• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 62 subscribers
  • Views 981 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.

UCD9248 PMBus communication - no devices found during address scan

pui ling lee
Prodigy 30 points
Other Parts Discussed in Thread: UCD9248

Hi,

I have my test board working fine with the UCD chip previously.

Suddenly something wrong on the board and now I'm not able to scan any UCD chip on my board.

How do I start debugging this?

I checked the V33A/V33D/V33DIO, they are 3.7V. Is this Ok?

BPCap is 1.77V.

I have the PMBus signals pulled up to 3.3V with 2kohm resistor.

The voltage level measured on the 2kohm ends are 3.3V and 0.7V.

The PMBus signals are connected to all of the UCD chip.

So what could cause the pullup drops to 0.7V?

  • 0
    TI__Genius 9845 points
    Do you mean UCD9248? The max V33D in operation is 3.6V.

    First step would be checking PMBus signal integrity. When Fusion GUI is scanning PMBus, can you see the data and clock lines switching? If yes, is logic High above 2.1V, is logic LOW below 1.4V? If PMBus has no switching, something is pulling the bus low. You may put PMBus devices in reset and enable them one by one in order to find out which one is pulling the PMBus.
  • 0 in reply to Zhiyuan Hu
    Prodigy 30 points
    Hi,
    Thanks for your quick response.
    Yes, I mean UCD9248. I know the spec max V33D is 3.6V. But why I get 3.7V now?

    I see logic low and high at 0V and 0.85V.
    I only see a low pulse when I click on "retry" to scan the devices.
    That is when I measure the logic low and high.

    We have all of the devices reset connected together. I'll find a way to try reset them one by one.
    If I don't supply Vin to the device, will it works as reset too?
    It will be easier for me to remove the Vin one by one instead of reset.
  • 0 in reply to pui ling lee
    TI__Genius 9845 points
    You need to check the 3.3V power supply. It is not controlled by UCD9248.

    Something is pulling low the bus. If possible, you can reset all UCD devices and see if it is something else loading the bus.

    If you can remove 3.3V for each device independently, it is equivalent to reset the device. However, if a device is not powered (say V33D=0V), the voltage on any pin should be less than 0.3V, otherwise the ESD diodes on these pins will conduct, and the voltage on these pins will be clamped at 0.3V. From this point, you cannot do it, as the PMBus will be clamped at ~0.3V if you don't power a device.

    You can check if any device on the bus is unpowered, which may be the cause. Worst case is you'll need to remove devices one by one until the PMBus works.

    Also, how do you power the 3.3V of PMBus? The USB-to-GPIO dongle has limited current capacity. You should use an on-board 3.3V power supply.
  • 0 in reply to Zhiyuan Hu
    Prodigy 30 points

    The power of PMBus is supplied from an on-board regulator.

    What might be pulling the bus low?

    I have tried reset all UCD devices and the bus is still low.

    I removed the input voltage that generates voltage to 3.3V for each device one by one. The measured voltage at V33 pin is 0.255V, which is less than 0.3V than you mentioned. However, one of the UCD device V33 pin gives 0V, which is suspicious. It seems that this device V33 is shorted to ground.

    I have removed all of the decap that are connected to the V33 net, but it still short to ground.

    If one of the device is dead, will it pull the PMBus low?