• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 796 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.

PGA2311 power rail sequencing requirement?

Benjamin Brooks
Prodigy 40 points
Other Parts Discussed in Thread: PGA2311, MAX660

Does the PGA2311 require that the VA+ power pin comes up before VD+, or just that VD+ to VA+ is separated by less that +/-0.3V? Can SCR latch up still occur if VA+ pin is shorted to VD+?

I'm seeing latent failures where the PGA2311 VA- rail over time collapses from -5V to around -3.5V (the PGA2311 is the only load on the -5V regulator).

  • 0
    TI__Mastermind 27715 points

    Hello Benjamin,

    I don't have information on the power rail sequencing requirements of the PGA2311, so I'll do some bench testing to find out the answer. 

    Are you seeing the VA- rail collapse during a specific condition, or does it seem to be random? Is the collapse occuring on multiple units?

    Best regards,

    Ian Williams
    Linear Applications Engineer
    Precision Analog - Op Amps 

  • 0 in reply to Ian Williams
    Prodigy 40 points

    Hi Ian,

    Thank you for your quick response & help troubleshooting. Yes, there are multiple PCBs where I have the PGA2311's VA- pin clamp around -3.5V when the -5V VR turns on. This failure mode only happens over time (after multiple power cycles of the part).

    As the PCB is currently designed, the PGA2311's VD+ and VA+ pins ramp up to 5V first (simultaneously), and the VA- pin ramps to -5V (on good parts, to only -3.5V on bad parts) slightly after.

    The -5V rail is generated with a switched capacitor VR from the same +5V rail that powers VD+ & VA+. The 5V rail is filtered with an LC (ferrite & cap). Otherwise the bypassing is 10uF + 0.1uF per power pin as recommended by TI's datasheet. 

    Regards,

    Ben

  • 0 in reply to Benjamin Brooks
    TI__Mastermind 27715 points

    Hi Ben,

    I have been unable to duplicate your issue in the lab. Even with multiple power cycles, and allowing anywhere from hundreds of milliseconds to several seconds to pass between VA+/VD+ and VA- coming up, I don't see any collapse of the VA- rail. 

    How long after your VA+/VD+ rail starting up does the VA- supply come online? In the case where VA- collapses, how much current is being drawn from that regulator? Which regulator are you using?

    Best regards,

    Ian Williams

  • 0 in reply to Ian Williams
    Prodigy 40 points

    Hi Ian

    Thanks for the follow up. Sorry for the long delay in response.

    The VA- rail is being powered from a TI MAX660 Switched Capacitor Voltage Converter.

    Looks like there was a power supply marginality on some PCBs where the caps being used were causing unexpectedly high source resistance and voltage ripple on the rail. 

    That being said, is there long term damage to the PGA2311 possible if the VA- rail is running (out of spec) between -3.5V to -4V with ripple as high as 500mV, even though the abs max rating on DS is -5.5V?

  • 0 in reply to Benjamin Brooks
    TI__Mastermind 27715 points

    Hi Benjamin,

    Thanks for the update. I'm glad you were able to discover the root cause of the issue.

    There shouldn't be long-term damage to the PGA2311 if the VA- rail moves around between -3.5 V and -4 V occasionally, but your maximum output swing will be reduced. However, if the negative rail gets much higher than -3.5 V, then you run the risk of the device going into MUTE (happens when either VA+ or VA- falls below +3.2 V or -3.2 V).

    What is the frequency of the ripple on VA-? The PGA2311 has good PSRR, but with a high enough ripple amplitude in the audio band some of that could be seen at the output.

    Best regards,

    Ian Williams