[SOLVED] how to design a thermostat without Microcontroller?

[Mohammad_Pirzadi]

hi every one
I want to have an electronic thermostat when the tempearture reaches to 37 °C turn a relay Off.
do you have a circuit?
thanks

 

[KlausST]

Hi,

I want to have an electronic thermostat when the tempearture reaches to 37 °C turn a relay Off.
do you have a circuit?

And never turned ON again?

***
Why "electronic" but "no microcontroller"?
Why not mechanic?
What relay? Voltage, current?
What supply voltage is available? what current?

***

Klaus

 

[d123]

Hi,

How easy or simple? I have this circuit, not fantastic but functional:



And this is two circuits pasted together to give the general idea:



I'm sure there are other solutions and the above can be improved upon a lot.

For no microcontroller, and for it to be "easy" to put together, you need at least a reference, a temperature sensor, a comparator, a BJT or MOSFET and a relay, then stuff like passive components and fuses.

 

[betwixt]

Anything that changes resistance with temperature or produces a voltage changing with temperature can be used to drive a comparator. If you want to avoid a hard switching at one particular temperature so the relay doesn't 'chatter', add some feedback so the upper and lower temperature points are slightly apart.

Brian.

 

[Mohammad_Pirzadi]

Hi,


And never turned ON again?

***
Why "electronic" but "no microcontroller"?
Why not mechanic?
What relay? Voltage, current?
What supply voltage is available? what current?

***

Klaus

hi Klaus
it should be ON again after temperature goes below 37°C.
I'm not interested in microcontroller because my studies are in Photonics and semiconductors for 6 years. moreover I need a simple circuit and microcontroller increase complexity.
it should be elctronic because I hate mechanic [lol].
the relay is used to switch the heater OFF.
The power supply of the heating source is 220V AC and thermostat should switch it OFF. the power supply of thermostat is 12v but if you have an idea to working it in 220vAC plz give it.

- - - Updated - - -

Anything that changes resistance with temperature or produces a voltage changing with temperature can be used to drive a comparator. If you want to avoid a hard switching at one particular temperature so the relay doesn't 'chatter', add some feedback so the upper and lower temperature points are slightly apart.

Brian.

Hi Brian.
this thermostat shoulde be placed into a incubator. when the temperature in the box reaches to 37°C the thermostat switch the heater OFF.

 

[dick_freebird]

If you know the specific temperature and have no need to
change it, you probably can't do better than a Selco 'stat
from Digi-Key or wherever; just a bimetal snap action
thermal switch, sold in increments of setpoint, cheap.

However for higher power you'd want to use one to run
a bigger relay and make sure you protect the contacts.

 

[FvM]

Using a simple on-off controller (also called bang-bang controller) for the incubator will result in a relative large periodic temperature variation due to the dead time between sensor and heater. To reduce variation range, you need to follow a different controller concept, e.g. PI controller with pulse width modulator.

 

[KlausST]

Hi,

it should be ON again after temperature goes below 37°C.

Please review your specification: ON above 37.00000°C, OFF below 36.99999°C. Are you sure you want this?
Usually it makes no sense to decide ON/OFF in micro degrees. Noise will cause malfunction.

microcontroller increase complexity.

This is a too general statement, thus I can't agree.
From hardware view: what can be less complex than an 8 pin microcontroller with inside clock and temperature sensor?
You may adjust timing and temperature levels precisely, without additional hardware...

the relay is used to switch the heater OFF.

I see the main job is to switch ON/OFF a heater. But why did you specify a relay? Why no MOSFET, BJT, triac...?
Btw: a relay is "mechanics"...

the power supply of thermostat is 12v but if you have an idea to working it in 220vAC plz give it.

Does this 12V supply already exist?

It's quite unusual to run such electronics from 220V AC.
12V DC is not unusual, but 5V DC, or 3.3V DC is more common.

We don't know your "real" requirements. Like cost, size, quantity, environmental (temperature, humidity, dirt, hygienic..), efficiency, power...
It may be a wall wart laying around, or an industrial open frame SMPS installed in a sealed metal box...

