I'm working on a new design that currently uses an ADCMP607 to convert user-input sine waves to square waves for timing input. There are lots of signal-conditioning front-end bits ahead of the comparator, ending with the ADCMP607. Despite having +30dB analog gain in front of the ADCMP607 (including some careful limiting using the AD8037 clamp amp), it seems clear that the comparator is dominating the tempco of the overall circuit. Spraying cold onto the components one-by-one, starting with the JFET input opamp, showed little effect until I hit the ADCMP.
Some other ADCMP comparators (such as ADCMP551 and ADCMP572) are specified with tempco's, any of which would be good enough for my application. However, none of those are as good a match otherwise to my application (which needs the programmable hysteresis pin, only needs a single comparator, and ideally keeps the CML outputs).
Is there any data on what to expect for tempco's from the ADCMP607?
Thanks in advance,