• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 2 answers
  • Subscribers 29 subscribers
  • Views 702 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.

PGA2500 overheating fast power and ground question

ben biles
Intellectual 840 points
Other Parts Discussed in Thread: PGA2500

Hi, I have 4 x PGA2500's on a pcb. I am using a shared +-5v supply between all 4 PGA2500's.

I have analogue ground and digital ground & 0v from +-5v supply all joined at one common point off the PCB

The digital supply is coming from the the -5v of the same power supply.

I have DCEN enabled by grounding to digital GND.

I am using 1uf caps between the DCEN capacitor pins.

The PGA2500's are warming up quickly so I assume something is very wrong in my design.

I have tried taking off 2 PGA2500's and now only have 2 soldering on. On my previous veroboard test design I disabled DCEN an eliminated the

chips overheating problem but I would like to enable the function in my design.


Can anyone advise me what i'm doing wrong here ? I have implemented diode protection with recommended diodes in data sheet but I have not included the

48v phantom power yet until I work out this problem !


Any ideas would be much welcome. perhaps I should not have connected 0v to GND ? or there is something wrong in my diode protection schematic that is causing dc to get to somewhere is should not ?

ben


Schematic Prints.pdf

  • 0
    TI__Mastermind 27715 points

    Hello Ben,

    The first thing I would do is to remove all but one PGA2500 for fewer variables during debug.

    The next thing I would do is probe all the input and output pins of the connected PGA2500 to determine the common-mode voltages or DC voltages present. The schematic shows them all to be AC-coupled, but it's possible that issues with the protection diodes or other input/output circuitry is generating an offset. If no input source is connected, it's possible that the input circuitry is floating to some DC voltage as well.

    Since you have the VCOMin pin grounded, that forces the input and output common-mode range to 0V. This also biases the input pins to approximately -0.65V. Therefore any external voltages that develop on the input or output can cause high currents to flow into the device, causing rapid temperature increase.

    The /DCEN function also works to minimize any offset at the output. Again, if you have a high-current condition from improper input or output biasing, this can add to the issues in the circuit.

    In summary, I would recommend the following debug procedure:

    • Remove all but one PGA2500 from the circuit
    • Probe all input and output pins of the PGA2500 while powered on
    • If any DC voltage is present, ensure that a source is connected to the input connector or that the inputs are grounded and re-test
    • If a DC voltage is still present, remove the protection diode network at the input of that device and re-test

    If you don't find any DC voltages present, then the issue likely lies elsewhere.

    By the way, what is the purpose of the 1.2k resistor across the input pins? I'd also remove the 6.8k pull-up resistors to the phantom power source if it's not being used during testing.

    Best regards,

    Ian Williams
    Linear Applications Engineer
    Precision Analog - Op Amps

  • 0 in reply to Ian Williams
    Intellectual 840 points

    HI Ian,

    I tried an empty board first.

    There is -5v at R1 ( the digital -5v supply )

    I tried populating 1 x PGA2500. there is now 2.5v at R1 with or without R1 populated !! I triedd adding DCEN capacitors 1uf, no change.

    I now have -2.5v on both the audio inputs and the PGA2500 still gets hot. -2.5v accross differential hot to GND or cold to GND.

    I'm starting to think faulty PGA2500 or power supply ?

    should I try disabling DCEN by bringing it high ( +5v ) if so should I use a pullup resistor?

    could that be causing some kind of problem with the -5v digital supply ?

    any ideas would be much welcomed.

    thanks,


    ben

  • 0 in reply to ben biles
    TI__Mastermind 27715 points

    Hi Ben,

    Yes, try disabling /DCEN for now by pulling it high with a pullup resistor of around 10k or 47k.

    I also recommend shorting AGND and DGND together directly on the PCB close to where the PGA2500 is soldered down. You may have an issue where these are not actually being connected properly.

    Finally, are all the capacitors on your board ceramic? If not, and you have some electrolytic capacitors in place, especially at the input network or power supplies, make sure they're installed in the right direction.

    Best regards,

    Ian Williams

  • 0 in reply to Ian Williams
    Intellectual 840 points

    Hi,

    I soldered on a new IC on a new board and the issue seams to have been resolved

    The PGA2500 is now using 29ma on +/- power supply.

    I now have biased -0.65v on both Vin+ & Vin- inputs as expected.

    I don't seam to have -0.65v bias on Vout+ & Vout- ( only approx -120mv on both )

    But this is a massive improvement on where I was !

    I think I'm going to have to review my solder techniques ( flux ) for 0.5mm pin pitch as this could have caused some issues before.

    Anyway, Thanks Ian for the advise Ian !  Going back to 1 PGA2500 and starting from that point point really helped.

    BEN