• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 47 subscribers
  • Views 264 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.

TLV3604: Abnormal waveforms occur.

Ryu Yamashita
Genius 3885 points
Part Number: TLV3604
Other Parts Discussed in Thread: PSPICE-FOR-TI, TINA-TI, TLV3605,

Tool/software:

Hi team,

When the following circuit was built, it was found that an abnormal waveform was generated on the falling edge.
The same occurs in the simulation.
What is the cause of this occurrence?

Best Regards,
Ryu.

  • 0
    TI__Genius 11865 points

    Hi Ryu-san,

    I apologize, but I can't help troubleshoot issues using LTSpice. Would it be possible for you to rebuild the circuit with PSpice-for-TI or TINA-TI to see if you face the same issue?

    Would it be possible for you to substitute the TLV3604 with the TLV3605? TLV3605 has ~60mv of internal hysteresis that can be controlled through the LEbar/HYST pin. That way, there is no feedback path that loads the LVDS outputs Q and Qbar:

  • 0 in reply to HO SIU
    Genius 3885 points

    Hi HO-san,

    This occurred in the simulation, but also in the actual circuit.
    Does this mean that oscillation due to feedback is a possible cause?

    Best Regards,
    Ryu.

  • +1 in reply to Ryu Yamashita
    TI__Genius 11865 points

    Hi Ryu-san,

    Could you please show some oscilloscope images showing the chattering occurring in the physical circuit? Please probe the same nodes on the physical circuit as your simulation.

    One thing you can try is to increase the feedback resistor value (R2 on your schematic) in order to lower the loading effect on LVDS outputs. Could you please adjust the value of R2 such that <35uA flows through the feedback resistor?

  • 0 in reply to HO SIU
    Genius 3885 points

    Hi HO-san,

    The oscilloscope results are attached.
    This phenomenon does not always occur and may or may not happen.

    Best Regards,
    Ryu.

  • +1 in reply to Ryu Yamashita
    TI__Genius 11865 points

    Hi Ryu-san,

    The quality makes it a bit hard to see the text on the oscilloscope image. In the oscilloscope result, what are the inputs to the comparator? Are they probed? If so, do you see any pulses where the inputs cross more than once?

    I'm curious to see what the output looks like with a periodic input. The oscilloscope result you shared seems to only have one pulse. Could you please test the circuit with a period signal (like a sinusoid) and see if this abnormal wave occurs at each output pulse? Could you also vary the frequency to see if the abnormal wave shows up there as well?

    Ryu Yamashita said:
    Does this mean that oscillation due to feedback is a possible cause?

    The implementation of hysteresis is positive feedback, so it should by design suppress oscillation (chattering) at the output. Could you try removing the hysteresis by depopulating the R2 (feedback resistor) and seeing if the abnormal wave still occurs?

  • 0 in reply to HO SIU
    Genius 3885 points

    Hi HO-san,

    I understand very well.
    Thank you very much.
    By the way, connecting a capacitor (100pF) in parallel to the output improved the symptoms.
    R2 has not been changed.
    Is there anything I can think of?

    Best Regards,
    Ryu.

  • +1 in reply to Ryu Yamashita
    TI__Genius 11865 points

    Hi Ryu-san,

    Apologies for missing this. In the case of connecting a 100pF to the output, that would suppress the abnormal output by low pass filtering the output and slowing down the overall output rise/fall from the LVDS output. If you're using an LVDS comparator for speed, that solution will work against that. Have you tried the tests in my previous post to see if this abnormal output arises with those conditions?