RF signal extractor comparator charging up

[cmorgan]

I've got a comparator circuit used turn small voltage changes in an RF signal from an RF detector into digital outputs. In this case I'm trying to extract a 1Mhz signal out of the RF detectors. It's based on some circuits I've seen for others trying to extract ADSB singles.

The circuit, attached here, works well except at times it stops working, I think because it charges up due to the diode, then after touching the circuit, in particular the diode side, or waiting a number of seconds, it starts working again.

I tried to resolve by adding a resistor between the +/- of the comparator, say 100k or 200k, but this entirely breaks the functionality of the circuit that I'm testing here, although it works with LTspice.

Thoughts on how to resolve?

[w2aew]

I've got a comparator circuit used turn small voltage changes in an RF signal from an RF detector into digital outputs. In this case I'm trying to extract a 1Mhz signal out of the RF detectors. It's based on some circuits I've seen for others trying to extract ADSB singles.

The circuit, attached here, works well except at times it stops working, I think because it charges up due to the diode, then after touching the circuit, in particular the diode side, or waiting a number of seconds, it starts working again.

I tried to resolve by adding a resistor between the +/- of the comparator, say 100k or 200k, but this entirely breaks the functionality of the circuit that I'm testing here, although it works with LTspice.

Thoughts on how to resolve?

Could be that the common mode voltage is going too high since you're not establishing a stable DC operating point for the inputs.  The common mode input range goes from VSS to VDD-1.1V.  Maybe a large resistor to ground at the common input point, like TP2 to ground.

[cmorgan]

Hmm.

We've got a 1M to ground, this is the circuit that feeds into the comparator attached below, sorry I should have included this with the other one.

Would it help to measure the +/- into the comparator to see what is occurring there? That was one of the next ideas I had just to see what voltages were there (assuming the probes on those pins won't affect the voltage/capacitance enough to affect the circuit).

Regards,
Chris