the right comparator

[xmat]

Hello.

I just wanted to ask what is the right comparator for the following task:

I want to feed into the comparator's (both) inputs two different signals (see picture).
The comparator should produce a pulse (from 0 volts to Vcc volts) for as long as the +input value exceeds the -input value.

Considering that there is a certain degree of noise to each signal that is fed to its corresponding input, I need a comparator that is quick and has a "flexible" hysterisis option;to prevent triggering of extra/multiple pulses at points A and B.

In simple words, I need a comparator that can handle variable inputs fed into both its inputs at the same time, and
has high response time while immunity to noise..

Hope you can help,

regards

xmat.

 

[williB]

https://www.electro-tech-online.com/custompdfs/2005/04/LM139.pdf
radio shack sells the LM339.. but it pains me to have to pay their prices..

Hysteresis control is accomplished through a feedback resistor..

 

[Styx]

problem is it is very low signal 200mV
As mentioned hysteresis (positive feedback) will set teh two trip levels, but it will also help with noise rejection.

the LM319 is extreamly good in noisy situations (I use it in a DeSat cct)

make sure your power rails are very stable (lots-o-decoupling caps) and the signal is as clean as possible.

 

[chemelec]

A Dual Comparator LM393
A Quad Comparator LM339
Both are Good Comparators.

However, I am Somewhat Doubtful about your "Pulsed Signal".
Is it Actually a Frequency?
Or Just a Varying DC Voltage?

As a Frequency, Unless the Pulses Occur Precisely at the same time on Both Inputs, there will be a Problem.

 

[xmat]

Thanks for the tips, guys.

My main problem is that for such low input signals, hysterisis reduces sensitivity (of an LM393) quite significantly (see figure below--blue).

While with no hysterisis there is the well known noise problem (red)

My goal is to have the gray-coloured pulse line, for a variety of signal levels (from 200mv to 5 volts).
Note that both input signals occur exactly the same time..so no such problem there.

Now, my question is if there is any comparator out there (and maybe with some extra elements connected to it) that can solve such problem.
It is really necessary not to "trim out"/reduce sensitivity over the entire dynamic area (200mV-5Volts) of the system;while maintaining its information as "reliable"/realistic as possible..

Hope that gave u an extra insight,

regards,

xmat.

 

[williB]

with the correct choice of feedback resistor , it should come pretty close to ideal..

 

[xmat]

Thanks for the help guys, really appreciated.