Radiant floor sensor?

  • Active since 1995, Hearth.com is THE place on the internet for free information and advice about wood stoves, pellet stoves and other energy saving equipment.

    We strive to provide opinions, articles, discussions and history related to Hearth Products and in a more general sense, energy issues.

    We promote the EFFICIENT, RESPONSIBLE, CLEAN and SAFE use of all fuels, whether renewable or fossil.
Status
Not open for further replies.

Sawyer

Minister of Fire
May 17, 2008
608
Northern WI
Jebatty mentioned a floor sensor set at 61F. I have a wall thermostat at present and am not happy with the swings in temperature. We had to pour a slab over the existing slab so my pipes are lower than they should be but I have to live with it.

Do you think if I go to a floor sensor I would “soften†the swings I presently experience? Presently my pump will run about ten hours to satisfy the thermostat. When it is -20 I would prefer that this long drain did not happen in the middle of the night and pull down my storage temperature.
 
The sensor would have to be embedded in the over pour at a sufficient depth without nicking the tubing. You also will need a thermostat to enable floor sensor. I have an Uponor that I got from pex supply.
You might have to boost the water temp due to additional slab thickness. What temp are you running now?

Will
 
There is a lot in your post, but I have experience with one part of your question. Regarding temperature swings. My circulator runs all of the time. Room temperature is maintained by a three-way valve positioned by a differential controller. The controller compares floor water supply temperature with exterior temperature. After the controller is adjusted for the building response, my interior is 70 °F 24-7. In other words, outside temperature changes will result in a change inside in a few hours. So the floor supply water temperature is adjusted a few hours before the change is needed. The number of hours depends on the flywheel effect of my house.

Now this part is speculation. I don't know about the response of your system, but my guess is it takes more than 12 hours to cool or heat in any significant way. So timing the heat input and also keeping comfortable daytime temperatures may be difficult. Slow response is one down side to in-floor radiant.

My control system is 20 years old. Better systems may be out there these days.
 
I assume the existing slab has the radiant and you poured over the top of that. Any insulation under the existing slab? around the perimeter concrete edge? vertical into the ground 2' and more to prevent frost from penetrating under the slab? Any and all will help keep the heat in the slab and reduce temperature swings.

If my assumption is right, you could safely bore a hole for the sensor into the top slab right to the top of the bottom slab without danger to the pex in the bottom slab. A Ranco or Johnson electronic control would work well, just extend the sensor wires.

From Jan 24 logged data: During the log period outside temp HI was 26, inside temp HI was 59; and outside LOW was 12, inside LOW was 55. Floor sensor set at constant 61 with 1F differential, floor ON at 61, OFF at 62. Floor sensor ranged between LOW 59 and HI 63.

You can see a little undershoot and overswing on the floor sensor, as it takes time for heat/cold to travel through the concrete. My sensor is located midway between two runs of pex, so about 6" from a pex run, and about 4' interior from the perimeter of the slab.
 
Willman said:
The sensor would have to be embedded in the over pour at a sufficient depth without nicking the tubing. You also will need a thermostat to enable floor sensor. I have an Uponor that I got from pex supply.
You might have to boost the water temp due to additional slab thickness. What temp are you running now?

Will

Will, I am running 130F water to the floor. Due to the added thickness drilling a hole for a well will not be a problem.
 
I use 100F supply to the floor. It seems to me that boosting the supply temp also will raise the slab temp, resulting in greater over-swings in air temp. Wouldn't you want to keep the floor temp at a temperature that results in the slab surface temp which maintains your desired air temperature? For this first heating season with my slab, I intentionally kept the slab temperature sensor at a constant to see how interior air temperature varied with outside temperature. As Performance shows, and this being a shop, except for very cold winter periods the 61F floor setting worked very well. Keep in mind that radiant "feels" warmer than the air temperature, and that radiant, at least in my case, virtually eliminates any drafts. I would not want the air temperature in the shop to be any higher than shown while doing work in the shop, as I usually just work a long sleeve flannel shirt, and even then on occasion it got a little warmer than desired.

Next winter I may consider bumping up the floor temp a little when the forecast is for outside temperature of below -10F, or get an outdoor reset that accomplishes the same. The wrinkle is that it takes a long time for the floor to respond to changes, 12 hours is a pretty good estimate, so a person has to act in advance of any desired change.
 
jebatty said:
I assume the existing slab has the radiant and you poured over the top of that. Any insulation under the existing slab? around the perimeter concrete edge? vertical into the ground 2' and more to prevent frost from penetrating under the slab? Any and all will help keep the heat in the slab and reduce temperature swings.

If my assumption is right, you could safely bore a hole for the sensor into the top slab right to the top of the bottom slab without danger to the pex in the bottom slab. A Ranco or Johnson electronic control would work well, just extend the sensor wires.

From Jan 24 logged data: During the log period outside temp HI was 26, inside temp HI was 59; and outside LOW was 12, inside LOW was 55. Floor sensor set at constant 61 with 1F differential, floor ON at 61, OFF at 62. Floor sensor ranged between LOW 59 and HI 63.

You can see a little undershoot and overswing on the floor sensor, as it takes time for heat/cold to travel through the concrete. My sensor is located midway between two runs of pex, so about 6" from a pex run, and about 4' interior from the perimeter of the slab.

Jim, I have 2†DOW under and around the perimeter. Due to shortsightedness the apron side of the shop does not have a perimeter barrier.

