TLV3202-Q1: AC Hysterisis

https://www.tina.com/blog/comparator-with-ac-hysteresis/ says:

Place a small capacitor to feed back the output edge transition into the comparator non-inverting input to provide momentary positive feedback. This sharpens the comparator response as well as reducing the tendency to “chatter” at the switching point. If the RC time constant of R1C1 is << the waveform period, the comparator trip point hysteresis will have settled back to 0V by the time the next threshold-crossing takes place. (Circuit is created by Neil P. Albaugh  TI-Tucson)

Hello Saraswathy,

Keep in mind that the AC hysteresis is 'temporary', and is only good for the period of time constant of the RC network.

The theory is the same as the DC resistor hysteresis - to bump the reference away from the threshold on the first crossing to stifle any further transitions.The ratio will be the equivalent capacitor Z and the input source resistance (R1).

But unlike the DC resistor hysteresis shift, the AC threshold quickly "recovers" back to the original threshold regardless of the current output state (time constant). So this can lead to "bursty" behavior on noisy signals. The actual cap value tends to be determined by trial-and-error in the actual system (hard to calculate).

Typically the feedback capacitor is in the 10's to 100's of pF. Making the capacitor too large can cause negative transients to appear on the inputs, so care must be taken to ensure that the resulting input feedback does not violate the input voltage range.

Yes; with AC hysteresis, you do not always have fixed switching thresholds. Assuming that the circuit has recovered, any signal that crosses the DC threshold will trigger the comparator to switch. But directly after the switch, the threshold moves (even up to the power rails), and slowly recovers back to the DC threshold.