Radiant floor heat spreader plates
Hi,
I am putting in a radiant floor.
Over the existing 3/4 wood subfloor, I am installing half inch PEX lines at 9 inch spacing with 3/4 inch spacers or sleepers between the PEX lines. Then the thin foam sheet underlayment and then laminate flooring.
I am going to use the type of aluminum sheet heat spreaders that have a groove for the PEX tubing to sit in.
The spreaders can be installed in two ways:
1) with the groove facing up and the alum sheet extending over the top of the 3/4 spacers just under the finish floor.
2) with the groove facing down, and the alum sheet extending under the 3/4 spacers (between the spacers and the 3/4 subfloor).
I was told that the performance would be about the same with either scheme, but I did a little test, and scheme 1 seems to do a lot better getting floor surface temperatures that are higher. Since the heat is supplied by solar collectors, the lower the water temperature I can use, the more efficient the system.
So, I am inclined to use scheme 1 (with alum spreaders right under the finished floor). My question is: Are there any disadvantages to this scheme, or any reason I should use scheme 2 instead??
—-
I am using PEX-AL-PEX for the tubing, so the expansion rates for the PEX and the spreader plates will be similar.
The alum sheet I am using for the spreaders is 0.019 thick.
Details on the spreader plate test:
http://www.builditsolar.com/Experimental/RadFloor/HeatSpreaders.htm
Water temp 92F,
Scheme 1: floor surface temp 82F
Scheme 2: floor surface temp 72F
Gary
Replies
I used method one under my bamboo floor and am very happy with the result. Less insulation above the plates made the most sense to me.
One.
The aluminum plates reflect heat, if it's under the tube heat is reflected into the living space. With the plates over the tube heat is reflected away from the living space, as your test show.
"The aluminum plates reflect heat, if it's under the tube heat is reflected into the living space. "Actually in this case it would be a radiator and not a reflector.But for it to radiate the AL needs to have 1" free air space, which it does not have in this application.Also it is a good conductor and in free the whole plate would be at the same tempature and it would radiaton almost as well in both direction. The slight difference would be the space that the tube takes up.The differnence is his experiement has nothing to do with the abilitly of the AL to radiate or not.The difference is the thermo resistance of the 3/4" ply spacers.
From email."Radiant is the key word here. It can be and is reflected in this case. Put a sheet of Aluminum out in the sun, the temperature of the aluminum will change some but the amount of radiant
energy it reflects will cook your dinner. Wirsbo publishes a complete design manual it explains the principles of radiant floor, radiant ceiling, ice melt installations. Wirsbo wholesaler supply
houses have them they are an excellent resourse . I keep one on the truck."Yes. Radiant is the key word.Place the aluminum between plywood and flooring like it is in this application the and then see if it will cook your dinner.The aluminum is only used as a conductor in this application to help transfer and spread the heat from the tubing to the floor.Any radiating is done what is done from the flooring.
Me thinks thou dos't protest to much! Simple physics only works one way!
"Me thinks thou dos't protest to much! Simple physics only works one way!"Exactly.What is the physics of radiant energy heat transfer and the physics of conduction heat transfer?
The reflective or radiative properties of materials are determined by its emissivity.
<!----><!----> <!---->
1.0 being the perfect radiator and 0.0 being the perfect reflector. <!----><!---->
<!----><!----> <!---->
Bare aluminum, even oxidized, has only a nominal emissivity between 0.20-0.31 and thus would make at best a poor radiator.
<!----> <!---->
In order for the plate itself to reflect or radiate it would need to be decoupled from the mass of the floor. It is however an excellent conductor in the application – coupled to a wood surface having a plus 0.9 emissivity. <!----><!---->
___________________________________________________________<!----><!---->
<!----><!---->
This link and images come from our resource library and will “shed some light” on the topic of plates and orientation.<!----><!---->
http://www.ngml.ksu.edu/research/project_archive/project_archive_publications/ashrae/1036_final_report.pdf
View ImageRBean
web: http://www.healthyheating.com blog: Journal of Indoor Environmental Quality
In the fourth photo they show plates. For obvious reasons they didn't bother to test and show pictures of a test with plates installed on top of the tube, witch a certain individual suggested wouldn't make any difference. It also is appearent that the heat is radiating more at the tubing than the edges of the plate. This leads me to conclude the whole plate is not the same tempureture as a certain individual suggested it would be. The 100,000 BTU radiant tube, gas fired heater I installed today required 18 inches of clearance to combustibile on the back side of its inefficent aluminum reflecter and 74 inches to combustibiles in the front. Or maybe they put aluminum on there for an insulater!
