• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 2 answers
  • Subscribers 46 subscribers
  • Views 1618 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.

ADS1278: Partially powered ADS1278 and /PWDNx pins

Dan Solvin41
Prodigy 95 points
Part Number: ADS1278
Other Parts Discussed in Thread: THS4524, OPA2350

Hi,

I have a situation with the ADS1278 where it is likely that the /PWDN pins could be active at 3.3V, and only the 1.8V and 3.3V rails are active.   The

3.3V /PWDNx voltage is created by a weak pull-up resistor, so is likely to be 0.1mA maximum.    The 5V rail will become active in short order, but

I'm concerned that the part could be damaged during this delay.

Thanks, Dan

  • 0
    TI__Guru 57306 points
    Hello Dan,

    Thanks for your interest in our ADS1278!

    It is ok for the digital power supplies to be brought up before the analog supply. Internally, the device will issue a global power-on reset once all three supplies are established. The digital configuration pins (i.e. MODE, FORMAT, /PWDN, etc.) are expected to be configured with on-board resistors before power-up in many cases, so I would imagine that the scenario you're describing should not be a concern as long as the current is properly limited.

    Best Regards,
  • 0
    Prodigy 95 points

    Hi Ryan,

    I have a similar question pending that involves the THS4524 buffer when powered down, as the /PWDN pin goes

    both to the ADS1278 and this buffer.   The board has a weak pull-up of 100K to 3.3V, but one of these parts seems to

    pull the signal to mid-range (1.225V or so), before 5V is active.   Based on your response, I'm now guessing the

    buffer is the culprit, but need to make sure I won't damage it if I use a stronger pull-up resistor to 3.3V.

    Dan

  • 0 in reply to Dan Solvin41
    TI__Guru 57306 points
    Hi Dan,

    What supply voltage are you using for the THS4524? Typically, the supplies are shared between the amplifier and the analog portion of the ADC. The /PD pin should be pulled up to the positive supply to enable the amplifier's output. If the amplifier is running off of 5V, then you should not share the /PD pin with the ADC's /PWDN pin.

    Are you pulling the /PD pin of the amplifier to 5V or to 3.3V? Also, what is the VOCM pin connected to? I'm not sure what the output pins will be driven to when the amplifier is powered-down (depends what is connected to the input), but when active, the outputs will be centered around the VOCM voltage (mid-supply by default).

    Best Regards,
  • 0 in reply to Ryan Andrews
    Prodigy 95 points

    Ryan,

    The THS4524 is powered by the same 5V as used on the ADS1278, and the board is attempting to pull the /PD to

    3.3V with a 100K pullup resistor.   The same signal does go to each parts /PD and /PWDN pins.  All the VOCM pins are tied

    to a OPA2350 as recommended, which is also powered by 5V.    It seems the unpowered buffer is pulling /PD down

    to about 1.22V in the time period between when 3.3V is active and 5V is active.    Your previous response indicated that

    the ADS1278 is likely not the cause.  This lower voltage on /PD is causing problems with other parts of the board.

    Dan

  • 0 in reply to Dan Solvin41
    TI__Guru 57306 points

    Hi Dan,

    Thanks for clarifying. I'm not the expert for the high-speed amplifiers. It may be more helpful that you post the question to their forum and ask them to explain what may happen if something tries to pull the /PD pin high before power is established. My guess would be that you could be violating an Absolute Max Rating and that the /PD pin is sinking some current. Is 1.22V measured at the /PD pin (i.e. the low side of the 100k resistor)? (3.3V - 1.22V)/100k = 20.8uA. This may not damage the part, but again, please consult an expert from the amplifier team.

    Best Regards,

  • 0 in reply to Ryan Andrews
    Prodigy 95 points
    Hi Ryan,

    I posted this issue in parallel to that forum. They essentially said the ESD diodes are not characterized at low voltages and
    they couldn't really guess what would happen if I force it closer to 3.3V by using a lower value pull-up resistor when the buffer is unpowered.
    They recommended a 5K series resistor, but that would require a board spin. I'll probably just have to try this and see what happens.

    Dan
  • 0 in reply to Dan Solvin41
    TI__Guru 57306 points
    Understood.

    According to their datasheet, /PD > 2.1V should be enough to enable the amplifier output for a 5-V supply. I suppose we don't know what other implications that may have.

    Best Regards,
  • 0 in reply to Ryan Andrews
    Prodigy 95 points
    Sure, but that doesn't really matter when it's unpowered. My key concern was whether forcing the /PD voltage higher (when unpowered) would
    damage that part. I'm anticipating that the ESD diode will pull even more current to ground as it's voltage is raised. I'll just need to play with the pull-up resistor value to get around 2.0V to 2.4V, which should be high enough to be seen as a logic '1' to other circuitry. I'll also make sure
    that the 5V comes up soon after the 3.3V to minimize the energy dumped into that ESD diode.

    Thx, Dan
  • 0 in reply to Dan Solvin41
    Prodigy 95 points

     I tried adding additional pull-up resistors to 3.3V (already has a 100K), first a 51K, then another 51K in parallel.   The

    voltage never got above about 1.50V.   It seemed clear that the harder I tried to pull it up, the ESD diode in the unpowered

    THS4524 tried to pull it down.    So, I gave up this battle as pulling up harder might have damaged the buffer.    The

    solution I took was to just lift the /PD pins of the buffer off their pads, removing them from the circuit.   Luckily an open /PD

    is still considered 'active' by this part.    This worked.

    Dan