Temperature/humidity sensors for concrete slab for when ASHP run in cooling mode - what do I need?

[markharro]

Our installer says that there is no option to wire up sensors to either the Vaillant ASHP controls or UFH etc so I am wondering if there is some sort of Raspberry Pi solution to give me a basic monitoring of slab temps/humidity so that I can decide if it is safe to run the ASHP on cooling mode? Our slab gets poured in about 10 days so I need to know what sort of sensors and how many of them to buy with a view (I am assuming) to having them embedded in the slab during the concrete pour? thanks

[Dan F]

The Vaillant controller has a humidity sensor and can be configured to limit flow temperature to avoid condensation based on current relative humidity.

 

Vaillant controls aside, it's still a good idea to put some temperature sensors in the slab (inside some UFH pipe off-cuts) in different locations.  Do you plan any type of home automation system that you could hook up these sensors to?

 

Assuming you can use 1-wire sensors, get a pack of DS18B20 sensors.   Buy ones with long leads so you can place the sensors where required and then connect them up within an internal stud wall. 

Edited March 17, 2023 by Dan F

[joth]

Yup condensation is a function of air humidity and flow temperature (assuming that it's impossible for the slab to be colder than the cold water entering it) so no in-slab sensor needed for the use case in mind 

 

That aside, +1 to DanF get some DS18B20 and drop in various locations. If you're at all handy with a soldering iron EspHome is easy to get running as a simple bridge to view the data in Home Assistant. 

I still have a few sensors burried just in case that I've not bothered to connect yet. But cost pennies to put them in. 

 

Note there's a lot of counterfeit sensors out there so if encasing in concrete may as well try and get the legit article

https://github.com/cpetrich/counterfeit_DS18B20

Edited March 17, 2023 by joth

[MikeSharp01]

We buried some spare UFH pipes, suitably blanked in straight lines from the manifold position to places in the slab I wanted to measure and two straight down into the sub soil under the slab, suitably sealed to the dpm. I can then push the sensors down the pipes with a wiremans tape and hey presto I have temp from a over the place. The DS28B20s are great and cheap but need calibrating a DHT22 has humidity also and I have some of them in the middle of the wall insulation. 

[Radian]

8 minutes ago, MikeSharp01 said:

DHT22 has humidity also and I have some of them in the middle of the wall insulation

And do you tremble when the conditions get close to the dewpoint? 🤣 I put one in the void between the joists supporting a warm, flat, roof. Quite fascinating.

[DanDee]

10 minutes ago, MikeSharp01 said:

We buried some spare UFH pipes, suitably blanked in straight lines from the manifold position to places in the slab I wanted to measure and two straight down into the sub soil under the slab, suitably sealed to the dpm. I can then push the sensors down the pipes with a wiremans tape and hey presto I have temp from a over the place. The DS28B20s are great and cheap but need calibrating a DHT22 has humidity also and I have some of them in the middle of the wall insulation. 

@markharro Put multiple tubes of 1cm interior diameter with the entrance at the base of the walls under the skirting, and later you will put any temp probes you find suitable for your system.

I don't think slab humidity sensor is something that you need to worry(I've just changed my mind)

For kitchen/bath I would put humidity probes in the grout joint for leak detection, and now for condensate from cooling, but really you need a way to manage the temp to avoid condensation rather than sensing when it happens

[Bramco]

24 minutes ago, MikeSharp01 said:

DS28B20s

18 B20s?  if so they don't need calibration but you do need a microcontroller based system to read the them.

 

I was planning to put in some tubes but didn't get round to it - so we'll see if actually need to run the UFH in cooling mode - although I can't see how you can do this with standard thermostats - they want to heat, not cool - so you'd have to have a separate system that could switch off the thermostat control and introduce a cooling control - you won't get your standard ASHP engineer or plumber specc'ing that for you.

 

Simon

[jack]

1 hour ago, Bramco said:

18 B20s?  if so they don't need calibration but you do need a microcontroller based system to read the them.

 

I was planning to put in some tubes but didn't get round to it - so we'll see if actually need to run the UFH in cooling mode - although I can't see how you can do this with standard thermostats - they want to heat, not cool - so you'd have to have a separate system that could switch off the thermostat control and introduce a cooling control - you won't get your standard ASHP engineer or plumber specc'ing that for you.

