We are building a new 2200 sq ft house, well insulated with sip panels. While Warmboards are clearly the superior product, they are also $7000 more than what we can get Roth panels for, and an even greater difference for Zurn Thermal Track. Here are the questions:

1) I know that thick aluminum is better than thin, but how MUCH better? Roth and Warmboards are .02", and the thermal track is quite a bit thinner at .004". I figure that even if the warmboards were so efficient that they saved us $30 a month in heating costs (over the thermal track), it would take twenty years to make up for the greater upfront cost. Does anybody have knowledge about how much more efficient the thicker stuff is? And does it spread the heat more noticeably?

2) Does anybody have experience with Roth panels? Our concern is that the foam may compress over time. We are also wondering about installing hardwood floors over them- they say to use nailing strips in between the panels, although it seems awkward to do this around the places where the tubing bends. And there is no aluminum on the rounded ends- is this noticeable in the floor heating pattern?

3) Does anybody have experience with Zurn Thermal Track? Does the floor heat evenly in spite of the thin (and therefore lower conductivity) aluminum?

thank you for any input!
Laura

We manufacture a product RHT floor panel system that is about 1/2 cost of the other floor panel systems.
This product is designed and manufactured in the USA by installers.
The key is that we use 1/2 pex pipe, letting circuit loop runs go upward in 330 feet. We provide the turns and a lay out, the turns are the most difficult to prepare on site. We also provide the heat transfer plates and pex, the 3/4 plywood fill is purchased localy and site cut minimizing freight transportation. cost is about 3.00 sq ft for all materials, pipe, plates, plywood, and turns.
I will be glad to tell you more if this s of interest you.
Dan

That's basically a sandwich installation with precut turn bends. Mixing in some heavy gauge plates can really make a sandwich good, if you target the high load areas appropriately.

Zurn thermal trak is rebranded thermalboard.

how much more efficient can only be guessed at if we know your heat load and heat source. Warmboard is probably a good 20 to 30 degrees cooler running than thermalboard, and has much more output which might also be an important concern.

roth panel should not be used with nailed wood floors, IMHO. It's a big pain to work with in that case, you basically have to rip up all the panels.

Thanks for your replies. Our total load is 35,000 btuh, and the basement level will be concrete slab, so we're looking at around 1500 sq ft of floor panels. The heat source will be a gas boiler. Wehave sip panels construction, with R27 walls and R42 roofs.

Three questions about the RHT panels:
1) It looks like in this sandwich construction, most of the aluminum is below a 3/4" layer of plywood. How much efficiency is lost by having this extra layer between the aluminum and the top flooring? Does the plywood act as thermal mass, or does it act as insulation?
2) I'm wondering about inverting the aluminum plates, so that they lay above the 3/4" plywood, and the u-trough for the tube lies in the gap between the boards. This would put the plates right up under the hardwood flooring. Seems like you would get much more direct heat transfer. Is there some reason why the system isn't designed this way?
3) Are the plates wide enough that they are continuous across the floor, or do they just run several inches on either side of the tubing? And how much effiiency is lost if there is not continuous coverage?

As for the thickness/thinness of the panels, are there some actual figures available for the btu output of these various systems, so that their relative efficiency can be accurately compared? Seems like a very murky area with lots of speculation, but maybe thats just the way it is for now.

NRT, are you saying that the warmboards actually produce the same output with 30 degree cooler water? thats a pretty huge difference.

thanks for your thoughtful replies.

warmboard should be 20 to 30 degrees better than thermal board, but that difference diminishes under low R floor coverings (tile) and especially under low load conditions (very well insulated for your climate).

I have no idea why anyone would put the aluminum under the plywood. We always put aluminum plates on top of the infill when we do sandwiches. It should make a difference in water temperatures. Sandwich, incidentally, should roughly split the difference between thermalboard and warmboard. that's rough of course, but a decent approximation.

sandwich plates are not typically continuous.

