GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Ceiling mounted radiators

rockies63 | Posted in General Questions on

I have very little wall space throughout my cabin due to built-in seating and cabinetry. All the heat will be supplied by an exterior wood fired boiler via piping. I plan on using low temperature radiators throughout (the cabin is only 540 sq’ – a main room, a small bedroom, entry/mudroom and a bathroom).

The bathroom can be heated with a radiator towel warmer over the toilet but the other rooms have no wall space. I was thinking of a ceiling mounted radiator in the entry mudroom and the bedroom but in the main room (221 sq’ with a 4:12 shed roof) there is room between 8′ to 10′ off the floor to install wall radiators around the room (essentially at the “loft” height).

Will radiators mounted 8′ above the floor result in too cool  a temperature at the floor level? The other option is to use trench radiators around the perimeter of the main room.

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. Expert Member
    Michael Maines | | #1

    Scott, if you have a decently insulated and air-sealed building envelope, ceiling-mounted radiant heaters should be fine. But if it's poorly insulated or air-sealed, you will probably have comfort issues--your head will feel hot and your feet cold.

  2. m854 | | #2

    You could consider kickspace heaters. They can fit under cabinets and under built-in seating. They have a water to air heat exchanger and a small fan that automatically comes on when it senses warm water in the hydronic loop. I'm not sure if they make a low temperature version, or if you could use higher temperature water.

  3. Expert Member
    Dana Dorsett | | #3

    Low temp radiant ceilings (using PEX and sheet aluminum heat spreaders) are almost as cushy as radiant floors, and will in fact heat exposed floor directly via radiation.

    Radiant cooling/heating radiators designed for ceiling mount are also available, but are more expensive than a PEX & heat spreader solution (and make you feel "right at home", if you basically live in an office. :- ) ):

    Radiant walls are pretty good too (if you promise not to drill through the PEX for hangin' the big screen TV :-) ):

    Installing 1' tall euro-style convecting radiators with the tops a foot below the 10' ceiling level can work, but it requires higher water temperatures, and would result in higher temperature stratification than a radiant ceiling solution. Convecting radiators installed tight to the ceiling would under-perform, since most of the heat transfer is still via convection with that type of radiation.

    Low temp kickspace heaters exist, but that solution is a last-resort band-aid solution, with lower reliability, lower comfort, higher noise than any radiator solution.

  4. Jon_R | | #4

    A small passive radiator will cause more stratification (ie, colder floors for the same average air temperature) than something that provides more mixing (like a fan coil). How much is a good question and even with that, your "too cool" needs quantification.

    Obviously a radiant floor will produce the warmest floors. Unless you have some special situation, response time isn't an issue.

    1. Expert Member
      Dana Dorsett | | #5

      >"A small passive radiator will cause more stratification..."

      Which is why it's preferable to have a LARGE passive radiator (such as the whole ceiling).

      I've experienced a house that had radiant floor in one section and radiant ceiling in another section of a large open floor plan space. While it's dead-easy to tell when passing from one section to the other (especially with shoes off), the comfort level in the radiant ceiling half was quite good, better than any small wall-radiator solution, comparable to large/very-large low temp wall radiator solutions.

  5. rockies63 | | #6

    Thanks everyone!

    Dana, the pex and aluminum spreader solution looks very similar to a radiant floor application. If I went that route I might as well run the pex under the floor but wouldn't that make the system have a slow response time (heating up the entire floor)?

    My cabin's floor system is rather unique since it will be built with a 12" thick SIP panel resting on top of piers and foundation beams and then a 7 1/2" deep false floor on top of the panel in which to run my other option might be to use trench radiators around the main room.

    Since I don't want to use any electricity for fans I was looking at this model.

    However, if I use radiators installed at an 8' height idea then I would use something like this.

    1. Expert Member
      Dana Dorsett | | #9

      >"Dana, the pex and aluminum spreader solution looks very similar to a radiant floor application. If I went that route I might as well run the pex under the floor but wouldn't that make the system have a slow response time (heating up the entire floor)?"

      Response time is one major difference between radiant floors and radiant ceilings. The thermal mass and R-value of ceiling gypsum is pretty small compared to subfloor + finish flooring- it's pretty fast.

