Maximum temp for suspended hydronic PEX floor heat

[StephK]

Hi trusted members of this forum. Your help over the years has always been much appreciated! Got a floor heat question for you now.

We have installed PEX radiant heat suspended in the joists under the floor of our 1910 house. I did a ton of research before doing so and was told not to worry, the heat would eventually rise to the room ..... but .... after all this work, it's not quite working as planned. The furnace runs continuously to keep the room warm. I turned the mixing valve so the temp was high, and the floors/room eventually did heat up, it took 2 days, but to the tune of 170 water temp. I thought once it heated up, I could turn the water temp down to the recommended max of 120. But, when it comes back on, it runs continuously again. I'm worried there's just too much floor for the heat to get thru!


Here's the details:
Bathroom - subfloor + 5/8 plywood + durock + tile.
Hallway,closet,bedroom - subfloor + hardwood floors.

Used metal shield plates underneath
R19 insulation under that

My question is, how hot can I safely run the water to make this thing work? My boiler is kicking it out at 180 and I have a mixing valve that seem to be able to output at 170. I was worried about the hardwood floors and also the thinset and tile. But, by the time the heat gets to these layers, it's NOT 170.

Thoughts??

 

[JohnCT]

We have installed pex radiant heat suspended in the joists under the floor

I have zero knowledge of your problem, so take this for what it's worth..

Seems to me that Pex is far less able to transfer heat than copper. Whereas copper conducts heat, PEX would seem more like an insulator. In fact, I've been told that PEX repiped houses experience hotter domestic water because less heat is lost to the air.

But anyway, I found this:

Underfloor Heating Systems
When radiant heat is desired in a typical residential home with wooden frame construction, PEX is most commonly installed underneath the plywood subfloor, attached directly to it. The PEX lines form a single loop inside the joist bay, then go to another bay and so on. Heat is absorbed by the flooring above and is further transferred to the space above. These systems require quality insulation to maintain efficiency and are often installed in combination with aluminum heat transfer plates.

That would suggest two things. The first is that the PEX is directly attached to the sub floor, not suspended and the other is that they are used in combination with "aluminum heat transfer plates". I would guess the plates are securely mounted to the subfloor and the PEX attached to the aluminum plate to facilitate heat transfer.

Good luck.

John

 

[Sponsor]

 

[StephK]

Oh yes, I just used the word 'suspended' b/c that's usually how it's referred to, as opposed to embedded in concrete on top. The pipes are attached directly to the subfloor, and the aluminum shields are attached to the floor over the pipes. The shields keep the pipes snugly on the floor and also radiate the heat.

 

[Plumber69]

How long are your runs? If you just have one long run it won't work properly.

 

[Plumber69]

180 is way too high.

 

[StephK]

it's 3 separate runs. Each ones is 250-ish feet. Laid out 6 inches apart.

 

[StephK]

180 is way to high

When you say that's too hot ... for what? The Pex? The water in the pex starts out at 160-170. But, by the time the heat gets thru the subfloors, durock, etc, it is not that hot. A thermometer on top the floor tells me the floor is 70. It's just frustrating, b/c I spent so much time researching, and designing the system with the company that sold it to me. They understood exactly how much floor we were dealing with.

The water comes from our furnace, which is being heated to 180 for the radiators in the rest of the house. I want to understand if 170 is too hot, is it too hot for the pex, crimp rings, etc. I know that would be way to hot for wood floors, thinset, etc. But, like I said, it's not that hot on those levels, or I'd be baking in the room!

 

[StephK]

PS - Not sure why my photo is a chicken ... but she's kinda cute, so I guess I'll leave her be.

 

[Dana]

A peak water temp of 180F is fine, but if it's running 24/7 at that temp it will dry out the subfloor quite a bit. Whether that causes any issues with rest of the stackup remains to be seen, but I doubt it's a real problem.

All PEX is good for 180F, some is good for 200F.