Most manufacturers have output charts for their products. Some lie. Some methods, like sandwich, don't really have easy charts available and so it is a bit "murky", as you say. But even with exact output given... then define what the efficiency difference is, running your system 30 degrees cooler (to take a very big example). Go ahead, I'll wait ;)

There is a lot more that is "murky" than just output at a given water temperature, for sure. You can be sure running much cooler is potentially more efficient, but how much is always hard to say.

OK,
I will try to reply,
Yes it does seam backwards to place the plates as we do, and yes we live in a democracy so as some of our customers do they place the plates grove exposed. to do this is easy and certainly accptable, we prefer the protection the plates offer to the pipe.
In our application it is quite fast these are the steps.

Detail;
1st. place turns first according to lay out, turns have a slight bevel so the pipe is captured.
2nd lay in 4 or 5 lengths of pipe(turn to turn) and cover with plates (1/2"x24"x5" Heat Transfer Plates single 'U' style .019" thickness)
3rd use site manufactured 3/4 ply wood strips ripped 6 3/4 inches fasten down with nail gun, screws, hand nail ring shank, what happens in this process is that the plates, pipe and plywood are smooth to the floor and the straight edge of the 3/4 plywood straightens out the floor.

If you chose to do the top down approach you follow similar steps,
1st Place turns
2nd place 3/4 X6 inch plywood infill strips,
3rd place heat transfer plates grove up, secure with your favorite (hand nail, nail gun,screw).
4th place pipe

Both systems work great,

Our average system temperatures run in the 100-120 degree range, less is better. the plywood becomes saturated with heat and works like a thermal mass.
We have had zero complaints with this system.

We do not have and charts, we just know it works. it is a tremendous improvement over staple up, system temps can stay quite low. the saturation of heat is held with in the envelope.
This brings to question how much heat in the warmboard system is saturation in to the wood, It certainly does not all radiate up....I think we have a thermal mass issue here, though not as prefered as concrete,

On another design note worth mentioning, a clever woman in up state NY applied this system using our 12 1/2 inch wide plates over a uneven barn floor, I think this is a great think out a the box design;
not level uninsulated slab,
She took long 2x4"S pressure treated, laid them out on a 12 inch pattern, shimmed to level and fastened to the concrete placed Bubble foil bubble ( I know smoke and mirror) between the 2x4"s . she was working on a slab error of like 2-3 inches over 60 feet and did not want more cement and foam, we went there with her.
Next using the 1/2"x24"x12.5" Heat Transfer Plates double 'UU' style .024" thickness she stitched the 2X4"s together, snapped in the pipe, 8 inch pattern, and captured the whole thing with a ply wood top.

Dan

Wood is not a thermal mass. it's practically negligable in its ability to hold heat. and it does provide resistance to heat transfer: if minimal water temps or max output is the goal, plates go on top. If you sprayed the tubing with insulating foam, all the heat "stays in the envelope" as well, but you'll still need a heck of a lot hotter water to actually emit any of it at a given rate. Wood isn't as bad as insulation of course, but the principle does not change; more layers, higher water temps.

I agree wood has an insulation property, R-1 per inch. At least that is what I am told when looking at R value of say a log building.
Closed cell foam similar to DOW board is R 5.5 per inch and has little ability to hold heat as mass. Wood on the other hand does. Further we also see that in the sandwich application there is surface exposure of the top 1 inch of the heat plate, or on an 8 inch pattern 15% of the surface of the plate is exposed on the upper layer.
I whole heartily agree that if you want more surface area place the plates top side rather than captured.
Point is it works both ways at low temp, I can only assume that there is a heat saturation effect.
More specific is the RHT Floor Panel system has its place. It is a direct substitute for a large number of the conventional systems offered by the cooperate giants at less than 1/2 the cost. By relying on the ability to rip plywood in the field using local lumber we present a some what greener alternative than shipping an entire above sub-floor system across country.
We have a large project coming up locally in the next couple months, I will suggest to the installer on site to place one zone plates up, we will photo documenting the project anyway and this may be a good start. The whole house will be heated with a condensing boiler (LAARS Neo Therm, Nice boiler I might add).
The Design temp is set at 110 degrees. Perhaps a side by side study of 2 zones will be of use. The only real difference I would expect is speed of recovery.
Dan

