GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter 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

How to employ a mixing valve in a closed loop radiant heat system

Hotsauce1183 | Posted in Energy Efficiency and Durability on

We are just wrapping up the new roof on my project house and now its [past] time to turn our attention to HVAC – or, more specifically this time of year, Heating! This 1975 house came with radiant heat ceilings as the only source of heat. Recently we have been having night time lows in the teens and daytime highs here in Northern Nevada zone 5 at 5000 feet are usually in the 40’s. I have done some preliminary calcs on heat gain/loss, and plan on doing a full Manual J model before laying out any cash for components.

My living room is 20×22 with a southern exposure which is the dominant direction the wind comes from. I am seriously considering doing both in the floor PEX supplemented with Runtal panel radiators or slant/finn baseboard. The thing I havent been able to figure out and havent found any examples of is how would you employ a mixing valve in a closed loop system so it can send optimum 180F water to the radiant panels and still be able to send ~125F water to the PEX portion of the system?

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. GBA Editor
    Martin Holladay | | #1
  2. Expert Member
    Dana Dorsett | | #2

    180F isn't an optimum temperature for ANYTHING (except a severely under-radiated heating system design :-) ), and it guarantees that the water returning to the boiler is going to be far too high to achieve. Panel radiators usually have specified output at a couple of different temperatures typically 180F and 140F, but the output is fairly linear with temperature down to water temperatures of 90F or so.

    Fin-tube convectors (Slant/Fin included) can be plotted linearly down to about 120F or so, but the convection height is so low that it gets pretty squishy at average water temps much below that, which can lead to paradoxical results at low load when tweaking in outdoor reset curves on the boiler controls.

    Since you're still in the design phase, there is no reason you can't design it as a single temperature system and maximizing efficiency rather than running the boiler at a higher than needed temperature to be able to serve under-radiates zones.

    Slant Fin publishes output charts at different water temperatures, eg:

    You'll notice that the output a 130F is roughly half what it is at 180F, but at 140F it's about 55-60% of the 180F output:

    Panel radiators usually publish just a pair of points (and sometimes only one), eg:

    You'll note that at 140F the output is also 55-60% of what it is at 180F. This can be extrapolated linearly to even lower water temperatures.

    If you design your radation to cover the 99% design condition load with 140F water or less, you will usually be in condensing mode for over 95% of the time once you dial in the outdoor reset curve, and meet or beat the manufacturer's AFUE numbers.

  3. Jon_R | | #3

    A radiant ceiling isn't ideal, but if it's already there, I'd use it as the supplemental radiator. Avoid high water temperatures when you can.

  4. Expert Member
    Dana Dorsett | | #4

    Most circa 1975 radiant ceilings were resistance electric, and pretty expensive to operate, sometimes difficult or impossible to fix. I was assuming his was one of those...

    If you want to make the whole thing a fixed temp 125F, if you don't have better documentation on the panel radiators, assume it's output with 125F water will be ~35% of it's 180F rating, giver or take a couple percent. That means it's 3x the amount of radiation, but you probably don't need very much anyway.

    To be sure, sitting next to a panel radiator running 125F water is a heluva lot more comfortable than one running 180F water. If the cost of the panel radiators is an issue (isn't it always?) Myson has some pretty decent lower priced "contractor" CV21 & CV22 or RCV21 & RCV22 models along with their more upscale stuff, priced WAY below Runtal UF-2 & UF-3 at equivalent output. It's not as pretty or thin as Runtal but it's not particularly ugly either. The xx21s stick out about 4" from the wall, the xx22s ~5" (but have more output per running foot), compared to the sub-2" of the UF-x Runtal.

    The RCV21/22 have a similar look to a Runtal, but it's 2x/3x as deep, and between a UF-2 and UF-3 in height.

  5. Hotsauce1183 | | #5

    Much thanks to Dana, Jon, and Martin. Your replies are very informative and educational as always. I kind of suspected that trying to do a dual temp system would be self defeating. On the other hand, how about running 140F water through the panel radiators first and from there into the floor heat? It seems like the panels would remove just enough heat from the water to make it more reasonable for the radient floor? I think that my target for return temps back to the heater is in the 95F to 105F range for best efficiency, Right?

