The attached circuit is a temperature activated switch using a 2N2222 transistor with the Base and Collector joined and connected to terminal B. The Emitter is connectd to terminal A. The variable resistor R2 value 10k is adjusted for the trigger point. I added the power supply portion of the circuit and the two transistors for switching On/Off depending on the application.
I constructed the circuited, tested it, learned Eagle CAD thanks to members on this forum, and got a board made. It does work.
In retrospect the base circuit has an operating adjustment via R2 of provides a temperature range of 32 deg F to 200 deg F (per original circuit description). For a single turn 10k pot, this range is too large and does not allow me to finitely adjust the temperature setting as much as I would like.
So here's the question - since R1 R2 and R5 seem to make up the reference voltage for the comparator, how can I determine replacement values for R1 R2 and R5 to establish a range of say 32 deg F to 120 deg F give or take some degrees as needed to match standard resistor values? I have one application that will be set at 50 deg F, another application that will be set at 90 deg F and another at 110 deg F. Right off the bat, it would seem using a 5k variable resistor increases the accuracy of the adjustment and then put the missing 5k on the other two resistors in some proportion. Then I thought "No this is about voltage division not resistor values". I've learned alot building this, but I don't know how to solve this on my own.

As a seperate issue, I can't figure out how to calculate the value of Rh which is the differential or "Swing" that controls the how far apart the on/off trigger is of the comparator. I'd like a fairly tight value of 2 degree F.

Help determining new values with explantions would be greatly appreciated.

 

The way to figure R1,2,5 is to find out what voltages correspond to the temperatures you want and calculate those 3 resistors as a voltage divider. I would start with R5 connected to +5V, decide to use 1 milliamp, and calculate for the voltages required.

Then add a resistor from the wiper to U1B, like 470k, and add a resistor from output U1B to the + input of U1B for the hysteresis (eliminating R8). If 470k is too high for the amp to work with, calculate the voltage divider for 10 ma and use 47k to the input of U1B.

Your circuit is wrong because Rh has no current limiter to work against and it is not in the measuring circuit. U1B has already flipped or flopped with a gain of 100 before your circuit gets to the hysteresis function!

Keep U1A as a comparator/driver, but don't use it as the hysteresis amp.

 

I'll try to do that when I get home tonight. Thanks for the input.