Hello all!! I am looking for options on my situation. I have installed the half inch pex tubing with the aluminum plates under my sub floor between joists. As per manufacturers I will require a minimum of 1 inch air space below the plates and then insulation. My issue is by the time I put the insulation under the floor and install my pot lights I will have to drop my ceiling by 2.5 inches. My question is, if I seal my house up with spray foam so its tight and I do not install any insulation under the floor between the plates and the drywall ceiling do you think i will be making a mistake?? Basically pex + aluminium plates + pot lights + drywall ( with no insulation) Looking for your opinions. Thank you very much! Ps I live in Canada

Heat is driven by temperature differential.

I you have more insulation above the heat source--sub-floor plus floor covering, you will drive heat to lower level through the drywall, which presents little resistance to heat transfer. What's more is the always greater heat load on the upper level compared to a basement level--typically triple--making the lower level warmer than the upper; in many cases unbearably so.

You must insulate under any sub-floor heating system.

Consider surface mounted light fixtures, or thinner insulation over the cans as I did in my own.

The one inch air space is more relevant in sub-floor suspended tube radiant systems where heating the air below the floor by convection is the primary means of initial heat transfer from the PEX. If you are using aluminum plates, the primary driver is conduction making an air gap less important. Having both in my system the practical outcome was not measurable. Much depends on the data you acquired in you room-by-room heat load analysis.

Yep, use IC light housings and just use thinner insulation over the cans...

Or you could just account for both the up and down heat loss in your design...build both a ceiling and a floor hydronic radiant heating system...

Or you could abandon the can lights and use some surface mount LED's. I have had very good luck with a light that will fit in a 4" J-box. My application was a retrofit. Not too difficult to install, a little tight depending on wire size (I think my fullest box was 3 14 ga. runs). Here is a link to the ones I used. Nice even light, have em on dimmers. Very satisfied so far. https://www.amazon.com/gp/product/B00X9F1ZB6/ref=oh_aui_detailpage_o01_s00?ie=UTF8&psc=1

Posted By icfbound on 17 Aug 2016 11:18 AM
Or you could just account for both the up and down heat loss in your design...build both a ceiling and a floor hydronic radiant heating system...

I have corrected for this common mistake many times, but can't find a hydronic design program that has a combined floor and ceiling from the same panel option... Fantasy I think.

A combined ceiling/floor HR emitter is easily designed using most HR design software including the free DIY HR design software on our website:

Borst Hydronic Radiant Floor Heating Design Software

Most software outputs the downward heat loss or back loss of the HR floor emitter. Normally, this downward heat loss or back loss is wasted heat, only contributes to the heating system inefficiency (e.g., the downward heat loss or back loss of a slab-on-grade HR emitter), and is something to be minimized by having very little R-value between the PEX and the living space and having a lot of R-value between the PEX and the outdoor exposure. However, if the HR emitter is between living spaces, this downward heat loss or back loss becomes a heat gain for the below living space and this can be easily accounted for when accomplishing a HR design. In fact, even if you are not designing a combined ceiling/floor HR emitter, this heat transfer into the other living spaces should be properly accounted for too. Some professional HR design software will accomplish this automatically (e.g., Avenir HeatCAD/LoopCAD) and identifies it as recovered back loss.

A combined ceiling/floor HR emitter can work well in an energy efficient building or a building in a mild climate. However, this approach doesn't tend to work well in buildings having high heating loads.

It’s certainly true that a lot of “pros” don’t fully understand or know how to use their HR design software and this commonly results in HR design mistakes.

Sure.

Let me know how that works out for you.

With one heat source and, effectively, two emitters, you have to have some experience to make it all work.

Wrightsoft will allow it. The experienced designer takes a pass.

Ok so it looks like I'm just going to install the foil bubble wrap 1" underneath the plates and skip the Roxul sound insulation under my floor. Anyone had experience with that? Thank all for your input

Well, given your current situation, my vote would be to place as much thermal insulation R-value as you can without having to alter your ceiling height. Who did your heat loss analysis, HR design, and sold you your HR components?

And why didn't this 'experienced designer' sort this out well in advance of installation? I wager this is another blue screw plate sale story...

I suspect there is only one schitzophranic on this post.

The insulation required below a suspended floor radiant panel is properly determined by the design air temperature of the floor below and the stack-up of the suspended floor above i.e. floor coverings.

Bubble foil is not insulation.

http://www.healthyheating.com/Page%2055/Page_55_o_bldg_sys.htm#.V7Wm1PkrJhE

Agreed Badger, bubble foil would be entirely inadequate.

Again,real thermal insulation is required. Since a combined ceiling/floor was not initially designed, the goal here now is to just make the ratio of the R-value above the PEX divided by the R-value below the PEX to approach zero thereby causing all the heat flow to move upward. We normally design so as to only allow 5-10% of the total heat generated to flow downward (back loss).

Icfbound, your comments are unprofessional and not helpful. While I am not a fan of plate systems in general because of their lower performance and higher cost, there are certainly applications (e.g., remodels) and people who can afford using plate systems to obtain HR heated floors.

One detail missing here Apollo1010 is how tall are the floor joists in this floor system? Bubblefoil is little to none in the insulation department IMO. Since the Joist bays are at least tall enough to accommodate light cans, just fill in the area around/between the lights and around the perimeter of the floor, not worrying about the small area on top of the light cans. That will be significantly better than bubble foil...

I have a problem with an installed radiant system where the power co., in order to purchase off peak power, required the tubes to be buried under 4" of concrete, and 1' of sand, tubes are set on 2" foam board. I have a pit sand with some small stones for my fill and have been told I used the wrong sand. I think a masons sand was the sand of choice. Any one have an idea to try to get the heat to the surface? Can I add water to the sand or some other liquid?

Jimmer, you might want to start a new post and NOT hijack Apollo1010's post... When you do, perhaps also post the BTU rating of your heat source, the diameter/length of your PEX circuits, your measured supply and return circuit temps, your measured circuit flow rates, and how long you have actually waited to achieve some slab surface temp at some existing indoor temp. You will certainly take a performance hit for having to transfer heat thru 1” sand and 4” of concrete, but you should eventually be able to heat the slab surface given that you have 2” of insulation below the slab...unless you not getting enough heat into the slab or you have a large amount of slab emitter back loss for some reason.

Hello Ronar,

The ceiling from basement floor to under ground floor is around 8' to under side of subfloor and joist spacing is 22" at 7.5" deep. The ground floor to underside of subfloor is 9'7.5" with 14" joist spacing. Both floors have plates running with pex at underside of joist bays.

But good point, I can install bubble wrap 1" below plates for air space, install cans and put insulation everywhere except for where the cans sit tight up to the underside of the bubble wrap.

Thoughts? And thanks all so much for the advice

Sorry forgot to mention my subfloor is fairly think as we put down new plywood over the old sub floor in some areas. Subfloor ranges from 3/4" to an inch think. Plus I will be putting engineered wood floors over that which are 9/16". So we are talking inch to an inch and a half over the plates

1.5” of wood is about R1.35. You will need about ten times this R-value below the PEX (i.e., R13.5) to have 90% of the total heat provided to this HR emitter flow upward (and 10% will flow downward). 3” of EPS insulation is about R13.5.

Well with 7.5" to play with, you should be able to make R13+ overall/average without too much trouble...