A floor temp of 70F means the thermal bond isn't as good as you think it is. Put some insulation over the top of the thermometer and give it 5 minutes to know if that's really the temperature. Usually 180F water in a plated subfloor radiant would deliver uncomfortably hot surface temperatures on the floor. Are you absolutely positive of the stackup? I have plated-up under the subfloor radiant zone with two layers of 3/4" subfloor + 3/4" hardwood on top and get ~73-75F surface temps in a 70F room with only 125F water. My stackup should only putting out about 2/3 the heat as yours since it's higher-R. Something other than water temperature seems to be the issue.

It sounds like you may have way too little actively heated radiant floor for the heat load of the bathroom (which is a problem for many smallish bathrooms. It's fine to run radiant under the bathtub/shower to get more heat into the room, but not under the toilet. If it's that warm under the toilet it can melt out the wax seal on the closet flange.

With any under the subfloor radiant you can't be using overnight setback strategies- set the temperature and forget. Letting it drop to 60F overnight, turning it on at 6AM and expecting to get the room up to 70F by noon isn't always realistic.

"Suspended tube" radiant means just that- the tubing is hanging a few inches below the subfloor.

"Staple up" radiant puts the tubing in contact with the subfloor, but doesn't have any metal heat spreaders.

Plated subfloor radiant can have either sheet metal plates clamped over the tubing, and screwed to the floor, or stiff solid extruded plates with a channel designed as a tight friction fit for the PEX. Any plated type should be able to deliver much higher surface temps at the tile that you are reporting.

 

[Plumber69]

Take a pic of your setup. What kinda sensor is telling boiler to come on. Try running one zone at a time and see if it runs properly. Maybe your issue is just with one zone

 

[StephK]

Thanks for all the comments.

Dana - you are right, it sounds like your flooring is thicker than mine, with better results. Ours is the staple up method with the extruded thin metal (alumumin?) sheets with a channel for the tubing. These plates are screwed to the subfloor. The entire area is covered, except directly below the toilet (although we did use the new plastic ring for fear of meltage). There are 2 runs in each joist, so 7ish inches apart. We hung R19 installation, but hung it even with the joists, we didn’t jam it up to the top. Could that be a problem, too big a gap?

I will do some testing. You’re right, I didn’t cover the thermometer when I took a reading. I’ll do that. I’ll also turn off zones and see what that does to the floor temp. Does pressure have anything to do with it?

This setup is so simple and so mechanical, I don’t get how yours performs so much better.

I will add that the rooms below this are not heated yet ... still a construction zone. They are probably 45-50 degrees. But I can’t imagine that would affect the floor temp that much ...

 

[Plumber69]

I think you want the insulation all the way up

 

[Jadnashua]

Can you measure the return water temperature? I've been told loops tend to work best if they are 200' or less...this, along with the right circulator, means it isn't too cool at the end so you don't end up with one end way warmer than the other. It also tends to work best to start the run near the exterior and work your way in, this puts the warmest water at the (likely) coolest area.

Your critical safety issue is the surface temperature of the floor. Could you have kinked the tubing somewhere, slowing the flow?

Depending on the thermal mass, radiant systems tend to work best when run constantly without a setback, or if there is one, not a huge one. On a high mass system, it can take a couple of days to stabilize.

If there is air infiltration, it can prevent the system from working well. Is your rim joist well sealed? Fiberglass insulation isn't the greatest at blocking air infiltration.

It can be tough sometimes to purge the loops of air so the flow works well.

 

[Dana]

I think you want the insulation all the way up

That's right- the insulation needs to be snugged up to the sheet metal plates & subfloor. The air space forms a thermal bypass, allowing air to pass freely around the insulation, undercutting it's performance.

But even that doesn't quite explain why the tile temps would only be 70F with 160F+ water.

StephK: What is the design heat load of the bathroom, and how many square feet of active radiant floor?

