Thunderbirds are go…

[Cidermanrolls]

Well, not thunderbirds, but inkbirds in my case.
Having noticed the wide temperature band I was getting with my electric fan heaters I decided to take the plunge when I spotted digital thermostats on offer over Black Friday weekend. I have now retrofitted these to the heaters in three greenhouses.
The key benefit is that most of us want the heaters to control to a minimum. If the heater stays on significantly above that then energy is wasted. For example, the heater in my 5C greenhouse was not turning off until the temperature reached 10C, an awful waste and expense. Last night each greenhouse saw fluctuations of <1C.
Fitting digital thermostats requires a decent knowledge of electrical safety, so is not for everyone.
I’m pretty sure the controllers will pay for themselves in one winter.

[Davey246]

The hunting of the temperature depends on lots of things.
Low power heaters are far better than high, but use more than one - you are then circulating more air at a lower temperature from the heaters. Position of the sensor for any thermostat is also very important, and what principle they use - internal heater thermostats are usually bimetallic strip types that sit in the in-coming air flow to the heater itself.
Not all thermostats are equal. They too vary in where the sensor is and in what it is. Also very important is what the switching part of the thermostat is - idealy this needs to be an SSR (solid-state relay) as they have an all but infinite life. Anything with mechanical contacts is bad news, How things are switched is also important - some electronic thermostats do not switch directly mains V to zero (off) in one "instantaneous step", which can give problems with fans, especially if, as they very frequently do, they use shaded pole type motors.

Here, two small domestic fan heaters set at 1kW each, at opposite ends of an 8x12 greenhouse, at bench level, controlled by a home-built thermostat using a disc thermistor on a flying lead as sensor, laid on top of pots in the middle of the greenhouse, controls temperature to something like 1C variation at the position of the sensor. Around the rest of the greenhouse, at bench level, the tempertaure varies but not a great deal more - maybe 1.5C total variation. Obviously, as heat rises, going higher in the house gives a warmer controlled temperature by a degree or so, lower down is less affected by temperature variation, probably due to the heat-sink effect of the floor.

[Paul in Essex]

I think they are brilliant, well worth the cost - not that they are that expensive. I have my greenhouse heater and propagaotr set up with them and they are extremely efficient. I have a small weather station that sends info to my pc and from there onto the interweb and here is today's snapshot of the temp in my greenhouse.

[Davey246]

Do not get too wedded to fancy electronics and temperature measurements as how temperature is sensed is another can of worms due to hysteresis of the sensors. You are also only measuring just one spot.

If you REALLY want to know what is going on, buy a thermocouple meter (insanely cheap these days), and use a bare thermcouple as the sensor. A bare thermocouple is the joint between two very thin wires that are bare for a few mm beyond the insulation.

So-called home weather stations that I have seen over the years most frequently have the temperature sensor element encased in plastic, often sleeved with metal. I have had one apart to use the tiny sensor bare and the hysteresis introduced by all of the crepe around the sensor has been horrendous, in other words, it does not react to temperature variations in situations such as greenhouse heating control in any useful way at all, it just does not see the the real variation.

The fact that you are only seeing only 5-6 switching cycles every hour makes me very highly suspicious, although if the outside temeperture was close to 5C, this may be real control. Down around freezing outside, good temeperture control inside any greenhouse would require tens of switch cycles per hour.

[Paul in Essex]

Well those fancy electronics work just nicely for me. With my shaky hands making anything electronic is a non-starter as I would most likely solder my fingers to the bench.

[Davey246]

The main thing is to think about what you are measuring and conrolling etc. and how.

Whether it is representative of your weather station or not, if you think about the type of "sensor" that I described above, the actual sensing element is measuring the temperature of a piece of plastic (a reasonable thermal insulator), which is following, to some degree, the tmperature of a metal sheath, which is following, to some degree, the actual variation in air temperature.

At low and/or fast variations in air temperature, even "accurate" mercury or alcohol-in-glass thermometers have huge hysteresis.

[Paul in Essex]

Actually the main thing I want to think about is how simple the whole system is and how easily I can keep tabs on it. I am not remotely interested in the nuts and bolts of how the system works or the absolute accuracy of the resolution. All of that is a layer or three of detail that I neither understand nor care about. If you understand it and are interested in it and get pleasure from making and setting up a more accurate system then that is great but there are other viewpoints.

[Davey246]

Not caring is costing you money and, potentially, plants, especially with such a very low switching frequency.
I do not bet, but am tempted to, that the temperature variation in your greenhouse is a long way different to what you think - the highs are higher and the lows are lower.
Being cooler will save on heating, but the highs will cost you more. The temperature variation if quick, will not be felt by the plants - they have huge hysteresis, but a 10 minute cycle will be felt by the plants.

A good off-the-shelf electronic thermostat/controller will be switching every few seconds to hold a steady temperature. But that is absolutely massive overkill, although that said, for something under £100, such a PID controller can be tuned by hand to set whatever cycle you like.

[Cidermanrolls]

I paid just £25 each for the Inkbirds. I think that is a good price for what they do.
My main concern was limiting the highs.
From what I saw this morning, using the same max min thermometers as before, the swings have been massively reduced.
I accept I do not know exactly the temperature everywhere in my greenhouses, but I do know my heating costs have just been reduced. From experience I know what minimum temperature I am happy to see on those thermometers and as I’ve not changed the heaters, I don’t plan to adjust that minimum target. What I’ve trimmed are the unintended highs.

[Stuart]

I'd go with Paul's thinking on this, I bought three Inkbirds before last winter when they were recommended on the Forum. Simple to set up and very accurate when checked with max/min thermometers, no wiring, just plug them between the socket and heater. They were so good I bought three more and they control six 2kw heaters. They're set to switch on at 3.5C and off at 5.5C though the differential is adjustable. They override the internal thermostats or, in the case of Parwins, external thermostats and have made my winter heating much more accurate than before. I wouldn't have known about them if I hadn't read it here.

Stuart