• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 63 subscribers
  • Views 165 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.

TPS782: Pin 4 issue

Stephane Leahy
Prodigy 10 points
Part Number: TPS782

Tool/software:

Hi I'm using TPS78218DDCT to provide 1.8V from a 3.3V supply and I'm seeing some strange behavior with microphones connected to the 1.8V supply. What does Pin 4 do? If I connect it to GND like the datasheets says or leave it floating, the microphones don't work properly, but if I connect it to 1.8V, then the microphones work well. Is it possible there is a mistake in the datasheet?

  • 0
    TI__Genius 10790 points

    Hi Stephane,

            I'm looking into the pin functions internally to see if there's an explanation for this. In parallel with that, I have a few questions:

    1. What loads are present on the 1.8V output, and what is their typical current demand?

    2. What behavior do you see from the microphones?

    3. Do you see this behavior across multiple TPS78218DDCT devices, or just a single affected unit?

    4. Can you probe the 3.3V input and 1.8V output with an oscilloscope and provide screenshots here?

    Best Regards,

    Alex Davis

  • 0 in reply to Alex Davis
    Prodigy 10 points
    Hi Alex,
    See the Q&A below. We are very confused by this issue and look forward to hearing about the pin 4 "GND" functionality.

    1. What loads are present on the 1.8V output, and what is their typical current demand?

    Total current draw of approx. 12mA

    2. What behavior do you see from the microphones?

    Our audio codec has large transients/pulses, roughly 1 per second.

    3. Do you see this behavior across multiple TPS78218DDCT devices, or just a single affected unit?

    2 out 15 PCB's were affected

    4. Can you probe the 3.3V input and 1.8V output with an oscilloscope and provide screenshots here?

    Both 3.3 and 1.8V have rippled below 20mV

    Cheers,

    Stephane

  • 0 in reply to Stephane Leahy
    TI__Genius 10790 points

    Hi Stephane,

    I was able to confirm that pin 4 does have an internal connection. The TPS78218 is characterized with pin 4 grounded, however, and should operate properly in that configuration, as shown in the datasheet.

    Does your application use electret microphones with a pullup resistor to 1.8V? If so, even small transients well below 100mV on the 1.8V rail could be significant when gained up by your codec. 

    How are you making the connection to pin 4 in your design? When grounded, is that connection a direct trace from pin 4 to pin 2, or are there longer trace segments or passives in between pin 4 and GND which could allow EMI to couple into that pin? Would it be possible for you to share the relevant portion of your layout and schematic? 

    Best Regards,

    Alex Davis

  • 0 in reply to Alex Davis
    Prodigy 10 points

    Hi Alex, thanks for the update.

    Could you explain what the functionality of pin 4 is? It could help us debug the problem.

    We aren't using electret mics, only PDM MEMS mics.

    We used a bare wire, as short as possible between pin 4 and 5.

    Below is the layout showing the regulator and PDM to PCM converter (TSDP1808X1NEGXZAX).

    Cheers,

    Stephane

  • 0 in reply to Stephane Leahy
    TI__Genius 10790 points

    Hi Stephane,

    Apologies for the delayed response. I unfortunately can't give an exact description of the circuitry that pin 4 is connected to, only that it relates to production testing of the device. I would recommend reworking the affected boards to tie pin 4 to ground and re-testing (if you haven't already). While I wouldn't necessarily anticipate the TPS782 to be permanently affected by floating pin 4, that use case is not one we have tested internally so it is possible that damage could occur. If the issue persists with new devices and with pin 4 tied to GND per the datasheet, I'd be happy to help debug further.

    Best Regards,

    Alex Davis 

  • 0 in reply to Alex Davis
    Prodigy 10 points

    Hi Alex, ok no problem. I think we figured out the problem. It turns out it was the CLK line that was overshooting, and for some reason applying 1.8V to the pin 4 was solving the problem. We added a capacitor on the CLK line and now it's working well with pin 4 grounded. Thanks again for your help!

  • 0 in reply to Stephane Leahy
    TI__Genius 10790 points

    HI Stephane,

    Glad that solved it!

    Best Regards,

    Alex Davis