1
\$\begingroup\$

I applied the same signals to both inputs of the LM393P double comparator. The number 2 comparator output gave the expected result (first photo).

enter image description here

However, when I apply the same signal to comparator 1, there is a distortion in the output 1 (second photo).

enter image description here

Why do output jumps (like a sawtooth) occur when the input signal falls below the reference signal? At first I thought the device was broken, and I installed a new LM393P, but the result has not changed.

What am I doing wrong here? And why am I getting this error only from output 1?

enter image description here

Note: Since the two comparators are in one integrated unit, the supply and ground line are common. I had to draw it as two separate comparators in Proteus.

\$\endgroup\$
1
  • 1
    \$\begingroup\$ The applied signal cannot be from a potentiometer, so the schematics do not show the problem. From where does the square wave come from and what are the parameters for it? It looks slightly drooping so is there some capacitor AC coupling? \$\endgroup\$
    – Justme
    Commented Jan 25 at 21:00

1 Answer 1

3
\$\begingroup\$

You are applying a voltage beyond the input range so bad behavior may be expected.

The LM393 is only guaranteed to operate down to 0V at the input, and you are applying a negative voltage.

That's assuming the numbers are showing about a real -60mV input. If the probes are x10, and the scope is set to x1, so the actual input is -600mV you would be outside the absolute maximum input ratings and things like phase reversal can occur (as well as damage to the chip if the current is not limited).

enter image description here

As to why the results are different, it could be some asymmetry due to the layout of the chip, in particular in the second case mentioned above, with unexpected currents flowing through the isolation junctions in the chip.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.