Author Options:

I need some help with comparators. Answered

Upfront I apologize if this seems like a dumb question, but both Instructables and google have turned up nothing useful for me. I have a need for a resistance comparing switch circuit. I can't find anything about resistance comparators anywhere except articles that you have to pay to read. I need something that will compare a set resistance to a resistance that will vary, I want it to switch off when the tested resistance is higher then the set one and off when its the reverse. I have no clue what to do.


Here is my original circuit, hacked up with a bit of hysteresis added. The cap is optional, but putting in a low value (47pF) can be beneficial.


Incidentally look up "Skin Galvanometer circuits" on Google to find the kernel of what you want.


you could use a standard comparator circuit, modified to input as shown in the pic below, such that

Vref = whatever reference voltage
Rtest = your DUT
REF = fixed resistance
R2 are a pair of matched resistors (use a good ohm meter to match 1% nom values to 0.1% or better)

the circuit shown would be used to power an SSR or high-ohm coil relay.

best wishes

Note that this circuit has no hysteresis, so when the test resistance is within the noise margin, the circuit will toggle unexpectedly... To improve it, one would use a high and low threshold (search for "threshold comparator circuits for more on the theory) to offset the uncertainty in output when Rtest approaches REF.

The purpose is to discriminate between flesh and metal, so anything within the noise margin shouldn't be a problem, but just out of curiosity, how would I go about making it histerize? Also, if I wanted it to instead switch on when the tested resistance is higher, would I just switch the comparator inputs?

tsk tsk..that comment about hysteresis and threshold comparators was homework pour vous, (for you)

See Goggle, Yahook, wikipedia, etc., and enter the magic words... Come on, yer half way thayre, Laddie....

Hmm. well... hysteresis seems to me that they've thrown in a resistor in between the input and the compared output. As for threshold comparators, either Google isn't forthcoming or I haven't a clue what I'm looking at.

What you want is a bridge circuit like this I've drawn it with an op-amp, which will work for this application, and not drawn in any feedback components. R1 and R3 are fixed, and will present a fixed voltage to the -ve input of the opamp. R2 is fixed, but R X is your unknown value. If R1, R2 and R3 are equal, then if Rx is smaller than R3, the output will be clamped to 0V. If RX is bigger than R3, then the output will snap towards 10V. Some hysteresis is necessary to stop it oscillating.