I do know the depth of the over pour so drilling to the lower slab is not a problem. What is the advantage of having the sensor near the tubing rather than near the surface which is radiating the heat? I am assuming the surface temp swing will be less but you know what they say about assuming.
 
jebatty said:
I use 100F supply to the floor. It seems to me that boosting the supply temp also will raise the slab temp, resulting in greater over-swings in air temp. Wouldn't you want to keep the floor temp at a temperature that results in the slab surface temp which maintains your desired air temperature? For this first heating season with my slab, I intentionally kept the slab temperature sensor at a constant to see how interior air temperature varied with outside temperature. As Performance shows, and this being a shop, except for very cold winter periods the 61F floor setting worked very well. Keep in mind that radiant "feels" warmer than the air temperature, and that radiant, at least in my case, virtually eliminates any drafts. I would not want the air temperature in the shop to be any higher than shown while doing work in the shop, as I usually just work a long sleeve flannel shirt, and even then on occasion it got a little warmer than desired.

Next winter I may consider bumping up the floor temp a little when the forecast is for outside temperature of below -10F, or get an outdoor reset that accomplishes the same. The wrinkle is that it takes a long time for the floor to respond to changes, 12 hours is a pretty good estimate, so a person has to act in advance of any desired change.

Jim, your 100F supply temp makes sense to me; I have noticed floor temperatures at 86F when the pump shuts off. DeltaT’s of 45 degrees both on the supply lines to the HX and also on the lines to the in floor tubing.
 
It would seem that the closer the sensor is to the pex, the quicker the sensor will respond, and therefore the more uniform the floor temp will be, and vice versa. But if the sensor is further away from the pex, a lower setting on the sensor might achieve the same result by limiting the over- and under-shoot of the change in floor temperature on the air temperature. I'm quite new at this, so others may have better advice. I have been satisfied with my first stab at setting the temperature for the floor sensor to achieve a desirable air temperature and likely will not change it.
 
I wonder why the temp swings so much? I went over budget on my system and didn't want to buy a thermostate at the time so I used a timer to controll heat in my shop floor. Usually I ran it once or twice a day for about 45 min at a time and the temp never varried more than 5 deg. I also installed my mixing valve wrong so I costantly had different temp water going threw the floor. I plan on putting in a thermostat with slab sensor this summer
to automate things more. Sometimes I even ran 185 F water threw the floor with no big swings in temp.
 
JT, my problem is not with the temperature swings of the workshop. it is well insulated and seems to hold the heat well as the thermostat calls for heat and the pumps runs for 8-16 hours heating the slab. I would like to have the pump cycle more often and maintain a more constant slab temperature.

It sounds like Jim is on the right track with a floor sensor and i will drill deep enough to get the sensor near the tubing. Perhaps a timer could be configured to do the job also.
 
I see. By the time the thermostat is satisfied the slab is to warm and it over shoots the desired temp. Mine worked pretty good with a timer ,but still want to get a in floor sensor thermostate.
 
After leveling the floor, after it settled on a "compacted site", I had to pour 8" of concrete on the back to get 2" in the front (the minimum that was recommended for an acceptable bond and strength) which takes a long time for heat to reach the surface in the rear 1/2.

I would have mud-jacked the floor but they could not raise it, the concrete would blow (volcano) before it would lift. :shut:
 
I have plans to build a workshop with radiant heat in the slab. I have a few questions:

1) What are people's thoughts on using a Taco iSeries three-way outdoor reset valve as the mixing valve for the workshop radiant/PEX zone? http://www.taco-hvac.com/uploads/FileLibrary/100-19.pdf
2) What do people think of the Uponor Setpoint Controller with Floor Sensor instead of a basic thermostat? http://www.pexsupply.com/Wirsbo-Upo...tPoint-Controller-with-floor-sensor-2068000-p
3) Regardless of which floor sensor I use, what's the ideal placement of the sensor and would it be better to embed a piece of copper in the slab, drill a hole for the sensor, or is there another option?

I know that neither the iSeries valves nor the Uponor Setpoint Controllers are cheap, but I plan on spending lots of time in the shop and would like it to be comfortable. Are there other options that are cheaper that would work as well if not better?
 
In preparing for the pour, I ran a 1/2" length of pex from the perimeter and in about 4 feet, and then ran the pex up the wall a ways to give me some working room. Then a plug on the end of the pex to be embedded in the concrete. After the pour and when I was ready to do the wiring, I shoved the sensor with cable into the pex all the way to the end. In fact I ran in two sensors, one for the control and the other a DS18B20 for data logging the floor temperature. The control is a Johnson A419, although I think I might have said above that it was a Ranco. Both controls are excellent.
 
Thanks Jim, I have the Johnson A419 control up on a search page and will get one on my next hydronic order along with some RIBs.
 
What do people think of the Uponor Setpoint Controller with Floor Sensor instead of a basic thermostat?

I use these to control my infloor in my bathrooms. Seems as if they rate fairly well. They require a 24v input. I used mine on my Taco controller. I used a piece of PAP from my thermostat location down into my screed. I pinched off the end. The sensor will slide down no problem, I used a stiff wire to push in after pre measuring to reach the correct depth. Same rules apply for sensor placement as thermostat. A good average of floor area. Not near a door or to close to outside wall.

Ideal slab/water temp for your application will have to be learned. All I can add is make sure your underslab insulation is the best and most you can do as there is no do overs.Also moisture management is critical. Can't comment on your other points.

Will
 
Pic showing floor sensor tube.

Will
 

Attachments

  • HPIM0194.jpg
    HPIM0194.jpg
    25.1 KB · Views: 378
Good pic that also shows about what I did for the pex tube into the concrete to house the sensor, between two runs of pex, 4' in from a wall, and not near an outside corner.
 
Status
Not open for further replies.