• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 29 subscribers
  • Views 586 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.

TAS5805M: Become Very Hot When Playback At Low Output Power

C%20K%20Ong
Expert 1240 points
Part Number: TAS5805M
Other Parts Discussed in Thread: TAS5805

Dear Sir/Mdm,

We have been using TAS5805M for soundbar and in production for almost a year.

Recently we found that TAS5805M would increase in temperature (>120 degC) until it triggered the over temperature protection and cut off the output. This issue happened intermittently.

Test Procedure:

1. Power up TAS5805M's PVDD and DVDD.

2a. PDN goes from LOW to HIGH with DVDD. I2C data is sent > 5 seconds after PDN is HIGH.

2b. PDN goes from LOW to HIGH (controlled by controller). I2C data is sent within 200ms after PDN is HIGH.

3. Write configuration parameters to TAS5805M, after I2S is stable.

4. Playback back audio on TAS5805M at about 10W~15W (mid volume).

5. Measure the temperature on the package of TAS5805M and observed.

Observations and Troubleshooting:

- We found that over temperature issue is easily reproduced with step# 2a of PDN.

- So far, we are unable to reproduce the over temperature issue with step# 2b of PDN.

- The same board will operate normally more of the time. Failure rate of the same board is about 1~2 out of 5 tests.

- During playback, when there is no issue, the surface temperature is about 36 degC.

- During playback, when there is over temperature issue, the surface temperature will increase from 36degC to 120degC within 2~3 min of continuously playback.

- When there is no playback, the TAS5805M's temperature is normal.

- When playback stopped, the temperature will also reduce back to normal.

- When writing parameters to TAS5805M, we did not see any NO ACK issue on I2C bus.

- After over temperature protection cut off the TAS5805M, power off the PVDD and DVDD, and repower the issue might not happen.

Other information:

- The TAS5805M's datasheet only stated minimum 5ms between PDN high to I2S, and then I2C. 

- In our design the I2S is >100ms after PDN is HIGH for both step# 2a and 2b above.

- We used three TAS5805M on our board with one for mid-range (20Wx2), one for tweeter (10Wx2), and one for subwoofer (40Wx1). PDVDD = 24V.

- Only the mid-range TAS5805M exhibit this issue.

Possible Solution:

We changed the controller firmware to pull PDN to LOW, then HIGH again before writing the parameters to TAS5805M. In this way, the PDN's LOW-to-HIGH till the I2C parameters configuration happened is >200ms and < 1seconds.

So far we are unable to reproduce the over temperature issue with this controller firmware.

Question:

1. We believe we have the correct startup sequence for PDN in both step# 2a and 2b. We want to know what could be the root cause of TAS5805M being over heat.

  • 0
    TI__Guru 52580 points

    Hi

      It's a quite interesting problem. From your description for now, the only possible reason I can think of, is in your step# 2a procedure, the I2C codes doesn't correctly written into the device somehow. 

      I want to double check two things:

      First, when the device temperature abnormally raised and triggered the OT, do we already test the output power during this situation? Can we make sure it's still between 10W to 15W? Not much higher?

      Second, if we can make sure the output power doesn't change too much, still within 10W to 15W, I suggest we test the output PWM frequency. If the frequency is abnormally low, also will cause OT problem.

  • 0 in reply to Shadow He
    Expert 1240 points

    Hi Shadow,

    We are sure there is no issue on I2C data, as we did not see any NACK on the I2C bus.

    However, we did not readback the I2C data, hence we cannot tell whether the data had correctly written into TAS5805M.

    1) We did not physically measured the power output, we "listened" to the same song during the tests at mid-volume setting. From the loudness of the song, we did not hear significant loudness difference between step# 2a and 2b.

    2) The PWM frequency did not change for both good and failed IC, both at 768KHz.

    We were suspecting may be a capacitor's size changed between the old PCBA and new PCBA could have caused the issue, so we swapped the TAS5805M between the two board. The result is the over temperature issue followed the TAS5805M.

    Following were the two IC used in swapping test.

    Best Regards,

    CK

  • 0 in reply to C%20K%20Ong
    TI__Guru 52580 points

    Hi CK

        From the marking, there's no obvious issue, all are normal TAS5805 device. The left side "No Issue" device, do you mean no matter you use 2a or 2b sequence, temperature will always normal, right?

        Do you have any TAS5805 EVM on your hand? Is there any chance you can put the OT device on the EVM and have some try? Can we reproduce this OT failure on the EVM?

  • 0 in reply to Shadow He
    Expert 1240 points

    Dear Shadow,

    Yes, the 'No Issue' device has normal temperature with step# 2a and 2b.

    Not possible to put the 'have issue' device on EVM as we need the EVM for other projects.

    We will be conducting more swapping tests today. I will update you on the outcome whether the issue will follow the TAS5805M.

    Best Regards,

    CK

  • 0 in reply to C%20K%20Ong
    TI__Guru 52580 points

    Hi CK

        Thank you, let me know if you find something else. If we can be sure the problem follow the device, maybe you can send some problem device to us, and we will try to do some test. You can contact with below e-mail:

    jesse-ji@ti.com

    wei-qiu@ti.com

  • 0 in reply to Shadow He
    Expert 1240 points

    Hi Shadow,

    We still unable to identify the rootcause.

    Best Regards,

    CK

  • 0 in reply to C%20K%20Ong
    TI__Guru 52580 points

    Hi CK

        As we discussed previously, if we can make sure the problem follow the devices, please kindly contact the e-mail address I provided, and send the failure device to us. We will try to do some analysis. Thank you.