Klaus

 

[d123]

Hi,

You can use the (cheap, small footprint) TL431 or TLV431 as a comparator (**broken link removed**) and the (cheap, small footprint) LM334 as a temperature sensor, or something simple (and smaller footprint than lm334 temp sensor circuit) like the LM35 Centigrade temperature sensor (cheap, easy to implement and small footprint). Three parts plus passives: reference, comparator and temperature sensor. I don't recommend using a BJT or a diode as a standalone temperature sensor, the BJT one needs an op amp to amplify the 60mV or so that its VBE changes across a wide temperature range... Microchip TC7106/7 datasheet shows how to use a simple (and cost, 1 cent) 1N4148 as a temperature sensor (with the ADC).

 

[Mohammad_Pirzadi]

Hi,


Please review your specification: ON above 37.00000°C, OFF below 36.99999°C. Are you sure you want this?
Usually it makes no sense to decide ON/OFF in micro degrees. Noise will cause malfunction.


This is a too general statement, thus I can't agree.
From hardware view: what can be less complex than an 8 pin microcontroller with inside clock and temperature sensor?
You may adjust timing and temperature levels precisely, without additional hardware...


I see the main job is to switch ON/OFF a heater. But why did you specify a relay? Why no MOSFET, BJT, triac...?
Btw: a relay is "mechanics"...


Does this 12V supply already exist?

It's quite unusual to run such electronics from 220V AC.
12V DC is not unusual, but 5V DC, or 3.3V DC is more common.

We don't know your "real" requirements. Like cost, size, quantity, environmental (temperature, humidity, dirt, hygienic..), efficiency, power...
It may be a wall wart laying around, or an industrial open frame SMPS installed in a sealed metal box...

Klaus

hi Klaus. thanks for your kindly replies.
I want to design an INCUBATOR. I provided a box which the heating source (40W light bulb), a little DC motor for rotating eggs and a thermostat embedded in.
the temperature in the box should be around 37°C and the accuracy is not too important. I want when the temperature reaches to 37°C the light bulb be turned OFF and once the temperature goes below 37°C the light bulb be turned ON.

 

[KlausST]

Hi,

a little DC motor for rotating eggs

Please understand: we all know what you want to use it for, but if you want us to help you with your design we (additionally) need technical/electrical specifications.

Klaus

 

[andre_luis]

I want when the temperature reaches to 37°C the light bulb be turned OFF and once the temperature goes below 37°C the light bulb be turned ON.

You migh be aware that thermostats are commonly based on hysteresis working principle, on which the shutdown temperature is a bit higher than the turn on temperature, when used as heat control. Although the propagation of temperature from the heat source toward the sensor, by itself, somehow provides this feature, as stated above, this gap is useful both to preserve the life span of the whole system and also to turn it more immune to unwanted triggering due to measurement noise.

 

[asdf44]

You asked for a circuit here is a circuit:

Use a temperature sensor IC like LM50 and send its output to a comparator. The LM50 puts out about 0.87V at 37C.

https://www.ti.com/lit/ds/symlink/lm50.pdf

Page 11 shows a complete thermostat circuit with a voltage reference and comparator. R1 and R2 should set a divider to 0.87V. R3 should be about 100X as large as R2 to set the hysteresis.

Almost any comparator or voltage reference could be substituted. The circuit might benefit from a relay driver in the form of a mosfet (See NUD3112)

Also you can just buy these controller for cheap:
https://www.amazon.com/Temperature-...ocphy=9001981&hvtargid=pla-395718471013&psc=1

 

[Mohammad_Pirzadi]

Design of an incubator temperature controller

Hi everyone. The previous post have requested to be deleted because was not complete.
I have a box which is used to be an incubator. A heating source(40W light bulb) and a Thermostat which controls the temperature. The needed temperature for eggs should be between 36 and 37°c. It means if temperature reaches up to 37,the thermostat turns the heater off and if goes below 37 the thermostat turns a light ON again. The power source for light bulb is 220v AC. I do´nt want to use a microcontroller. Thanks for your kindly replies.

