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.

LM2904: LM2904 crossover distorion happans

Part Number: LM2904
Other Parts Discussed in Thread: TINA-TI, LM2902, TS321, RC4580-Q1

Hello team,

My customer is using LM2904 and add the push-pull output to increase curent drive as the below schematics. When the custoemr check this circuit by the simuration software, they did not find issue. (I also checked using TINA-TI which I attached below. ) However they found crossover distortion when they check their own board.

I think the crossover distortion happens potentially when we use LM2904.
Woudl you please advise to improve crossover distortion?  There is the Input signal information in LM2904.tsc of TINA-TI file.

Best Regards,
Akihisa Tamazaki

<TINA-TI file>
/cfs-file/__key/communityserver-discussions-components-files/14/4520.LM2904.tsc

  • Akihisa san,

    There is a time delay every time the amp op output current crosses about -50uA in either the positive or negative direction. This is due to the unique topology of the LM2904, LM2902, TS321 output driver. It is easy to see the 3 VBE crossover in the data sheet schematic. Due to the finite slew rate it takes a certain amount of time to slew 3 VBE. This behavior appears to be missing from the model. I will take note of that.

    Try the RC4458I as lowest cost substitute for LM2904; however Ta max is 85C not 125C.
    Do you need 125C?
  • Ron-san,

    Thank you for comment.

    My customer already change the circuit to fix it. (But they did not disclose the new one).
    The customer wanted to know why crossover distortion happens, so they asked us.
    Thank you for your support.
    I will post if the customer ask again about it.

    Best Regards,
    Akihisa Tamazaki

  • Akihisa-san,

    The schematic shows the reason. The output NPN darlington and PNP emitter follower output stage is driven by a single node marked green.

    This green node is is slew rate limited by marked capacitor. It should be obvious why there is a delay between sourcing and sinking current.

  • Ron-san,

    Thank you for your comment. 
    I see.

    The cutsomer application is Automotive, and need Q100 device. 
    How about RC4580-Q1?

    Your advise would be so appreciated.

    Best Regards,
    Akihisa Tamazaki

  • Akihisa-san,

    RC4580-Q1 will work.

  • Ron-san,

    Thank you so much fo your support.

    The customer would change RC4580-Q1.

    Best Regards,
    Akihisa Tamazaki