  6. charlie_sullivan | | #6

    The series panel-->floor approach sounds appealing in the abstract, but if you consider what happens in different weather conditions, when the circulator pump short cycles, it would be extremely difficult to keep the same balance between the different sources. I suppose it would be possible with the right reset curve, but it might take a lot of experimentation before you got something to work,

    It seems unlikely to me that you'd need both the radiant floor and the additional emitters. And if you do, it might be worth investing in envelope upgrades first.

  7. Expert Member
    Dana Dorsett | | #7

    If there is a lot of window area with 1975 vintage thermo-panes or (gulp) single pane glass the heat load could easily be enough to require both radiant floor and convecting panel baseboards, but it's silly to be forcing yourself into a dual-temp system unless there are no better options. The BTU/hr per foot of Myson CV22 & RCV22s is over 300 BTU/ft @ 125F, which is pretty substantial. At 140F it's over 500 BTU/ft-hr. Going single (and lower) temp might cost you a few hundred on a large room like that, but it won't cost you a thousand, even if you were heating only with the convecting basboard radiators. (With Runtal it would be a much larger cost, even at 140F.) As a rule lower temp panel radiators are more comfortable than running them at higher temperature.

    Some opt to run baseboard as a second stage for their radiant floors rather than in parallel or series. That way whenever the room temp is below the programmed offset from the set point, the baseboards kick in and heat it up quickly, whereas most of the time the floor is carrying the load on it's own. That also allows more use of overnight setback- the room comes up to temp quickly- more quickly than the floor temp, but once it's at temp the floors provide the lion's share of the heat, for higher comfort.

    But seriously, you are getting WAY ahead of yourself on the design particulars. Before you add any complexity (dual temp or 2 stage), run the Manual-J, and tell us how many BTU/hr per square foot of available floor you're dealing with, and what the BTU shortfall is (if any). If you already have the preliminary calculations, c'mon now, don't be shy- share!

  8. Hotsauce1183 | | #8

    Dana: The views here in this valley are spectacular so the rooms facing the best views, which are to the south, practically have glass that is floor to ceiling. In adition to being 1975 double pane they were, probably a long time ago, argon filled. At this elevation (~5000') most older double panes with argon suffered seal failure and this is the case with this home. It is my plan to:

    Replace all of the windows with high quality windows designed for higher elevations.
    Raise the bottom of the windows ~2' so they are 4' high instead of 6'.
    Pay particular attention to installation detail such as window sealing and flashing.
    Increase the insulation of all the south facing walls by removing and replacing the UV burnt and weathered siding and adding at least two layers of 1.5" polyiso sheets under the siding to help compensate for 2x4 studs with inadequate fiberglass batts.
    I will probably also replace the bats either with newer and more effective fiberglass or else using another material like denim or cellulose.

    When I bought the house it was my intention to have all of the insulation and heating issues resolved by now but here it is in December already and its getting hard to do things in the right order when your morning shower is in a bathroom that is in the 40's. Of course we are running spot heaters all over the house and using the radiant ceiling heat in the rooms we occupy most but I shudder to consider what my electrical bill is going to be when I get that first bill that covers a full month of high heating requirements every day.

    The radiant ceiling has been more effective than I thought it would be, but only in the smaller rooms with less blown glass seals. In the larger rooms, with the most glass, it cant keep up with the heat loss and we have to supplement with whatever is available. I have been taking readings in the attic with an infrared thermometer and I was surprised that there was only one or two degrees difference in the temperature of the insulation over spaces heated with the radiant ceiling heat versus areas that werent being heated at all. Then I dug down into the insulation just a couple of inches and found a 6 to 8 degree difference. This convinces me that the insulation is absorbing a lot of the heat that was originally intended to be directed downwards. I am going to the big orange or blue box store in the morning for a couple of sheets of polyiso with the intention of completely removing the cellulose insulation from two or three rafter bays and replacing that with a double layer of foil faced polyiso. Then I will repeat the observation and see if the ceiling temps under those bays are higher or not. BTW I am also using ceiling fans in those rooms to help harvest the radiant heat off of the ceilings. I have also found that in rooms without a ceiling fan the temperatures of the ceiling are 4 - 8 degrees higher as is the temperature of the insulation above in those areas. This is proof that there is a lot of [fully expected] heat loss when using this system. I would be interested to hear from others who are living with radiant heat ceilings and I will be happy to continue to report how we are dealing with this and how our heating and insulation project is progressing.