Unfortunately, there is a shortage of researchers with dollars but infinite ways to assemble radiant systems (thus, the limited number of assemblies tested). <!----><!----><!---->
<!----> <!---->
There are however, mathematical means and the FEA software to play “what if”. The plate manufacturers would be able to provide the FEA models for most assemblies – I would venture to guess they already have this stuff on file. <!----><!---->
<!----> <!---->
So long as there is a temperature difference between the fluid in the pipes (or temperature of an electric element) and its surrounding mass there will be a difference in temperature in the "chain" so yes the whole plate is not the same temperature when it heating or cooling.
<!----> <!---->
The temperature profile at the surface (infrared footprint) will vary depending on the spacing and depth of the tubes and the conductivity of the assembly.
<!----> <!---->
We recommend locating and reading these three resources:<!----><!---->
<!----> <!---->
Depth Perception, by John Siegenthaler, P.E. <!----><!---->May 18, 2000<!----> The depth of<!----><!---->
hydronic tubing can affect the performance of a heated concrete slab. <!----><!---->
<!----> <!---->
Is 2 Foot Too Far? By John Siegenthaler P.E. , <!---->January 31, 2001<!---->, Some<!----><!---->
radiant installation can afford to push the envelope.<!----><!---->
<!----> <!---->
Impact of Surface Characteristics on Radiant Panel Output, University of Illinois at Urbana-Champaign, Department of Mechanical and Industrial Engineering, ASHRAE 876 TRP<!----><!---->
<!----> <!---->
Not sure what you mean by “inefficient aluminum reflector” or “aluminum on there for an insulater”.
That's a valid test, GaryGary. Your method #1 is exactly the right way to do this IMHO, especially the PEX-AL-PEX. You might even want to look into pre-made heavier extruded plates to see if they might fit in your budget. They will be even more efficient at emitting heat into your room. Especially with such low water temperatures.
Have you done a room-by-room heat loss analysis? What is your design day temp? How many heating degree days (HDD) in your climate? Do you have climate data indicating % of clear days? Ah, looking back now I see you are in Bozemen..should be just about the perfect climate for a solar system.
These are all things that are nice to know as you design your system.
What provisions, if any, have you made for back-up heat for extended periods of cloudy weather? Do you have a large storage tank that will store fluid at a higher temperature and then mix it down for your floor? Are you going to drain back the fluid from the collectors when the circulators are off to avoid freezing or flash-to-steam problems?
These are all things I've thought of when pondering whether or not to add solar to my current electric resistance/propane sourced system.
One thing for sure, don't try to economize on the plate coverage. Put them just as close together as you can without touching.
Edited 11/17/2006 7:35 am ET by johnnyd
Thanks all -- appreciate the help!Looks like method 1."Have you done a room-by-room heat loss analysis? What is your design day temp? How many heating degree days (HDD) in your climate? Do you have climate data indicating % of clear days? Ah, looking back now I see you are in Bozemen..should be just about the perfect climate for a solar system."I used the Ipex software to do the heat loss and loop sizing. For the time being, our existing forced air system will be the backup heat. At some time in the future, I plan to add a condensing boiler, and get rid of the forced air altogether. I sized the loops to be good for our full design temperature. Bozeman is pretty good on sun, but our house is about as bad a passive solar design as one could imagine -- my simulation says the collectors will be able to provide nearly half the heat."What provisions, if any, have you made for back-up heat for extended periods of cloudy weather? Do you have a large storage tank that will store fluid at a higher temperature and then mix it down for your floor? Are you going to drain back the fluid from the collectors when the circulators are off to avoid freezing or flash-to-steam problems?"The solar heating system I have put in has 240 sqft of collector and a 500 gallon hot water storage tank. Its a drainback system. Its described in way too much detail here:
http://www.builditsolar.com/Projects/SpaceHeating/SolarShed/solarshed.htm
(any comments on it would be appreciated)Thanks,Gary
Wow, sounds like you've really done your homework well. Good luck on the project, and let us know how it works out. We love pictures!