      Slow response times for radiant floors is mostly attributed to radiant slabs and under the sub-floor systems. With a SIP floor those types of radiant floors aren't really relevant (though I suppose you COULD go out of your way to screw it up installing an under the subfloor system in your utility-chase framing.) Most above-the-subfloor approaches to radiant floors that are quite responsive, almost as responsive as radiant ceilings and radiators, eg WarmBoard, Roth, or at the lower cost end, RHT. Any of those should work just fine with a SIP floor. But they tend to be more expensive than radiant ceilings (with the possible exception of RHT).

      Trench radiators suck- I mean REALLY suck! There is no way to get them to convect well at low temperatures, and they fill up with grit, dog hair & other crud fairly quickly, reducing the already marginal air flow through the fins. (The fan-assisted versions are really no better than kickspace heaters, and need more frequent cleaning.) Cutting space for trench heaters into a 12" SIP could also add some complications and cut into the thermal envelope's performance.

      But perhaps you're already way out over your skis on this- have you done any sort of heat load calculation to determine just how much radiator is needed, and at what water temperature?

  6. zenhomes | | #7

    Entire ceilings made into radiant heating surface work very well, especially if using low temperature heating source. Using the right controls to manage dewpoint these ceilings work amazingly well for radiant cooling. Cool Ceilings

  7. Expert Member
    Akos | | #8

    Radiant and fast don't go hand in hand. If the outdoor boiler is similar to the typical setup up north here, speed of the radiators is not an issue. Any radiant device will get the place up to temperature by the time your boiler is warm.

    If you want quick, go for either electric baseboards or a heat pump. If you don't want electric, can't beat a through the wall direct went propane heater.

    If your finished floor is not in, getting a floor heat setup in there would not take too long. I like the Ultra Fin product better than the heat plates. Much quicker to install, a 500sqft cabin is maybe a day and a half job. If you are DIYing it, it is probably the cheapest solution after hydronic baseboards.

  8. Expert Member
    Dana Dorsett | | #10

    >"If your finished floor is not in, getting a floor heat setup in there would not take too long. I like the Ultra Fin product better than the heat plates. "
    >"My cabin's floor system is rather unique since it will be built with a 12" thick SIP panel resting on top of piers and foundation beams and then a 7 1/2" deep false floor on top of the panel in which to run plumbing."

    I'm not convinced Ultra-Fin is going to work here, and it would have to be installed before the subfloor goes down, since there's a fat SIP blocking access from below. In either case, Ultra-Fin isn't nearly as responsive as above the subfloor or radiant ceiling solutions (though it is comparatively quite inexpensive and installs very quickly.) There is barely enough vertical clearance to meet the installation spec, and it would be hard to thermally isolate the cold water plumbing from the heating system with that approach, creating a potential legionella incubator.

  9. vashonz | | #11

    Are you committed to hydronic?

    We saw electric infrared radiant heat panels at a homeshow earlier this year. Looking at it as an option for our bathroom. My limited experience (at the homeshow, nothing real world), was that it felt nice, warmed the people and objects, which then warmed the air.

    1. Expert Member
      Dana Dorsett | | #12

      The efficiency and operating cost of electric infrared is usually much higher than hydronic systems, especially when compared to the operating cost of an outdoor wood boiler burning scrap wood.

  10. rockies63 | | #13

    The cabin is off the grid up I the Rocky Mountains of BC Canada. There will not be a wood stove inside so I'm using a Polar outdoor G2 wood fired boiler for both space heating and DHW.

    Being off grid I only want to operate small pumps to move the water around so no fans or electric heating. Dana, I suppose the above subfloor radiant system is the best option if your analysis of trench radiators is correct (and I'm sure it is). BTW, if I had used them the trench radiators would have been cut into the false floor system, not the SIP panel.

    However, does anyone know of ceiling radiant systems that come as an already complete box that could be attached to the ceiling? In the main room there are ceiling beams running from eave to ridge under the sloped shed roof (the beams are spaced 3' apart) so radiant ceiling boxes could run between them but in the mudroom/entry, bathroom and bedroom there are flat ceilings with storage spaces above so I'm hoping for radiant ceiling boxes that could be recessed into the ceiling to help hide them. If not, I guess the solution would be to lay out pex piping on the ceiling like you would for a floor system.

    Finally, there will be a backup through the wall direct vent propane heater for emergencies or for times when I'm away for long periods.

