Theoretical radiant output

dcwittlo

Where can I find a more complete formula for calculating radiant floor output which accurately accounts for tube spacing, heat spreader, water temperature, flow rate and floor covering R-value?

Jamie Hall

A single formula is not going to do it. That is not to say that the temperature of the radiating surface (the floor surface exposed to space, which is the governing variable in heat transfer to the space) can't be modelled — it can, and all the items you mention factor into the models. Several firms have the capability of doing this more or less accurately for you — at least two pop up on The Wall from time to time and may see this thread.

That said, a calculation which comes within 10 percent is going to be the best your can hope for in a real world situation.

hot_rod

Any radiant panel output is based on the floor surface temperature and the ambient air temperature of the space.

In a residence you want to keep the floor surface below 82° or it becomes uncomfortably warm to bare feet.

So here is some math for various radiant panels, floor, wall and ceiling. From Caleffi Idronics 25

So for a room at 70° ambient with average floor surface of 82, 82-70 X 2 = 24 btu/ sq ft output.

Often when you see higher floor output it is because the room temperature is lower. At 65 ambient

82-65= 17X 2= 34 btu/sq ft. That may be fine for a shop, but a little cold for me at 65°

Jamie Hall

Thank you, as always, @hot_rod . For @dcwittlo , that is exactly the kind of approximating approach to which I was referring. As the reference notes — and I noted — going from the variables which you mention to the heat output can also be estimated using various sources — but to get to anything much better than 105 you are going to have to go to finite element heat transfer modelling. And you don't want to do that…

hot_rod

Under a wood floor the subfloor and any final coverings subtract out.

From Uponor

Thjerse infrared show the challenge getting the entire surface a consistent temperature. From left copper tube in 4" extruded aluminum plates

Suspended tube

Staple up EPDM

Pex in 4" extruded plates

Warmboard.

The back 1/2 of each sheet has berber carpet, no pad

The RadPad calculator is another way to run some output options.

I do have some FEA models also, but the story is the same.

mattmia2

what are the 2 cooler circles in the warmboard?

dcwittlo

OK, now we are getting somewhere. Next part of the calculation is: How many BTU/hr/ft^2 if one is given the spacing of PEX tubing, flow rate thru tubing, supplied water temp of tubing and maybe the width or area of the aluminum heat spreader. Does it much matter how thick the aluminum is?

dcwittlo

Sorry, my last post was a little late. I think you already answered these questions.

hot_rod

https://forum.heatinghelp.com/discussion/comment/1898868#Comment_1898868

I routered across the sheet to make a return bend for another project.

It clearly shows the conductivity of the aluminum

hot_rod

The grip on the Pex is more important than the thickness of aluminum plates. The extruded plates grip so well that you need to tap the tube in.
The thin flashing thickness plates can make some noise as they warp and the tube slides. A creaking noise from tube movement, an “oil can” noise as the plates expand and contract. Thin plates work if you lock them between layers of wood.

Another key is to use a boiler or mixing device with outdoor reset so the plates don’t see wide temperature swings. Constant circulation is ideal.

DCContrarian

@hot_rod: Thanks for posting those photos, very interesting. Is the temperature scale the same for all of them? Because I'd expect the Warmboard to be noticeably warmer and evener than the plates, and it doesn't look that way. I'd also expect the copper pipe with plates to out-perform PEX, but that doesn't seem to be the case either.

Interesting, it looks like the carpet does more to reduce output on the Warmboard than on the plates.

hot_rod

https://forum.heatinghelp.com/discussion/comment/1898876#Comment_1898876

What you see with the Warmboard may be the 12” spacing. All the others are 8” on center in a 14” joist bay, so the heat stripping is more pronounced.

Same SWT, same flow, same room ambient temperature.

Notice that the carpeted back half reduces the surface temperature, but it does also spread the heat across the surface a bit also.

The most consistent temperature spread I witnessed was the early SolaRoll mat systems. Both the width of the mat and the counterflow flow path with a hot supply next to a cooler return across the mat did the job well.
Dealing with 5/16” non barrier tube was another story,