Radiant Ceilings Higher than 12' And Other Questions

Luke_I

Hello! I'm in the process of evaluating possible heating systems for a new-build house, and a radiant ceiling installation seems very attractive, but I'm wondering if I'm barking up the wrong tree, so I've got a few high-level questions... First, though, a bit of background:

I've got a 2500 sq. ft., 3 bedroom ranch laid out with a 40 ft. x 16 ft. open floor plan living/dining room, with a single-surface sloped ceiling (10 deg. slope, ~16 ft. ceiling height at the high end, 13 ft. at the low end), and large 225 sq. ft. fixed, double-pane fiberglass-frame floor-to-ceiling windows at each end, one of which is oriented to the south, for some nominal passive solar contribution. I will be building in central Connecticut. As would be expected, the heat loss for that one room is pretty spectacular; using a few calculators online, I'm getting between 33k and 45k BTU/h for that one room, with delta Ts of 60 and 80 deg. F, respectively. (Fortunately, the remaining rooms are more conventional, and thus are less of a concern at the moment.)

My original plan was to use (hydronic) baseboard heat throughout the house, with trench-style units in the main room to keep the walls unobstructed (and conventional/cheaper baseboard units everywhere else). This looks like it would work fine for the bedrooms, hallway, etc., but it is proving difficult to get enough linear feet of baseboard to actually meet the heating requirement for the main room, and frankly, I think ringing the room with trench heaters would look terrible anyway. The limited maximum temperature (and thus, BTU/h output) for radiant floor installations eliminated that as a viable option. However, a radiant ceiling installation running at ~100 deg. F seemed borderline plausible, from a simplistic BTU/h standpoint.

So, my first question is this: typically, I've seen a 12 ft. upper limit on ceiling height discussed. How borderline is this 12 ft. limit? Would that last 1-4 ft. really kill the performance/comfort levels, and how would the loss in performance be quantified (i.e. BTU/h doesn't cover it, since technically, the same amount of energy is being dumped into the control volume, but I imagine the perceived intensity at floor level is degraded, inverse-square law, and all?)

Second, the 10 deg. slope of the ceiling is oriented away from the seating area (i.e. couches are against the shorter wall); thus, there is a further intensity reduction. Given the shallow angle, I can't imagine this would have a severe negative impact on the installation? Having said that, how directional is a planar ceiling radiator?

Third (and lastly, for the time-being) was I too quick to discount the possibility of a radiant floor installation? While getting 47k BTU/h out of ~640 sq. ft. obviously isn't workable, are the heating requirements for a high-ceiling room practically reduced, due to having the favorable heat gradient that a radiant floor produces (i.e. not trying to heat the entire room volume to a steady-state condition, only providing a heat gradient shallow enough to have the first 5-6 feet of height be comfortable?)

Any thoughts, first-hand experience, etc. would be much appreciated!

Thanks!

Regards,
-Luke I.

STEVEusaPA

I've never heard of a 12ft ceiling height limit for radiant heat, sounds made up.
Are you sure about the heat loss for that room, seems high.
You could put radiant in the floor, ceiling and walls, and/or supplement with panel radiators.
I think I recall a thread/comment from Mark E. about 'washing' large window areas with heat, but I'll have to dig a little deeper to find that thread (or maybe Mark will chime in).
Radiant heat will heat the entire room with only a slight gradient from floor to ceiling. That is what provides the superior comfort.

Luke_I

I've seen multiple references to the 8-12 ft. 'limit', for instance:
http://www.zurn.com/media-library/web_documents/pdfs/zpm02101-pdf.aspx (Pg. 84)

https://www.pmmag.com/articles/100112-radiant-ceilings-offer-options-for-heritage-homes-and-other-buildings

https://forum.heatinghelp.com/discussion/87515/radiant-in-ceilings-instead-of-a-floor (4th post)

From a irradiance standpoint, it makes sense, I guess..? For a point source (perhaps not applicable in this case?) intensity will fall off as the square of the distance, so going from 12 ft. to 16 ft. roughly halves the power per unit area at floor level.

I'll be honest, I was pretty shocked at the heat loss estimate. I'll do the hand calcs tomorrow (snow day here in CT), and see how those compare to the online estimators. Still, that main room is basically a 40 ft. long, 16 ft. x 14 ft. tube, with glass ends... The windows will be decent (U = 0.26, thereabouts), but that's still only R-4, vs. a barely-to-code R-20 wall.

I'll look around for Mark's thread(s); in any case, from what I've read elsewhere, it's standard practice to 'wash' windows. That was going to be my original approach with the trench heaters.

I have considered radiant wall panels (+ ceiling + floor), but I'm trying to be mindful of cost, so I'm trying to avoid covering every available flat surface with Warmboard or similar, if possible. But, if push comes to shove, you're right, it definitely would get the BTU/hs required.

-Luke I.

Gordy

Radiant heat tends to travel in straight lines perpendicular from its emitting surface. That being said the 10 degree slope which isn’t to aggressive you may not get the full effect in the seating away from the slope.

Am i picturing the slope getting greater as it gets to the windows?

12’ isn’t outrageous. You may be looking at a little higher AWT.

Biggest thing is a well built panel detail. Give a serious look at Roth, warmboard r, or sun board. Something with a continuous aluminum sheet, and the ability to go 6” centers with tubing. This will drop your water temps.

I think I would double triple check that load for the room also.

Radiant floor coupled with it should meet the loads, but require a different water temp. Than the ceiling......floors being lower than ceiling.

You could if needed take the ceiling to say 115degrees verses the floor. What is the floor covering going to be. Something to think about.

Trench baseboard sounds like a dust catcher to me.

Luke_I

Straight lines in a direction normal to the emitting surface, I imagine?

12 ft. was being discussed as a soft limit I'd seen discussed; the design as it currently stands has 16 ft. at the high side, 13 ft. at the low side, so... A bit higher than 12 ft.

Yep, I'll let you know how my recalc comes out. Some quick partial checks suggest that the online numbers may be out-of-whack (serves me right for getting lazy).

Notionally, the floor would be engineered hardwood.

Yes, I expect you're right about the trenches. (Dust, crumbs, pet fur...) Easy enough to vacuum, though.

Gordy

50-70 btus a square foot is unbelievable. This is new construction.

My guess is half the low end probably less. Of not then window, and insulation upgrades should be considered. Insulation the gift that keeps on giving.

Use the slant fin heat loss calculator.

Luke_I

Gave the Slant Fin calculator a run:
80 deg. F delta T: 28.6k BTU/H
60 deg. F delta T: 21.5k BTU/H
50 deg. F delta T: 17.9k BTU/H

So, not quite half, but definitely a bit better than previously estimated, and back within the realm of radiant floors.

Exposed wall factor is 0.05, window factor is 0.26, floor over warmed space, infiltration factor is 0.018. Are any of these way off the mark for reasonably-insulated modern construction?

Also, while a floor is back in the running, I remain curious about the ceiling range [non-]issue; I like the idea heat transfer by radiation, rather than conduction/convection, as long as it would actually work.

Gordy

Curious what did you use for indoor temp, and outdoor design tem?

Jamie Hall

If it were mine, I'd do both the ceiling and the floor. Set the floor up with enough heat to be comfortable -- say shoot for 80 or maybe a bit less. Then make up the rest with the ceiling.

I'd consider trenched units but only under the windows. They will be there not so much to heat the room, but to kill the cold draught from the windows on chillier nights (like this one...).

Couple of points about the ceiling. It is a diffuse, extended surface. Therefore the inverse square law does not apply, and the felt radiation will fall off actually slightly less than linearly, rather than as the square of the distance. Much better! Also, since the radiating surface is diffuse, the angle of the ceiling is far less important.

Luke_I

Inside temperature of 70 deg., with -10 deg., 10 deg., and 20 deg. outside temperatures.

Yeah, my original concept was trenches under the windows, for exactly that reason. However, if I can avoid them, I wouldn't be upset.

Diffuse radiator: that's what I was wondering, regarding directionality, etc. Good to know. Still, if I had to pick one (money is money, end of the day), it sounds like a radiant floor would be the safer bet, in terms of comfort? (Not having to worry about possibly marginal felt warmth, etc.)

Gordy

Is -10 design temp for your area? You use the coldest day in a normal winter not the record coldest.

70is a good indoor temp.

Gordy

I’m partial to ceilings. Higher outputs. No furniture to block output. Something to think about.

STEVEusaPA

@gordy What brand panels are they on the ceiling?

Luke_I

Outside temperatures: fair enough, -10 deg. is excessively low, though we have been below zero at night several times over the last week (with highs of 10 deg. F during the day), so around 21k BTU/H probably isn't unrealistic.

Edit: just watching the weather forecast, and we're supposed to get down to -12 deg. F at night this weekend... Having a little extra capacity to handle these outliers probably isn't a bad idea?

Gordy

Roth

Gordy

If you have never experienced radiant ceilings it’s a great feeling. Especially when sitting, or laying on furniture because the radiant energy still hits your body. Well like the sun.

Luke_I

Gordy: yes, I've read up on the various pros of ceiling-mounted installations, and it's a persuasive list; but ideally, I'd like to hear some first-hand opinions on how the system 'feels' on a higher (14-16 ft.) ceiling, with reasonably normal water temperatures (<110 deg. F).

Gordy

https://www.uponorpro.com/~/media/extranet/files/cdam/cdam_manual_7thed_0711_ch9.ashx?version=072920111126

Keep in mind on the 12’ note it is for their system panel design.

Leon82

7 degrees was the New Haven design day when I was calculating things

Luke_I

Gotcha, at 7 deg., the estimate is 22.5k BTU/H.

Gordy

The house I use to own had radiant ceilings from the 50’s. Design day is -4 in my area. However through the decades that house has been in temperatures of -27 F, and able to hold setpoint in the mid 70s. Initial attic insulation was 5” of vermiculite...........

Luke_I

So, I dusted off my college textbook on heat transfer, and it looks like the intensity difference between 12 ft. and 16 ft. ceiling height (for direct radiant emission, not counting wall bounce, etc.) is about 15-18%, which agrees with Jamie's 'bit-less-than-linear' estimate.

So, since either floor or ceiling heat appears plausible, my next question is whether there are installation cost differences between the two?

Gordy

According to your calculations then what is the difference from 8' to 12' ceilings? Another 12-18% less?


If you use something like pre fab panels Roth, WB r, so your cost would be the same since same panel can be used on floor also.

Luke_I

Using the view factor plot (see below) for two finite parallel plates, 40 ft. x 16 ft., with spacing of 8, 12 and 16 ft., I get the following view factors.

(X = 40 ft., Y = 16 ft., L is varied as described)
8 ft.: F = 0.60
12 ft.: F = 0.38
16 ft.: F = 0.32



Figure credit: Incropera, Dewitt, Bergman, Lavine, Fundamentals of Heat and Mass Transfer, 6th Ed., John Wiley & Sons, 2007.

So, no, there's a much larger difference between 8 ft. and 12 ft., compared to 12 ft. and 16 ft.. Of course, this is an extremely simplified case where any radiant energy that doesn't directly fall on the floor is lost to the environment (i.e. no walls).

That's what I figured, cost-wise. Labor rates aside, I assume that using good-quality Omega-section reflectors, but home-brewing the supporting panels using plywood and foam for a ceiling installation is inadvisable?

Gordy

What is finished ceiling? Drywall?

Luke_I

Yes, half inch drywall would be the plan.

Harvey Ramer

You could install radiant windows. That would get your heat load for the room down to a manageable number.

Gordy

They would be a nice asset. Not to wallet freindly though. Depending on electric rates.

26 btus SF isn't terrible. Decent window treatments do a lot on those cold nights to curb sucking the heat off the radiant components.

Luke_I

Yes, I plan on using cellular shades to help manage the nighttime radiant losses. Radiant windows might be a little much up front, and the electricity rates around here are pretty bad, so it wouldn't get any better in the long run.