1 Reply Latest reply on Sep 3, 2020 12:42 PM by bhfletcher

    PicoZed Voltage Divider R69 - is it needed?

    400garrison

      !  We hit a snag on the VCCIO_EN signal but I think we have it all sorted out but wanted to run it by you guys.

      PicoZed Power Architecture

      VCCIO_EN doesn't provide a 1.8V signal, it's more like 1.4V.  This was surprising to us as the HW user guide implies a 1.8V signal (although is not explicit).  See below excerpt..

       

      The 1.4V wasn't sufficient to turn on the fan2558s10x regulator we selected and so our power rails were not able to turn on. 

       

      I reviewed the 7015 SOM Rev E schematic and found that there is a resistive divider on VCCIO_EN.  Can you explain why this resistive divider is present?  We had to remove R69 from the SOM to get a desired 1.8V logic level.  Do you think this is appropriate for our application, or are we missing anything?

      TLV61230 Power Block

       

      I suspect there was some reason for this voltage divider on VCCIO_EN but I can't find any documentation on it.  Please advise.

        • Re: PicoZed Voltage Divider R69 - is it needed?
          bhfletcher

          Hi Adam,

           

          The PG output is open-drain, which is why we have the pull-up. The weak pull-down was added to more quickly shut down the circuitry in a power-off event. That is why you see the voltage divider. This circuit originated on MicroZed, and PicoZed inherited the design.

           

          The pull-up value has gone through some iterations due to another issue on the board that was resolved. You can read about this in the MicroZed Rev G errata, Issue 3.3.

           

          On MicroZed Revisions F-05 and earlier, an issue was identified where this signal was affected by backfeed

          from the VTT regulator. Resistor R35 was changed from 1.0K to 1.5K on Revision F-06 to accommodate

          this other issue. Revision G maintained this 1.5K resistor for consistency.

           

          The VTT backfeed problem was corrected on Revision G, so R35 could be replaced with 1.0K-ohm without

          issue.

           

          Therefore, I think you have two options:

          1. Reduce the resistance on the R67 pull-up which will raise the output voltage on PG_1V8. This maintains the shutdown behavior as designed.
          2. Remove the pull-down, which also raises the output voltage on PG_1V8. However, the customer is responsible for determining if the shutdown behavior is acceptable, meets eFuse requirements, etc.

           

          Bryan

          1 of 1 people found this helpful