 

Simon

 

I'm not sure whether it's a common feature of thermostats, but if you can get one with normally open and normally closed outputs, you could easily wire a summer/winter switch so that the right output is controlling cooling/heating.

[MikeSharp01]

1 hour ago, Bramco said:

if so they don't need calibration

I find they do need calibrating so I use a 5w Calorimeter and a traceable thermometer, put a few of them in at low temp 0 or thereabouts, and then let it track up, they can be several 10ths out so I mark the units and programme the curve fit, usually close to straight - y=mx+c, into the microcontroller - works a treat.   

[MikeSharp01]

1 hour ago, Radian said:

And do you tremble when the conditions get close to the dewpoint?

Not yet as we are not finished but I have one at the most risky place - top of north east wall, according to the WuFi analysis we did, so I might then start to worry.

[markharro]

Thanks for comments...so a few questions...as @Dan Fsaid its my understanding also that the Vaillant system has a humidity sensor also but I think that that is simply for the air in the room. My understanding is that the cooling risk is that you get condensation within the slab. Thats why I thought you needed some sort of sensor or sensors in the slab. But @jothseems to say no but then goes on to say fit the sensors in any event!?

[Dan F]

4 minutes ago, markharro said:

My understanding is that the cooling risk is that you get condensation within the slab

Condensation on the surface of the slab, not in the slab.  How much condensation you get here depends on the relative humidity in the room and the floor surface temperature.   The Vaillant calc will be quite conservative and assume your floor surface will be same as flow temperature, which in practice might not be the case.

 

6 minutes ago, markharro said:

But @jothseems to say no but then goes on to say fit the sensors in any event!?

No, to humidity sensors in the slab.  Yes to temp sensors in the slab, even if they may not be used to control heating.

[JohnMo]

Do you need anything, a thermostat that can do heating or cooling mode, run the flow temp at a fixed 15 degrees - job done. Little or no concern with humidity anywhere and a floor surface temp cool enough to manage house temps.

 

I did a drain down of an UFH loop in the summer and the water coming out was at 25 degrees.  So if you flow at or around 15 degrees you should be doing about 10 degs of cooling over the floor area.

[Conor]

I'm with @JohnMo.

 

we run ours just fully manually using the room stat timer. If you do get condensation, it'll be minimal, limited to exposed pipes and gone again within a couple of house after the system turning off. You'll only be using it when it's warm!!!

 

we had condensation when flow temp was set to 12c, at 14c there was nothing 

 

keep it simple, use your energy and time for something else.

[IanR]

I don't use my slab temp probes for cooling condensation avoidance, but I do control cooling flow temp from a calculation in loxone that works out the dew point from temp and relative humidity of internal air and the flow temp is set half a degree above dew point. 

It might reduce flow temp by 1° or 2° degrees at times from what you could do with a fixed cooling flow temp, so possibly not worth the trip.

 

As @joth says though, still worth putting the temp probes in the slab, if you can use the data. It's good feedback.

 

I use the temps from mine to allow some control if I'm getting effected by solar gain (during heating season). I've got two probes, one by some SW windows where the main solar gain is and one in a NE room that gets none.

 

When I start to see a couple of degree difference in the the two temps, I circulate the UFH (without heating or cooling) to redistribute the energy from the solar gain. 

 

It's more effective if I allow the solar gain areas to get a bit warmer than ideal, ie. 23°C or 24°C, and no one ever complains about the living areas being a bit warmer in winter. 

Edited March 18, 2023 by IanR

[Bramco]

13 hours ago, jack said:

I'm not sure whether it's a common feature of thermostats, but if you can get one with normally open and normally closed outputs, you could easily wire a summer/winter switch so that the right output is controlling cooling/heating.

 

I'm no expert, but I think a lot of them are simply a switch, so the switch is closed to indicate a call for heat, i.e. NO.  Also, there's normally only 2 wires going from the thermostat to the heating wiring centre, so you wouldn't be able to get both the NO and NC signal to the centre.  Wireless thermostats might be a way of doing what you suggest.