 

[andre_luis]

It means if temperature reaches up to 37,the thermostat turns the heater off and if goes below 37 the thermostat turns a light ON again.

You did not realize yet that there is a difference between the desired working and the accomplishable working. If you want that a temperature remains as close as possible of the target value 37 degree, you need to consider a better approach, as for example the PI control above mentioned at post #7, on which heating addition are made softly. Another point you should consider is that in a heating project, the proper sizing of the heat source and the load (thermal capacity) should be compatible with the allowable thermal range, ie you not only take care to inject heat but also to have adequate thermal isolation, so that with the addition of small packets of heat is enough to keep the air temperature stable.

 

[dick_freebird]

Here's one other "out of the box" (or rather, retail-in-box)
solution: get a cheapo forced air heater. They have the
thermostat(duh) and I picked up a couple the other day
for $12 each (cheaper, it turns out, than many snap-action
thermostats or semiconductor thermostats). You'd have to
find the range and then make it permanent-ish (like pull the
knob off). Fan forced will cut the hysteresis to a minimum
(snap-action only, not embedding the thermal transport
as an adder to the toggle-swing).

Now if the designing of it is the goal, and not just getting
on with the poultry business, then maybe this is irrelevant.
But might be worth a look just on the basis of "cheap and
effective".

 

[Mohammad_Pirzadi]

You did not realize yet that there is a difference between the desired working and the accomplishable working. If you want that a temperature remains as close as possible of the target value 37 degree, you need to consider a better approach, as for example the PI control above mentioned at post #7, on which heating addition are made softly. Another point you should consider is that in a heating project, the proper sizing of the heat source and the load (thermal capacity) should be compatible with the allowable thermal range, ie you not only take care to inject heat but also to have adequate thermal isolation, so that with the addition of small packets of heat is enough to keep the air temperature stable.

dear Andre, please read my post then spread your thoughts! I think you did'nt realize yet what I need! It's simple. believe! the incubator is a box which a 220V AC light bulb is embedded in as a heater and a thermostat controls it.no additional device or accuracy is needed! the temperature should be in 36-37 degree.

 

[d123]

Hi,

After multiple circuit suggestions, have you done any research and/or simulation/prototyping of your own yet?

I don't think 36ºC on, 37ºC off is a sensible design approach, it is a very narrow window - the relay will be going on and off every second and the light bulb will be flickering like a strobe light. Presumably it can be extended to e.g. 32ºC on, 40ºC off or similar wider range.

 

[FvM]

I don't think 36ºC on, 37ºC off is a sensible design approach, it is a very narrow window - the relay will be going on and off every second and the light bulb will be flickering like a strobe light. Presumably it can be extended to e.g. 32ºC on, 40ºC off or similar wider range.

Even with very small hysteresis, the control system will unlikely act very fast. Reason is the "plant" transfer function, which can be modelled as combination of dead time and multiple low passes. If the temperature sensor e.g. detects exceedance of the 37 °C set point value and the heater is switched off, the hot light bulb will further heat up the incubator. The actual value over shoots respectively and similarly under shoots after heater switch on.

It might be that with careful selection of heater power and well considered sensor placement, the temperature variation can be held in an acceptable range. Unless you are able to predict the control system behavior in an exact simulation (very unlikely, I guess), you need to check it empirically.

 

[Mohammad_Pirzadi]

Hi,

After multiple circuit suggestions, have you done any research and/or simulation/prototyping of your own yet?

I don't think 36ºC on, 37ºC off is a sensible design approach, it is a very narrow window - the relay will be going on and off every second and the light bulb will be flickering like a strobe light. Presumably it can be extended to e.g. 32ºC on, 40ºC off or similar wider range.

I do'nt have any circuit to be simulated:thumbsdown:
accuracy is not important. thermostat should turn a relay ON and the relay turn the light OFF. LM35 or a NTC is a good choice but a functional circuit has not founded yet.