Problem with pull up resistor and op amp

Discussion in 'The Projects Forum' started by nokturn, Jul 31, 2012.

  1. nokturn

    Thread Starter New Member

    Jul 31, 2012
    5
    0
    Hi guys,

    first off, I am new to this forum and also new to electronics. I have been developing software for many years and am now trying to make a little design.

    The design are two PWM generators which each have a jack input and a jack output. For the jack input I add a pull up resistor so that when nothing is connected, the input is still high and thus the generator is driven. So far, so good and it works well. The outputs work well, too, I measure 0v and about 3.8v (low and high).

    The question boils down to a problem I'm having when I connect the output of one generator to the "gate/enable" input of a second. In that case, the output (an op amp as comparator) varies between the previous 3.8v and now 0.6v... which I believe is sinking current from the pull up resistor of the input jack... and that's enough to enable the other generator although that is contrary to my intentions.

    I tried adding a diode after the op amp but that didn't work, it was as if nothing was connected through the jack at all and the pull up resistor set the generator enabled.

    I am aware that I am probably missing some very basic concepts, but except for that my little project worked out well so far. I googled for hours but found no examples or anyone having similar issues. Hope someone can help :)
     
  2. wayneh

    Expert

    Sep 9, 2010
    12,090
    3,027
    A schematic of what you already have would help a lot. Unless the pullup resistor is too low a resistance, the op-amp should be able to pull it down IF it's a rail-to-rail output.

    But anyway a schematic is worth a thousand words.
     
  3. nokturn

    Thread Starter New Member

    Jul 31, 2012
    5
    0
    Thanks a lot for the quick reply... I also imagined it should be able to pull it down to 0v but I think the op amp sinks and sources depending on low/high of the duty cycle... it compares the trigger/threshold of a 555 (556 package) against a set voltage to provide duty cycle control.

    Attached is the schematic, the gate inputs and outputs are labelled.
     
  4. nokturn

    Thread Starter New Member

    Jul 31, 2012
    5
    0
    To clarify: when I connect Out A to Gate B, the voltage at the op amp output of A cycles between 3.8v and 0.6v which causes the second 555 to be active even when it should not be.
     
  5. mcasale

    Member

    Jul 18, 2011
    210
    12
    Not sure why you are using op amps as comparators.

    The outputs will tend to oscillate -possible rail-to-rail when the inputs are close to the switching point. You may want some positive feedback to give some hysteresis. This is very common practice with comparators.

    Also, the input impedances on the op amps are not matched, so you need to accommodate the input bias currents on each input.
     
  6. wayneh

    Expert

    Sep 9, 2010
    12,090
    3,027
    The LM358 datasheet has the likely answer to your mystery. With a 5V supply, it can't really get to 0V with even the small load of the pull-up and the reset pin. Upping the 10k might help get close enough.
    [​IMG]
     
  7. nokturn

    Thread Starter New Member

    Jul 31, 2012
    5
    0
    @mcasale: Thanks, Ill have to read up on op amps and good designs, this one was basically just adapted from a schematic I saw on the internet... it seemed to work reasonably well on the breadboard.

    @wayneh: Thanks, Im going to try this right now and change it for a 22k... if this works, then it would explain why one gate input works perfect while the other doesn't (tolerance?). I double and tripple-checked wiring and even the volt meter shows almost the same numbers so it's not a connection issue...
     
  8. nokturn

    Thread Starter New Member

    Jul 31, 2012
    5
    0
    Hey, I exchanged both 10k pull-ups for 22k pull-ups and it worked! Thanks so much, I spent so many hours trying to find the mistake!

    So basically is there a clean way of handling this sort of situation, a pull up resistor and optional (pluggable) logic input? So far the examples I've seen only concerned themselves with a pull-up/down resistor connected in conjunction with a switch.

    Thanks again!
     
Loading...