 

[Bramco]

13 hours ago, MikeSharp01 said:

I find they do need calibrating so I use a 5w Calorimeter and a traceable thermometer, put a few of them in at low temp 0 or thereabouts, and then let it track up, they can be several 10ths out so I mark the units and programme the curve fit, usually close to straight - y=mx+c, into the microcontroller - works a treat. 

 

Hi Mike,

 

For this type of application, I reckon if you just used them without calibrating, then you wouldn't know the difference.  If you were using them to measure something that needed an exact temperature, then it would be necessary.   For a central heating/cooling application, you're deciding on a setpoint according to how the house/room feels. if c above is 0.2 or 0.3 out, then your heating/cooling would kick in a bit higher or lower temperature and you'd adjust the setpoint accordingly.  Accuracy is great but maybe a bit of overkill for this application?

 

Simon

[ToughButterCup]

16 hours ago, DanDee said:

@markharro .... but really you need a way to manage the temp to avoid condensation rather than sensing when it happens

 

Quite the most polite way of saying  ...  Put the horse in front of the cart eh?

 

 But @DanDee, we do what we can because we can, because we're engaged, most of all because it's fun.

[MikeSharp01]

1 hour ago, Bramco said:

Accuracy is great but maybe a bit of overkill for this application?

You are correct but sadly I am just like that.

[joth]

Yes, in slab sensors are largely pointless for control purposes, my suggestion to put them in is purely because they cost nothing to add now but impossible to do later. 

I use mine mostly for logging/monitoring/diagnostics. It's nice to to see how warm the slab in each zone is getting, especially if you have fancy floor finishes or zoned controls and what to confirm how it's working. 

I do actually have mine set as an emergency shut off for the pump if the floor does get too hot. Also I can use it to dynamically control flow temp to eek maximum value out of overnight cheap rate (run a very high flow temperature initially and then back it off as the slab warms). But these are really not things I thought about until a couple years after the build.  Don't worry about using slab sensors on day 1

[MikeSharp01]

I know the great JS Harris, formally of this parish, tried to get everything to 1/10th of a deg and his hysteresis down to less than 1 degree but never quite managed it. I have this idea that a bit of AI from all the inputs, including sun / weather forecast should help close the gap a little.

22 minutes ago, joth said:

in slab sensors are largely pointless for control purposes,

I guess this depends on what the plan is. I though a bit of research would be fun so I put the sensors in the floor, about 2m in front of each south facing window, we only have north & south windows of any size really, so I could sense when the solar gain was happening, as @joth points out - will also know if the blinds are open or shut, and allow me to get ahead of the cooling in summer and in winter / shoulder months manage the blinds / pump the water from south to the northern loops to even out the temperature and spread the heat around a bit / use the northern loops to cool the southern loops etc. We have 43 tons of concrete in our slab so it's good chunk of store. I appreciate there are issues, after all it's heat in the building no matter where it is, and that it's just a hypothesis but it will be fun to play with when its up and running. 

[DanDee]

23 minutes ago, ToughButterCup said:

 

Quite the most polite way of saying  ...  Put the horse in front of the cart eh?

 

 But @DanDee, we do what we can because we can, because we're engaged, most of all because it's fun.

 

I'm struggling to understand how you got there, it looks more like Jeremy Clarkson going off the tracks whilst field drilling.

[ToughButterCup]

22 minutes ago, DanDee said:

I'm struggling to understand how you got there,...

 

You're at the end of a lengthening queue. 😔

[Alan Ambrose]

>>> I was planning to put in some tubes but didn't get round to it

 

That's what bag-ass SDS drills and masonry and diamond core drill bits are for. Even better if you don't mangle the UFH or DPC in the process.

[SteamyTea]

1 hour ago, MikeSharp01 said:

I have this idea that a bit of AI from all the inputs, including sun / weather forecast should help close the gap a little.

Just a lookup table is all that is needed.

Call it Bayesian Statistical Control if you like.