Thanks to both of you for your suggestions. We will most likely go with the plates, on top of the infill. And once we know our heat load calcs, how do you then determine the spacing of the tubes? Seems like you could just go with all 8", or all 12", and then adjust the controls for each zone according to whats needed in the room. Do either of you (or anybody else) have thoughts or suggestions on this?

Another factor in the tube spacing is that if we go with the RHT panels, if we choose the 8" spacing then they are the single channel plates, and they aren't wide enough to cover the whole gap between tubes. The 12" come in double UU channels, and therefore give more aluminum coverage. Any thoughts on the need for continuous aluminum coverage?

You don't want more than a few inches between plates if you are on top to avoid heat striping. but continuous is not necessary. We use a 5-3/4" wide plate 9" o.c., or a 4" wide extrusion for high output at 7" o.c. (both picked to minimize plywood waste as much as anything.. 1" gap between strips).

Dan: you know there is more to expect than just slower response. What affect does throwing another 1-R down on top of your floor have? That's the difference between "wood" floors and "carpet". It's almost the difference between underfloor and overfloor systems. It makes a difference. If you're going to put the plates under the infill, save the cost, expense, and resource usage of your extra layer of plywood entirely and stick to the joist bay.

And the plywood was likely shipped just as far as Warmboard.

And you're really missing the boat if you are counting on the wood to act as a thermal mass. Pound for Pound it holds half the heat of concrete, and you have a lot less pounds per square foot than you would with concrete. "not as ideal" as concrete is a bit of an understatement: it's useless as a thermal mass. Utterly useless.

OK,
First pipe pattern, If using the RHT floor panel system, turns are on 8 inch pattern, the 5 inch plates are easier to line up and work fine.

Ok, Thermal Mass, first agreed wood is not ideal and pales in comparison to concert.
Post form another green builder, solar section;
Concrete has a specific heat of .156 Btu/lb/°F and a density of 144 lb/cu ft. Pine is .45 & 34. That means it would take roughly 3 times the pounds of wood to store a thousand Btus as concrete, and since it weighs only about 1/4 as much the volume would be about 4 times as much per pound. It would take roughly 50" of pine thickness to store the same amount of heat as 4" of concrete. Oak is .57 & 50 so it would take only about 30" of oak!

My point was that there is stored thermal mass.

Next Wikipedia, search Thermal mass;

Properties required for good thermal mass
Ideal materials for thermal mass are those materials that have:

* high specific heat capacity,
* high density

Any solid, liquid, or gas that has mass will have some thermal mass. A common misconception is that only concrete or earth soil has thermal mass; even air has thermal mass (although very little.)

Again this is no substitute to concrete. We are in full agreement here.

Last, movement of plywood vs specialty items made of wood.
Plywood is a huge commodity,made in mass production runs at maximum efficiency vs waste. If not manufactured locally/regionally it is most often transported by rail. Rail is the most cost effective mode of transportation short of sea container.
So if the stars line right your ply is made regionally, or second best moved geographically closer by rail. These are the contributing factors on why I can buy 3/4 ply 4x8 sheet for 25-30 dollars in anywhere USA. Am I fooling myself?
Done for the day, best to all.
Dan

Dan, obviously wood has some mass. while all is pure energy, you can pick up wood and feel that it has weight. Kudos for noticing this ;)

even though it technically has mass, it's not enough to be useful for any purpose we are discussing. That is why we don't build wood passive solar walls or collectors or tell people to add layers of wood to their floor if they are short cycling their boilers. Because it doesn't help and it only serves to confuse the issue to talk about the "mass" of the wood floor. Why not also talk about the "mass" of PEX pipe or the aluminum plates themselves? Because it too would be silly.

