This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM393-N: A Capacitance on a comparator output

Genius 13570 points
Part Number: LM393-N
Other Parts Discussed in Thread: LM393

Hello,

 

The below application notes recommends to add a series resistance when putting a large capacitance on a comparator output.

 

http://www.ti.com/lit/an/snoaa35/snoaa35.pdf

 

Then I would like to know up to how much capacitance can be put on the comparator output without a series resistance.

Is there any theory or do we have any case how to think about it?

Now I’m looking at LM393-N, but any part number case would be helpful.

 

Regards,

Oba

  • Hi Oba,

    capacitive loads are unwanted because in combination with the pull-up resistor they form a low pass filter and introduce an additional switching delay. 5k1 and 50pF, for instance, give an additional delay of 260ns when the output toggles from low to high.

    When the output toggles form high to low, on the other hand, the capacitive load causes a brief short circuit. This can result in a huge ground bounce, in an erosion of signal ground integrity and in a massive decrease of stability. Spontaneous oscillation of the comparator can set in.

    I think up to 50...100pF might be acceptable. But with higher capacitive loads I would limit the short circuit current to <20mA by inserting a current limiting resistor.

    Kai
  • Hello Kai,

    Thanks for your reply.

    >I think up to 50...100pF might be acceptable.

    Is there any theory to calculate these value like 50 and 100pF that might be acceptable?
    Or is it just from like a experience?

    Regards,
    Oba

  • Hi Oba,

    as I said, the pull-up resistor and the capacitive load form a low pass filter and give an additional delay. See section 8.2.2.4
    of datasheet of LM393 (not LM393-N !).

    To be honest, I never allow any capacitive load being directly connected to the output of a comparator. I do always have this current limit resistor inserted, because I don't want my signal ground being eroded from the brief short circuit currents. This is the more important the faster the used comparator is.

    On the other hand, I know very well, that some people do connect capacitive loads to the output. Even the datasheet of LM393-N has circuits showing them. In figure 20 you can see a capacitive load of even 10nF. In very certain applications this might be acceptable. But I would not recommend this as a usual practise.

    Kai
  • All valid points, thanks Kai
    Chuck