we always have PERFECT air sealing in our joist bays, right?)

The RB layer in direct contact with the fiber can be an air-barrier to limit convective losses through air-permeable insulatio, but that's about it for it's true performance enhancement in this app. You can get the same air-barrier benefit with paper, Tyvek, or 6-mil poly, but using a kraft facer on the bottom side is good enough if the room below is conditioned or semi-conditioned space (could have moisture-trap ssues if it's an open pier foundation in cold climate though.)

Nothing like hands-on proof. I suggest insulating one bay with fiberglass + tyvek and another with fiberglass + radiant barrier film and sticking a temperature probe into each one. The warmer one will be whichever one achieves the better air seal :-).

What is not needed, is by definition, wasted.

Dana put forth a very succinct and typically scientific argument opposing the wasteful use of foil "reflectors" so often hawked by the questionable and bought by the ignorant. With a fleeting grasp of "radiant" energy the layman, and his vendor, are easily duped by claims defying physics.

We will not use suspended tube for any reason, lest the climate is so mild, thus the heat load so low, that floor " warming" is all that is required. Heavy extruded plates, as Rob suggests.

When designing a radiant floors, wall and ceilings we look to the heat load and then the heat source for our ultimate choice of emitter. If you have chosen a specific heat source then the construction of the home, the design of the HVAC systems including domestic hot water, must follow the heat source.

Naturally, this is rarely the case, so compromise on comfort, economy or both is the unfortunate norm.

If your PEX is suspended (not true staple-up, which hasn't been done by the well-informed for more than a decade) then your radiant floor started out as a poor hydronic convector and after super-heating the air in a tight joist space, finally heats the floor above to create the radiant panel we call "infloor" heat.

The first order of business it to reduce heat-flux downward and the second to isolate the necessary air movement to the smallest effective area possible. This appears to be roughly an inch below the suspended PEX. To the degree that any "foil", vapor barrier or rigid insulation my accomplish this, some value may be added.

In my own recent remodel of my 1921 farm house R19 and R13 fiber-glass under extruded aluminum plates, allowed me to operate my systems below outdoor design conditions with a EWT of 120°F. Nearly perfect any heat source and all under 3/4" pine sub-floor topped with 3/4" oak flooring.

Each system must follow an ACCA Manual 'J' heat load in order to predict the ultimate performance of any radiant panel. Much like a wall "system" Dana educates us on so regularly, all radiant panel not hanging on a wall or standing on the floor are part of a constructed "system". Therein, in many errors may occur.