• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 31 subscribers
  • Views 938 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

LM4675: Loss of Audio, AMP remains in shutdown after power restored.

Vaughn Watts
Prodigy 50 points
Part Number: LM4675

Design with 5.65V(Max) to 4.75V(Min)  input to Codex and AMP circuit; divisor circuit from 5V to shutdown pin using R1@47K, R2@22K, R2 to ground and shutdown. .01pf CAP added between R1 and R2

Problem is that originally when removing 5V input, the AMP would not reset on restore of 5V-we had "loss of Audio".  This happened when the power was remove and restored quickly and up to 3-5 seconds afterwards.  The AMP was in a shutdown mode and would not come out even though the Codex was happy.  We put the values of 47K and 22K into a divisor  circuit connected to shutdown and after testing 17 units we had only 1 failure.  The failure resulted in removing the power for more than 10 seconds and then the AMP was locked up after restoring the power.  We have not been able to reproduce the 10 second lockup again but know Murphy is waiting.  We replaced the 22K with a 33K and then the 10 second "loss of Audio" re-appeared multiple times and with ease.  Seems like the 22K is in the correct direction but we are not sure of the final value.  20K or other?  Do you have a suggestion?

In summary, it seems the divisor with 47K and 22K fixes the "loss of Audio" for the quick loss of power and restore of power, but adds the loss of Audio for when the power remains off greater than 10 seconds before restoring it.

We don't have any app notes on this that we can find from TI.  Can someone validate our logic and values along with others reporting this issue.  We found the current solution on a blog from 2010 and decided to try it lacking any other input.  Are we chasing the wrong solution for this problem?

  • 0
    TI__Mastermind 20085 points
    Hi Vaughn,

    Welcome to E2E, thanks for your interest in our devices.
    I don't understand very well your shutdown circuit, so Could you provide me a schematic?

    Best Regards
    José Luis Figueroa
    Audio Applications Engineer
  • 0
    Prodigy 50 points

    Originally we did not have the divisor circuit on the board.  When that happened, the AMP would not recover when power was removed and restored within 3 seconds.  After adding the divisor circuit (via reworking the board), the unit would power down correctly and power up correctly (audio working) when the power was removed and restored within 1, 2, 3 or faster seconds.  However, we got one failure after 10 seconds one time. We replaced the 22K for R2 with a 33K and the 10 second failure stated to occur almost on demand.  We have seen the 22K for R2 solve the AMP lost of audio (comes out of shutdown) 99.99 % of time.  BTW measuring the input 5V at R40 it is steady at 4.95V.

    We think we are close to the correct values but maybe we want to do 20K for example to give some more margin.  However, we just don't know and need your input as to how to solve this problem.

  • 0 in reply to Vaughn Watts
    Prodigy 50 points
    Please let me know if the schematic that I pasted got to you. I can't tell from this post output.
  • 0 in reply to Vaughn Watts
    TI__Mastermind 20085 points
    Hi Vaughn,

    I can't watch it. Could you submit it again?

    Best Regards
    José Luis Figueroa
    Audio Applications Engineer
  • 0 in reply to José Luis
    TI__Expert 8560 points

    Jose,

    JPG of the schematic attached.

  • 0 in reply to John Walker
    TI__Mastermind 20085 points
    Hi guys,

    It is not necessary to add a voltage divider or capacitor in the shutdown network.
    It is always recommended to add a pull-up resistor in the shutdown pin and the maximum shutdown level should be Vdd.
    I think that a pull-up resistor could resolve your issue.

    Best Regards
    José Luis Figueroa
    Audio Applications Engineer
  • 0 in reply to José Luis
    Prodigy 50 points
    The suggestion is exactly what we did before we added the rework to add the divisor and cap. Before the rework, the AMP would remain in shutdown unless the power was removed for more than 3-5 seconds. Any less time between power removal and restore would keep the AMP in shutdown. That's why we went to the divisor and cap. Adding a 33K to R2 created an opposite problem in that the AMP would exist shutdown as long as the power was restored under 10 seconds but would remain in shutdown if power was restored after 10 seconds. We don't understand this opposite effect at all. As we used the 22K now in a test bed, we are getting 100% passing test.