Mass collects and stores heat. Wood does not do this to any level that is worth considering. You are using wood as a conductor, which it is not good at doing, and that is all. The only benefit mass brings to the table anyway is either collection of passive solar energy, time-shifting your heat generation (say,to off-peak times), or cycling control on a low mass heat source: wood won't help with any of this. so practically speaking, it is not a thermal mass. Which isn't a big deal, because mass itself is only useful for those three things anyway, but acting like "wood has thermal mass" isn't much more useful than throwing down a carpet and claiming it too is thermal mass.

Point taken on rail shipping.

OK,
glad you see my point on Ply,
The rest is hair splitting, I am not promoting banking thermal mass in wood in place of concrete.
We are both on the same side of the conversation.... almost
The nature of heat transfer loads the 3/4 ply up w/ heat, it spreads out over the surface.
Again it can be used both plate up, or plate down, same water temp, low works fine.
I understand you are correct about the resistance of the wood, my point is that that resistance has a curve where it is overcome and becomes "mass"
This one needs to go to the lab, Perhaps myth busters.
Dan

Dan, I"m not sure what to say if you can't acknowledge that adding an R1 to your radiant assembly does in fact change the water temperature it requires to operate, and dismissing your additional layer of insulation as "mass" is quite simply wrong. This is not hairsplitting. The two systems are not equivalent. Similar, perhaps, but not the same, and there are significant benefits to aluminum on top.

Not to be harsh, but this doesn't need a lab. It needs a basic understanding of heat transfer. Would you really tell someone that it would have no effect on their water temperatures to throw a second wooden floor down on top of the first, because it's just "overcome" when it heats up and becomes "mass"? That is almost exactly what you are saying here.

In reality, the water temperature requirements of a system can easily be 5 to 30 degrees different with an extra wood layer, depending on starting heat load. You can extrapolate that easily from the output charts of any installation method you care to check: look at the R1 output vs the R2 output. In this case, it's really about an R-0.75 difference and if you want to be really charitable with the "upward" channels you could call it an average 0.5 R for estimation purposes. but still, the wood doesn't "disappear" when it heats up any more than insulation does. It remains a resistance to the transfer of heat through it. The "curve" where it is overcome is the water temperature curve, not just a timed event.

This isn't esoteric, it's not obscure, it's not difficult to understand. I am not saying plates down don't work any more than I am saying joist plates don't work. I am quite simply saying you sacrifice performance to put the plates down, and there is absolutely no reasonable argument against that unless you want to start a discussion on how floor coverings over radiant don't matter either.

You might argue why that sacrifice is or is not worth it, but to say there is no sacrifice is simply wrong. I don't regard (if we use the 0.5 estimation) up to 15 degrees of water temp as even a "small" sacrifice. It's quite significant in many cases. That's almost a full COP you lose on Geothermal powered systems.

Rob,
I agree you are correct in part, however this system works well at low temp.
Clearly the best way to lay the pipe for heat transfer is upward, for speed and protection from the trades the way we show capturing the pipe works well also. We will keep looking at some alternatives, and perhaps a alternate application based on the up and the down approach. Our goal is to provide what works best and is cost effective. Another thing I am looking at is offering a 48 inch plate, this to could add to the install efficiency time.
Short on time today,
Dan

With the plates up, is there an issue with installing a hardwood floor, other than obviously not nailing the tube?

Do the plates and tube lie flat enough for a tounge & groove install?

Will the floor squeak?   Should you leave a few inches between the plates for nailing?

you can nail right through plates without consequence. squeaks are a wood on wood issue.

Here is a link to our heat transfer plates, You will find sizing data here. The pipe is fully captured by the plate that holds it tightly.
http://www.blueridgecompany.com/radiant/hydronic/316/rht-heat-transfer-plates
Dan