 

[Plumber69]

Whats the temperature outside? And is everything thing ran in a cold crawl space

 

[Plumber69]

I remember I put one in. The return water would mix with the send thru a box. I think this tricked the boiler into thinking it was up to temp. Although it would initially take longer to get up to temperature in the rooms.

 

[StephK]

OK, I have accumulated some details. Sorry if this is TMI.

Boiler water temp - 180 (coming from high-efficiently Munchkin boiler)
Return water temp - 145 (It's hard to read the tape strip thermometer, but each of the 3 zones are very hot to the touch. Can't hold them for too long)
Mixing Value produced mix for pipes - currently set to 170

I have not messed with the thermostat. Set and Forget at 63. Boiler runs continuously b/c the floor heat circulator calls for it (This is the part I'm unsure of ... Should the boiler really run continuously, just b/c the circulator is calling for it? Shouldn't the furnace water stay at temp for bit, then maybe come on again?)

Floor temp with blanket covering thermometer - 80 Wood floors feel barely warm to the toes, tile never really feels warm.

Staple up method, pipes covering the entire floor area every 6ish inches. Sheet metal with channel cut for pex, screwed to the subfloor everywhere except the turn loops.

Bathroom - 7/8" subfloor + 5/8" plywood + durock + tile.
Hallway, closet, bedroom - 7/8" subfloor + hardwood floors.

Mistake made - R19 fiberglass bats, installed level with the bottom of the ceiling ... not squished against the metal/pipes. joists are 2 x 12s. I'd stay there is a 6" gap. Many sheets of drywall hung so no fixing that.

Room underneath is unheated as of now, but it's the kitchen and will eventually be heated. Temp is probably 50ish in there now.

Outside temps 35-40.

I've attached two files. One is the loop map with lengths for each zone. Plus an annotated photo of our plumbing.

Thanks again for all the insights!!

 

[Dana]

You won't get high efficiency out of the Munchkin running it at 180F with a 145F return. The entering water temp at the boiler needs to be ~125F or less to get condensing efficiency. The 35F delta seems a bit high, but isn't a disaster- a 20F-25F delta-T is more typical at 180F out in a well design system, but 35F is fine. If the 145F is just the return from the mixed down 170F it's close to ideal. At lower temps the delta will be smaller. Most boilers can tolerate up to a 50F delta, in-to-out.

Any time the zone controller is sensing a call for heat the circulator will keep running, but that doesn't necessarily mean the boiler is firing. If the heat load is greater than the minimum modulated output of the boiler and the water temp is lower than the boiler's set point the burner will keep firing too. It's more likely that the water temp is going above the setpoint and it stops firing until the temp drops below the setpoint, cycling on/off on some duty cycle as the circulator keeps running continuously. If the burn cycles are short (<<3 minutes) and frequent (>>5 burns per hour) it's robbing some efficiency and putting unnecessary wear & tear on the boiler. Measure them as a sanity check.

Which model Munchkin? Is it under any sort of "outdoor reset" control, or are you just running it fixed-temp @ 180F?

It's possible to blow the joist cavities full of cellulose or fiberglass over the batts without replacing the drywall. That may be doing (for several reasons), but isn't likely to be the cause of the problem unless you have large leaks in the band joist in the bays under the bathroom floor.

To get a handle on the heat load of the bathroom...

What is the framing and insulation in the bathroom walls?

Are the windows 1950s vintage aluminum sash single panes, or something more efficient than that? (Low-E double pane? Clear glass wood sash single pane plus clear glass storm windows?) Describe them.

How many square feet of window, and how many square feet of exterior wall?

Are other rooms able to maintain a reasonable indoor temperature?

 

[Plumber69]

Are you feeding 3 zones with 1/2 inch PEX?

 

[Dana]

are you feeding 3 zones with 1/2 inch pex

He doesn't have three zones, but rather three circuits of half-inch PEX, the shortest being 261', the longest 276', as spelled out the sidebar of the system plot: