Pipe thermostats

[oldbutnotdead]

Very irritating things. Currently using a pair of (admittedly cheap) Corgi pipe stats on a woodburner gravity circuit, wired so the heating pump will not run unless flow and return are 50 deg C or hotter.
Trouble is the accuracy is wild and the hysteresis is worse- about 15 deg C- which is not ideal. Searching on here brings Honeywell up as the best of breed but at £80 for a pair I'd rather buy an Arduino and code my own system. Anyone know where the things can be had for a more sensible price- its a bimetal switch ffs, £40?
TIA

 

[Jackrae]

Buy yourself a pair of these and your problems are solved.
https://www.ebay.co.uk/itm/220V-Ele...014132?hash=item41fcf86f74:g:UOAAAOSwL1RcirvM
I've been using one for several years with no problems

 

[oldbutnotdead]

Ahh, yeah they look v promising, ta

 

[oilhead]

If you are checking for the operating differential by turning the knob, then you will never get a good reading. Why not trust the component and set at indicated temperature and let it work properly. If they are too high/low in practice, it is just a simple tweak to get closer.
After all, these units work competently on DHW cylinders, and your application is not exactly something that needs controlling to 1/5th of 5/8ths of a degree. Similar components are available from many merchants under different names at a much cheaper cost than honeywell

 

[oldbutnotdead]

The.accuracy is annoying but (as you say) can be worked round. The hysteresis is the expensive bit- to make the return stat open at 50 deg i have to have it set so it doesn't close til return reaches 65 odd. Which doesn't happen til flow hits 85 odd, shame the burner starts kettling before that point.
For this particular do zero hysteresis will be best, i don't really want to spend the next month sampling thermostats when shared wisdom could point me at what will work

 

[oilhead]

Are you checking the hysteresis by turning the stat up and down? If so, this is not a valid test as the bi metallic element will not operate that quickly, but work much closer on a fixed setting. I've seen many simple pipe stats on such systems that operate effectively. How exactly are you wiring them if both stats need to be 50o ?
If you want mechanical stats that are more sensitive, a capillary stat (www.bes.co.uk part no 19410) will give you better performance as the bulb can be attached firmly to the pipe and insulated.

 

[oldbutnotdead]

No, leaving stats set and monitoring temperature with contact thermometer. And thinking about it, these cylinder stats are doing exactly what they're designed to do (temp drops sufficiently below setpoint to make it worth firing the boiler. Boiler keeps firing til temp a bit above the setpoint). In my application, as soon as set point is achieved, heat is removed from the system- if there's enough wood on the fire then temps stay above setpoint, if there isn't then i need the pump to shut off and let the gravity circuit get back up to temp.
So I need a thermostat that makes and breaks at the setpoint (+-1 degree) not a cylinder stat. My bad for not thinking it through properly...
EDIT They're wired in series (both flow and return have to be at or above 50 deg c before any power goes to the pump- that via the 2 port valves for downstairs and upstairs. Rough description- mains in to RET, out of RET into FLOW. Out of FLOW into 2 port valve microswitches (those in parallel). Controllers for downstairs and upstairs open their respective valve when the area is calling for heat- if set temp has been achieved then pump fires, if not then gravity circ only until set temps achieved)

 

[Jackrae]

<£20 for answer No2 and they have digital displays to show what is happening on both lines
These are the instructions and you will see the differential on the controller can be set between 0.3 and 10 centigrade degrees
https://www.homebrewbuilder.co.uk/stc1000 instructions.pdf

 

[oldbutnotdead]

Excellent spot- ta, they look very appropriate (plus always nice to have something to look at)
EDIT There are so many lovely toys that company make

 

[oilhead]

Maintaining 50o on return will surely be difficult. Once the temperature starts the pump, then you are going to pull cold water through the return, thus switching it off almost at once. A wider hysteresis would help smooth out the pulses. I see you now state you are using cylinder stats. These have a void between the surfaces, whereas a pipe stat has a solid base in contact with the surface being monitored.

 

[oldbutnotdead]

Maintaining 50o on return will surely be difficult. Once the temperature starts the pump, then you are going to pull cold water through the return, thus switching it off almost at once. A wider hysteresis would help smooth out the pulses. I see you now state you are using cylinder stats. These have a void between the surfaces, whereas a pipe stat has a solid base in contact with the surface being monitored.

Yes it is tricky. There's built-in lag in the system (remember the stats are on the gravity circuit, not the pumped circuit otherwise return would never hit 50)- ret of 50 = gravity flow at about 70, if there's not enough fuel to maintain 70 flow with the additional load of the rads then flow temp will drop eventually as will return.
The stats i'm using are sold as cylinder or pipe- for pipe fixing the lugs for the springs are set so there is contact between the (presumed) wax container and pipe