Thermostat design properties

Question

Cut-in, cut-out thermostats are commonly used with refrigeration devices in order to hold a consistent temperature range.

Their general operations is such that when the temperature by the probe reaches a particular temperature set as the upper limit, the thermostat cuts-in and the compressor is turned on providing cold air.
And when the temperature by the probe reaches the lower temperature set point, the thermostat cuts-out and turns off the compressor.

What materials are used in within an analog or mechanical cut-in, cut-out thermostat? Likewise, how is the internal circuit designed such that one probe feeds both cut circuits and what control is put in place for the cut-out to override the cut-in signal?

To help scope this question: based upon the materials used in an analog cut-in, cut-out thermostat, would it be reasonable to attempt to test and adjust the set points of a thermostat by using jars of water at known temperatures and measuring the output off of the relay within the thermostat?

Answer 1

A bimetallic strip is used. Two different metals like steel and copper expand at different rate and a strip of the two bound together as result bends with temperature changes. A contact placed at the end of the strip will close when the temperature is right - and the temperature can be tuned by turning a screw that pushes the strip closer or farther from the contact.

In simpler systems the natural inertia of the system is used to create the hysteresis - as the refrigerator compressor works, it takes time for the temperature drop to reach the bimetallic strip, and switch it off, so the temperature is brought below the cut-off point by some hard to control factor.

In more complex systems, a two-level switch (either through two strips or just mechanism that shorts two switches at two different temperature levels), in connection with an analog RS switch (usually based on a relay) creates a better-controllable hysteresis.

As temperature is too high, both switches are engaged and so the cooling compressor starts. Temperature drops.

At cut-in temperature the "On" switch disengages, but power to the relay is still supplied through the "Off" switch.

At cut-out temperature the "Off" switch disengages, and the cooling process stops.

With temperature rising, the "Off" switch engages, but since the relay is open, it doesn't supply power to the coil.

Further rise causes the "On" switch to power up the coil and the relay switches both the compressor and the "Off" switch circuit on. With temperature drop, the "on" switch will disconnect but the coil powered through the relay will "remember" the state until "Off" opens.

(and I'm sorry, but I don't know about calibrating thermostats with jars of water. My vote would be "against" as they are rarely submersible.)