Operational amplifier as a comparator

Status
Not open for further replies.

sxy

Member
Hi
What happens in Operational amplifier used as a comparator;
if V1=V2 ?
i.e. the two inputs are equal.
Thanks a lot
 
Last edited:
All OpAmps or Comparators have offset voltages, input referred, and noise.

And they have parasitic capacitive and PSRR related problems.

So if V1 = V2, the + and - inputs, then out put can either go positive or negative,
due to offset, or oscillate due to noise and these other effects.

Note in actual practice V1 = V2 + Voffset + Noise (ignoring some of the other
effects/parasitics) , Voffset can be either positive or negative.

To get rid of noise effects one generally uses hysterisis on the comparator/OpAmp.






Regards, Dana.
 
Last edited:
Thanks for your answer.
How do we know from the output that the two inputs are equal to each other?
thanks a lot
You don't, even with a dedicated comparator - it's either ('equal or higher' and lower), or ('equal or lower' and higher), there's only two possible states.
 
How do we know from the output that the two inputs are equal to each other?
If you use it as a linear amplifier in an appropriate configuration instead of a comparator (and it is properly calibrated), you can then see when the inputs are equal by reading it with and ADC, or using two comparators in a "window comparator" configuration.

Or it the "equality" is not too critical, just use two comparators directly, to get equal / greater / less output combinations.
 
Example window comparator. Make the difference between the two reference inputs as small as needed for the "equal" state.

Or use the two outputs separately and logic to determine equal or which is higher.

Note for that circuit to work you must use comparators such as LM311, 323, 339 etc. that have open-collector outputs.

 
Thanks for your answer.
How do we know from the output that the two inputs are equal to each other?
thanks a lot

Posts 4 and 5 answer that question........

If you want to eliminate noise (but would still have Vos uncertainty) then
supercool your comparator and its circuitry to absolute zero. Of course that
does not necessarily eliminate popcorn noise or interaction with galactic
noise induced problems (like neutrinos, gamma rays.....other yet to be
discovered particles).


Regards, Dana.
 
When talking about analog circuits, there are two terms we never use: "exactly" and "zero noise"

You can never tell anything exactly and there is always noise.

You can tell the value or relative value of something within a tolerance if specified, and you can refer to noise below some limit.
 
I would set up your opamp as a differential amp. When the output of the diff amp is at, or near, zero, then the two input voltages are equal.

Set up the gain of the diff amp to be whatever is needed for the system requirements.
 
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…