  9. Dana1 | | #9

    Insulation does not absorb heat- open blown cellulose has fairly little thermal mass. It is normal to have a gradient in temperature through any insulating layer, and it only slows, does not direct heat. But even at low density celluose is fairly air retardent.

    Polyiso has about 1.7x as much R-value as cellulose at any given thickness, but if it is thermally bridged by rafters (which have only about 1/3 the R value of cellulose) the increase in overall benefit isn't nearly 2x. At a given center-cavity R-value the thermal bridging of the rafters is higher with polyiso due to the lower R-value through much shorter path through the themally bridging wood.

    Pointing an infrared thermometer at a foil facer will give you a dramatically lower (and false) reading, due to the very low infrared emissivty of bare aluminum.

    Ceiling fans only decrease comfort due to the wind chill effect. This means it takes a higher room temperature to be comfortable. The additional heat loss of an 8F higher temp-sealing through the insulated R10+ ceiling is probably a lot less than the heat loss through the R2 glass in contact with 2F higher average room air temperature. The higher temperature ceiling radiates more heat directly onto the humans, which makes it more comfortable, making up for the very low radiant temperature of the cold glass.

  10. Hotsauce1183 | | #10

    In the interest of brevity I did leave out a lot of details. Of course we dont sit around in the wintertime with the ceiling fans roaring away. We typically turn down the thermostats for the radient ceilings at night to the low 60's. The next morning I usually get up around 5:30 or 6:00 and I immediately go to the rooms that we will be occupying the most that day. I turn the thermostats up to 70-something and turn on the ceiling fans in two of the three rooms we are trying to heat. Two of the rooms heat up to the desired temps in 30 - 45 minutes. The longer time is for the larger room with more glass. The third room, which is the same size as the second one, and with the same window but without any ceiling fan, takes double the time to reach the same heat. From those simple observations it is clear to me that the ceiling fans do help harvest heat from the hot ceilings, especially in the larger room with 12' cathederal ceilings. During the day, especially on a really cold day, we will turn off the fan when we enter the room, but will turn it back on again when leaving if we plan on not returning for a sufficient length of time. This has worked very well for us in helping to keep the house comfortable and make the most of the available utilities.

  11. GBA Editor
    Martin Holladay | | #11

    It's your house, and you can operate your ceiling fans any way you want, but I agree with Dana Dorsett: Operating ceiling fans during the winter usually raises your energy bills. Here are links to two articles on this issue:

    Using Ceiling Fans To Keep Cool Without AC

    Cold Floors and Warm Ceilings

    Dana is also right about the insulation temperature. It is absolutely normal for measured temperatures in a thick layer of attic insulation to range from the indoor temperature (close to the ceiling drywall) to the attic temperature (at the top of the insulation thickness). That just means that the insulation is doing its job. The insulation is slowing down the flow of heat from the warm side of the insulation to the cold side of the insulation.

  12. Hotsauce1183 | | #12

    Hey Guys - can we back up a few yards here. We seem to have lost sight of the original point of the conversation which was that in a house where electric radiant heat ceilings were the only source of heat using a ceiling fan was an effective way to hervest more heat quicker from the warm ceiling and get it down into the room. In this house rooms without ceiling fans take more than twice as long to warm up as rooms that do. Plus, this conversation seems to have turned into an unintended debate for which I apologize. No disrespect was intended.

  13. charlie_sullivan | | #13

    The way you describe using the ceiling fans to help the radiant ceilings warm up a room faster while the room is unoccupied makes sense to me. By avoiding having the fan on while it is occupied, you avoid the primary disadvantage of it.

  14. Jon_R | | #14

    With electric heat and an unoccupied, highly stratified room, I'm sure that destratification increases efficiency. With occupants, I expect that the air velocity at the occupants needs to be less than .1 m/sec (not noticeable, no negative effect on thermal comfort).

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |