Underperforming radiant floor heating, how can I improve it?

pipecurious

Hi Everyone,

I've been doing a lot of reading on here and was looking for some advice.

We bought a house (built 2005) a couple years ago that has all radiant floor heating (in unfinished basement, first floor, and upstairs bedrooms). It's a two-story, ~3000 sqft (not counting basement), living room has cathedral ceilings. We live in a cold climate (less than an hour from Montreal) and we've noticed that on some of the cold and windy days (especially single digit Fahrenheit temperatures and 20+ mph winds) we get in the winter, the house doesn't stay warm. We keep the thermostats all set at 73 degrees F, but on these cold and windy days our master bath was only able to get to 66 degrees F and the living room was only able to get to the upper 60's as well. One of the bedrooms upstairs is our nursery and that couldn't get above 67 degrees F. Overall the house felt noticeably colder than normal.

Overall the house is well air-sealed and we don't have any major air leaks. The house is well insulated, considering the year it was built, we have 2x6 walls with interior fiberglass batt insulation and exterior foam insulation (1 inch). Rim joists are sealed and insulated with fiberglass batts. Overall the house has a lot of glass (many large triple pane casement windows). Upstairs rooms have fairly large double pane, double hung windows, and on the very windy days I did notice some air leakage around the double hung windows, so I sealed those with some flexible rope caulk, but that only helped to increase the room temps there by a degree).

For heat source we have a 154 MBH oil boiler, but we seldom this (it's just backup) as we have an outdoor wood boiler that is tied to the oil boiler via a heat exchanger. The plumbing is primary secondary, and each room is on its own zone with a circulator pump and mixing valve (6 zones total). This whole setup is original to the house.

I noticed the radiant floor temperature gauges located right after the circulator pumps for the first floor and upstairs bedrooms were reading 150 degrees F). So after doing some reading I wanted to see what radiant floor setup we had. I checked the first floor joists from the basement and can see that there's radiant barrier bubble insulation stapled up, a few inches below the radiant lines, and the radiant floor lines behind that are fastened to the subfloor with PEX clips, so they hang below the subfloor by about an inch. This seems like some version of either plateless staple up or suspended radiant floor from what I've been reading.

My question is, what can I do to improve the setup so that we don't see temperature drops in the rooms? I'm assuming if heat loss calculations were done when the system was designed, the design temp is surely lower than single digits, as historically we've had much colder weather in the winter (the last few years have been mild). Might I see an improvement if I put insulation in the joist bays (e.g. rockwool)? What about adding plates? How about both? At some point we'd like to replace the oil boiler for a heat pump and I figure plates would be necessary to get the flow temp lower to allow for that. Is there anything I'm missing that would help?

Any suggestions are very much appreciated. I'd like to make changes sometime in the spring/summer once the heating season is over so that next winter the house will stay warmer.

Jamie Hall

You're not missing much there. Better insulation below the piping (but don't cover it) will help — the bubble wrap isn't doing much for you. Plates would help — but only if you could move all the piping so everything was in contact with the floor rather than spaced.

That said, can you measure the actual floor temperature when the system is running full out? Not the air temperature just above the floor — but the actual material temperature? That will be the best guide to whether the system can actually provide the heat you need, as there is a definite upper limit on how warm the floors can be! 80 F is about as warm as is nice to live with…

bjohnhy

Post some photos of you boiler, circulators, mixing valves, manifold, etc.

What are the return temps from floor loop?

DCContrarian

When it's running full out, how are the floor temperatures? Are they comfortable to walk on? Because the only way to get more out of the floors is to make them hotter, we can talk about ways to do that, but if they're already at the edge of comfort that may not be the way to go.

There's basically two ways to approach this. One is to get more heat out of the floors, making the surface hotter. If you're already seeing 150F for the water temperature the easiest step of raising the water temperature has already been taken, so it's going to involve reconfiguring the piping below the floor. The other approach is to leave the floors as they are and add more heat emitters. Those could be radiators, fan units, towel warmers, all kinds of things. If you're already running pretty hot water it would be pretty straightforward to add them.

If you have a way of measuring the existing floor temperatures we can estimate the existing heat output, which will gives an idea of how much capacity needs to be added. We can also estimate from fuel usage, and from return water temperature and flow.

hot_rod

You have a joist bay installation method of tube installation. The tube heats the air in the joist bay, which convects to the floor. Those generally need to run pretty hot 150°F plus.

An outdoor reset control is helpful, it would ramp that 150 up when temperature drops outside.

An infrared camera is a good way to assure it was tubed properly. Use it for looking at the structure for leaks also. The phone attachment IR cameras actually work well enough for this. This is a common tool on hydronic service trucks around here.

You'll find many other things to look at with an IR camera😎

Screenshot 2025-01-10 at 3.55.21 PM.png

pipecurious

Thank you, everyone, for the replies. I will take some photos when I get a chance later today and post those. I don't have a thermometer to measure floor temperatures but I am going to get one this week and I'll take some measurements and report back.

We don't find the floors ever to be uncomfortably warm, but the warmth is noticeable at times. So from what I understand, there may be room for increasing the flow temperature, as long as the floor temperature isn't too high?

Is there an ideal timeframe for measuring the floor temperature? That is, should I turn up the thermostat for a zone and check it at certain time intervals to find the maximum temperature?

hot_rod

On the coldest day of the year a system could be expected to run non stop if it were sized correctly. So a floor temperature on that day would be realistic.

Thermal equilibrium is reached in all hydronic systems when neither the supply water temperature is rising or the return. So all the heat being delivered is going into the space. At that point you could measure floor temperature, and the delta T temperature difference between supply and return.

You have what is known as a low mass system, it should warm the floors relatively quickly.

On a mild or low load day, the thermostat may turn off before you get accurate surface temperature readings. So the best results will be a cold or design day where the system can run long enough and stabilize.

pipecurious

OK, thank you, that is helpful information. I got an infrared thermometer and did some measurements. Based on what you wrote, Bob, I'm not sure how accurate the temperatures are that I recorded, in terms of what could be expected on a design day, but we had some outside temperatures in the low teens (degrees F) and I recorded the supply temperature to the radiant floor zone in the living room to be 150 degrees F and the return to be around 110 degrees F (I used an infrared thermometer, and adjusted the emissivity until the reading I got was the same as the temperature gauge on the supply for that zone). I checked the temperatures of the floors in the living room, right after the zone finished calling for heat and the room reached the temperature set on the thermostat, and the highest I got was around 80 or 81 F depending on the spot I checked (some areas were as low as 77 F). In the kitchen the floor temp was between 75 and 80 F, depending on the location checked, and upstairs I got around 80 to 82 F. Do these temperatures allow for some flexibility in making adjustments and/or modifications to the current system?

I've attached a photo of the manifold with the zone piping, in case that is helpful too (sorry for the angle, it's too tight a space to take a photo straight on). The first circulator on the zone is isolated and disconnected because we replaced the old indirect water heater with an electric heat pump water heater. The circulator after that is for baseboards in a finished space above our garage. All the rest are for radiant floor zones for all the main rooms in the house.

I'd like to have the option in the future at some point to run a lower flow temperature so that we could switch to an air-to-water heat pump, but I'm not sure if that's feasible given our current set up.

bjohnhy

Is your thermostat set up for a setback? turn down a few degrees at night?

Can you point out the circulator for the zone that is functioning the worst?

Those mixing valves are mixing the temp down to 150F? You said you were measuring temp just above circulator.

And the return from said zone is 110F?

To have a 40 degree difference of supply and return for a floor loop usually means the flow is sub-optimal. It looks like the Grundfos circulators are all set on Hi setting. What model are they? What model are those mixing valves? Some of them can be installed with a mesh filter that can get clogged and reduce flow.

Can you post photo of the air vent, expansion tank, main system circulator?

pipecurious

No, the thermostats stay at a constant temperature. Most of the rooms we have set at 73F.

When we've had colder weather (single digit F with strong winds) all the zones seem to perform poor in the sense that they can't get the rooms up to the set temperature. The living room zone has a lot of large windows though, so I'm assuming that that zone has the most heat loss (but I may be incorrect in that assumption). That zone is the 5th circulator from the left in the photo I posted before (there's a bit of surface corrosion on the copper piping you can see).

Yes, for the living room zone the mixed temp was 150F (there are temperature just above the circulators on all the radiant zones and for that one it read 150F). I checked with an infrared thermometer (Fluke 62 Max) and it read right below 150F, and when I used that to measure the adjacent return pipe I was getting 110F.

Yes, it looks like all the circulators are set on the highest speed. All of the circulators, including the main circulator, are Grundfos UPS 15-58 FC. We've lived in the house for about 2 years, and I haven't touched those, so not sure if that's what they were at when the system was put in or if the previous owners increased the speed over time. One thing I will note which I realize I probably should have mentioned before is that a few of the circulators make noise (kind of like marbles rattling around in a can). My understanding is that the circulators should be fairly quiet, so I'm not sure what that noise is. I tried purging air from the system a few times, one zone at a time, using the isolation valves, and it doesn't seem to have improved or worsened the noise at all. Could the circulator pumps need replacing?

I've included some photos of the main circulator, the air vent, and the expansion take below. There is another air vent lower on the boiler too. Sorry for all the photos but with how everything is positioned there was no way to show everything clearly in one photo.

Main circulator (photos from two different angles):

expansion tank and air vents:

top air vent:

boiler:

living room zone circulator pump:

expansion tank:

pipecurious

Update: I thought the previous temperatures I measured might've been inaccurate because of the emissivity of the copper pipes. I wrapped some paper underlayment around the supply and return of the living room zone and I measured 139F supply and 134F return.

bjohnhy

Those supply/return temps are consistent with good flow (maybe too good). You have near exact problem as user Coldhouse here

https://forum.heatinghelp.com/discussion/199925/many-years-of-issues-with-radiant-floor-heat

if you haven't read through that one, most of it applies to you as well.

I cannot emphasize enough on air sealing the house. Especially those ceiling penetrations to the attic. I moved into a 2005 "tight" house and on the 25F days it was using about 40K btu/hr. After some moderately aggressive air sealing that number is now 25k btu/hr on those days.

pipecurious

Thanks, I am planning on getting a thermal camera to see if there are any improvements I can make in air sealing.

Does the small difference between supply and return suggest I should lower the circulator speed?

Teemok

Lowering Circ. speed will not improve floor output. Very accurate supply and return temperatures (type k senor taped and insulated) and a few thermal camera shots of the floor when you have had the system calling for heat but are unable to meet set-point would tell a lot. Photos of the tubing and insulation in a joist bay? There are often more than just one problem. Stands to reason, if one short cut or misunderstanding was deemed acceptable there likely others.