  11. Expert Member
    Akos | | #14

    All hydronic celing/wall units I've seen were site built. The only pre-made panels are electric. There are some commercial products out there, but I think the cost would be much more than site building with heat spreaders:

    Before going too far on specing the heat surface, I would do a rough heat load calc. Judging by the 12" SIP panel, I'm guessing the place is well insulated. This means that with a floor heat setup, you might only have to heat 1/2 the floor area or part of the ceiling, which is a big cost save if you are going with any of the pre-made panels.

    Since you are running off a wood boiler, there is no need for low temperature heat. For the ceiling panels, you can reduce the area you need by a fair bit by bumping it up.

  12. rockies63 | | #15

    I found this article printed in a 2014 Plumbing and Mechanical magazine about radiant ceiling panel systems. It basically discusses site building the system and then covering it with drywall. They use a rather time consuming system of cutting foil faced rigid insulation panels into strips, gluing it to the ceiling and then installing metal radiator plates and pex piping.

    I wonder if you could use Warmboard radiant tubing panels on the ceiling instead of making your own insulated ones?

    I must admit that I hadn't considered a radiant ceiling simply because I assumed it would have the same slow response times as a traditional radiant floor system. My 2nd choice had been baseboard or wall radiators but ceiling radiators are moving into 1st place.

    My one concern with the ceiling system is placing it under a SIP roof panel. I know the seams between SIP panels need to be sealed with tape and foam but I'm wondering if I should add a layer of drywall onto the underside of the SIP panel to help seal the surface, then the Warmboard, then another layer of drywall?

    This is another brand of installation panel that can be used on the ceiling.

  13. rockies63 | | #16
  14. Expert Member
    Akos | | #17

    I've only done floor heat, so not sure how much trust to put into my opinion. Sealing the SIPs is a whole other problem, there are some articles on this site on how to deal with it, adding in radiant panels won't change that much.

    With celings, you are not limited by temperature the same way, no chance of burning toes or warping flooring. If it is out of reach, I can't see why you can't operate it 160F or 170F. Assuming a 20F delta, at 170, you can get 2800 BTU out of a 4x8 sheet. My guess for a well insulated cottage that size, you need maybe 3 panels. These would feel very hot on full tilt, so I would not put it close to where anybody is sitting.

    You can definitely make the the panels out of any of the warmboard style panels. DIY it is also not that hard. Similar to the foam link you sent, but I would use plywood strips between pipe runs. I would also not bother with anything at the end of the runs and turns. It is a ceiling after all, nobody will be walking on it, as long as you are supporting the drywall at 16"OC should be fine.

    The good thing with a high temp setup is you can get away with a single delta P ECM pump for your entire heat/DHW. The only electronics you need is to turn the pump off when the boiler is not running.

    boiler->indirect->ceiling panel->pump. Put the ceiling panels on a TRV and a differential bypass around them (ie TRVs closed, the pump is just circulating water through the indirect). No need for any controls outside of that.

  15. rockies63 | | #18

    What I found fascinating about the article I included (in comment #16) is that the floor of a radiant hydronic system will actually be warmer using less piping. In the article he used this example: Imagine a room with a design heating load of 3,000 Btu/hr., and a corresponding indoor temperature of 70 degrees F. The room measures 20 feet by 30 feet. His calculation says that the floor surface temperature will be about 75 degrees and the room air temperature will be around 70 degrees with an upward heat flux of 10 btu/hr/ft2.

    Then he states: "If the size of the radiant panel in this example was cut in half, the necessary upward heat flux would double from 10 to 20 Btu/hr./ft2. This would bring the average floor surface temperature on a design day from 75 degrees F up to 80 degrees F. Although still a tad low, such a temperature may appease those looking for barefoot-friendly floors. Reducing the panel area to one-third of the room’s floor area would boost the average floor surface temperature to about 85 degrees F, a recommended maximum for floors in which there is prolonged foot contact".

    For a radiant ceiling he writes: "Another approach is radiant wall or ceiling heating. Since occupants don’t rest their feet on these surfaces, neither is constrained to a maximum average surface temperature of 85 degrees F. Instead, a practical surface temperature limitation for a gypsum surface is about 120 degrees F. This is based on not causing long-term degradation of the joint compound".

    "A radiant wall with an average surface temperature of 100 degrees F, releasing heat into a 70-degree F room, would yield an output of 1.8 x (100-70) = 54 Btu/hr/ft2. A radiant ceiling operating under the same conditions would yield 1.6 x (100-70) = 48 Btu/hr."

