Comparator rising/falling edge propagation delay

[patrickfreninger9]

Posted on May 22, 2015 at 11:17

I am using a STM32L152RC microcontroller and have to detect the length of pulses of about 0.7 V and 3 ?s using the COMP2 comparator. At 2.7 V VCC the worst case propagation delay is 2 ?s in fast mode. A minimum delay is not specified. The worst imaginable interpretation suggests that my pulse could be shortened or streched to 1 ?s or 5 ?s.

Can I make any assumptions about the equality of the propagation delay for the rising and falling edge? Note that my overdrive is not 10 mV as used for characterization but at least 300 mV in both directions. How does this influence the worst case propagation delays? #stm32l1 #comparator

[waclawek.jan]

Posted on May 25, 2015 at 08:22

Yes it's probably quite better than the max value given, but it's unlikely the manufacturer will guarantee you that unless a significant incentive (i.e. unless you represent a massive buying power). One of the reasons is that making what is basically analog circuit on relatively cheap digital process is tricky and full of tradeoffs.

If you need to ensure a well-defined pulse shaping, you would probably need to resort to an external comparator with tighter characteristics. JW