• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Subscribers 48 subscribers
  • Views 1377 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Question about the behavior of LMV324

Toshiya Hiramatsu
Intellectual 630 points
Other Parts Discussed in Thread: LMV324

Hi,

The customer is using the following circuit.

They said that we observed the spike wave shown in red circle by operation with this circuit.
Is this behavior correct? 　Are any causes thought?
Could you please teach me?
  • 0
    Expert 1330 points

    I think it should work OK. Looks like some unexpected coupling between the signal generator (PWM) and the op amp, or some bad ground loop involving the "PWM", opamp, and/or scope.

    Regards   ---   Roger

  • 0
    TI__Guru**** 166636 points

    Ideally this circuit would be a two pole low pass filter, but the real world is more complex.

    Lets look how the opamp will respond to an “instantaneous” positive input step.

    All capacitors become shorts. So a +5V step at point ‘A’ would causes a 500uA current change on the opamp output pin (5V/10k) or will cause a 5V step on point ‘B’ if the opamp had no response (open circuit). As expected the empirical result is somewhere in between.

     

    The LMV324 output stage has a PMOS drain and NPN collector on the output (see datasheet).

    That makes them current sources which are inherently high impedance and not able to respond to an external current change. Therefore the opamp has to make a correction which is limited by the slew rate which causes the recovery ramp signal.

     

    Improvements can be made by slowing the input signal edge rate with a RC low pass filter and /or increasing the first 10K resistor and adding an output resistor to ground will help attenuate the glitch.

  • 0 in reply to Ron Michallick
    TI__Guru**** 166636 points

    I made a small mistake, only half the input signal makes it to the output pin because both caps are "shorts".

     

  • 0 in reply to Ron Michallick
    Expert 1330 points

    Ron, You are right of course, I should have known better, the 324 is not fast, & looked at the scope Y axis scales a bit closer! Its been a long week..........  Roger

  • 0 in reply to Roger
    Intellectual 630 points

    Roger - san,

    Thank you for your reply.
    I must apologize for the delay in response.

    I'll propose to the customer that I confirm the following two item.
    1. The unexpected coupling.
    2. The some bad ground loop.

    Best Regards,
    Toshiya

  • 0 in reply to Ron Michallick
    Intellectual 630 points

    Ron - san,

    Thank you for your reply.
    I must apologize for the delay in response.

    I'll propose to the customer that I confirm the following two item.
    1. More slowly input signal edge rate.
    2. Increasing the first 10K resistor.

    And,I have two question.
    1.Why is "Adding an output resistor to ground" effective?
    2.Dumping factor of the circuit seems to be small.
      Is not this thing related to this phenomenon?

    Best Regards,
    Toshiya

  • 0 in reply to Toshiya Hiramatsu
    TI__Guru**** 166636 points

     

    A1) An output resistor creates a voltage divider with the input resistor attenuating the glitch when R(load) << R(in).

    A2) If you mean "Damping" factor, then that term can apply to this case.

     

  • 0 in reply to Ron Michallick
    Intellectual 630 points

    Ron - san,

    1.I was able to understand.
    2.Sorry! "Dampling" is correct.

    Well, following new information came from the customer.
    ------------------------------------------
    We changed first 10k resistor to 100k and experimented.
    Result : Glitch has become small very much.
    And, we connected the 200ohm resistor with the op-amp output and experimented.
    (The first 10k resistor is unchanged.)
    Result : Glitch has become small very much.
    ------------------------------------------

    The customer confirmed result same as your comment.

    I think glitch's cause is the following.
    Maybe the input slew rate is too fast and current change is too big.
    So, the op-amp is not able to respond to an external current change.
    Is this correct?

    Best Regards,
    Toshiya

  • 0 in reply to Toshiya Hiramatsu
    TI__Guru**** 166636 points

     

    Toshiya, your understanding is correct.

     

  • 0 in reply to Ron Michallick
    Intellectual 630 points

    Ron - san,

    Thank you.

    Best Regards,
    Toshiya