    This is the first time I've read anything that says that decreasing the heating surface results in a warmer room. With such an open area to work with I'd say a radiant ceiling is the best choice.

    1. Expert Member
      Akos | | #21

      Running a smaller heated surface is very common in low load construction especially for floor heat.

      For the floor to feel warm you want it above 80F, so you want to size your radiant surface area to provide that. Covering the entire house in radiant floor is counterproductive, it cost more and you also don't get the warm toes feeling, which is one of the main reasons for heated floors.

      This doesn't make the room hotter or get more heat into the room, it just gives the perception of the room being hotter.

      I didn't know about the temperature limit of drywall, now it makes sense why people recommend not to build pot light covers out of it. I guess if you want a high temp radiant panel you have to stick to cement board, tile, wood or metal.

      With the 120F limit, drywall radiant panel doesn't make much sense for ceilings, you are much better to go with walls as you get an extra 50% heat output out of them.

      There is a lot of good information out there (pmmag is defiantly a good source), just be careful with your design. Most people think of hydronic and imagine a panel wall filled with pumps, valves and controllers, you pretty much never want that. Stick to simple, check out:

      For example, for my own home I have 5 programmable thermostats for each zone. They are always running in fixed setpoint, I could have saved a fair bit by running some zones with TRVs and consolidating some under 1 or 2 thermostats.

  16. rockies63 | | #19

    Akos, if you have an interest in hydronic heating systems you should look through the articles on their site.

    I did a search for systems using wood fired boilers and also for solar water heating combination systems and found a lot of good information. They take you through the whole design and setup process and you can also search for common mistakes made when installing a system and how to fix them.

    Another site mentioned in one of their articles was:

    Specifically the forum called "The Boiler Room".

  17. DunSpannerin | | #20

    Try Modern hydronics done right volume-4 by the master himself John Siegenthaler at...

    or for all his articles on this site, try....

  18. rockies63 | | #22

    Akos, in the first article I mentioned in comment #15 it states:

    For example, a ceiling operating at an average surface temperature of 102° releases approximately 55 Btu/hr./ft2 into a room maintained at 68°. This is almost 60% more heat output than a radiant floor with a mean surface temperature limit of 85°.

    You write: "With the 120F limit, drywall radiant panel doesn't make much sense for ceilings, you are much better to go with walls as you get an extra 50% heat output out of them".

    However, further along in the article on partial coverage the author states:

    "A radiant wall with an average surface temperature of 100 degrees F, releasing heat into a 70-degree F room, would yield an output of 1.8 x (100-70) = 54 Btu/hr/ft2. A radiant ceiling operating under the same conditions would yield 1.6 x (100-70) = 48 Btu/hr."

    It would seem that the ceiling is right in league with the wall for producing heat. The main advantage to the ceiling is that you never have to worry about where to hang a shelf or picture.

    1. Expert Member
      Akos | | #26

      Working from the Uponor radiant design manual ( )
      • Radiant floor thermal transfer
      coefficient = 2.0 BTU/h/ft2/°F
      • Radiant wall thermal transfer
      coefficient = 1.4 BTU/h/ft2/°F
      • Radiant ceiling thermal transfer
      coefficient = 1.1 BTU/h/ft2/°F

      For a wall, running at the same temperature you get more BTU out of the same area than a ceiling (I remembered incorrectly 50%, but see it is 30%). There are lot of issues with walls and nails, so ceiling does have some benefits. Floor heat will always need larger area because of the lower temperature limits.

      Overall, I would just look at which one is the simplest for you to install and gives you sufficient heat. In a well insulated house, they all work and work well. I'm partial to heated floors especially for any tiled surface (entrance, bath, kitchen).

  19. rockies63 | | #24

    In reading through the articles Mr. Siegenthaler suggests adding radiant tubing into the walls around a shower, in the ceiling over a tub (especially if it has windows around it) and in the low wall fronting a kitchen island that has an overhang and bar stools in front of it. He cautions that radiant tubing in exterior walls will have to have extra insulation in them to prevent the heat from migrating outwards.

    A neat trick he did in his own home was run a radiant pipe underneath the granite countertops in his kitchen.

  20. Jon_R | | #25

    I think that radiant interior walls are underutilized. Perhaps skipping the picture hanging height. I'd rather not see any radiators or vents.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |