• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 63 subscribers
  • Views 700 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.

Schematic Review for PMIC: TPS65217D

Amanda Ross95
Intellectual 2740 points
Other Parts Discussed in Thread: TPS65217

Hello,

Would it be possible for your team to review a schematic (just the connections to the TPS65217D) for Server Technology? 

I can email you the specifics since I'd rather not post anything from their schematic on the forum.

Please let me know.

Thanks,

-Amanda

  • 0
    TI__Genius 17380 points
    Hi Amanda,

    First, be sure that someone has gone through the schematic checklist: www.ti.com/.../slva686

    If there are any questions generated during the review, feel free to post them here. I have just sent a connection request so that you can send the schematic over after identifying if there are any questions based on their initial schematic checklist review. At that point we will have feedback to you within a week, if not sooner.
  • 0 in reply to Kevin LaRosa
    TI__Intellectual 2740 points
    Hello Kevin,

    They have added 22uF of capacitance on the SYS_PWR pin and have gotten rid of any boot loop behavior they were seeing. That is the only change in the schematic that I provided. Here are the issues that are still outlying:

    1.) PMIC can boot Sitara from either Vbat or Vac with no issues. With both supplys present, either supply can also be removed and we see no issues. But if we boot from Vbat first then apply Vac, the PMIC does not switch over to Vac rather the PMIC keeps running off Vbat. Now if we removed Vbat (with Vac still present) the PMIC reboots.
    -I understand that Vbat needs to be ~200mV less than Vac, or the PMIC’s power-path will prefer the Vbat connection and ignore the Vac input. But even if that is the case, why does the PMIC still not see Vac once Vbat is removed after originally booting from Vbat.

    2.) Other issues seen are occasional rebooting (presumably PMIC susceptibility) when plugging/unplugging USB and/or Ethernet cables. This is only seen on one board type (we have two) that has a combo Ethernet/USB Magjack. There may be some ESD/transient susceptibility issue here. The 22uF cap added on the SYS_PWR rail has helped (but not eliminated) this issue. Between the two board types, the components are nearly identical excepting for the use of right-angle connectors rather than vertical connectors for the outside port connections due to having two different form factors.

    #2 seems hard to pinpoint but I think there is probably a solution to #1 that we haven't thought of. Please provide feedback as soon as possible.
  • 0 in reply to Amanda Ross95
    TI__Genius 17380 points
    Hi Amanda,

    Sorry for the delay.

    For the case (1), because VBAT = VAC, it looks like in this case VAC is never detected (rising edge + 200 mV > VBAT). As such, when VBAT drops, it doesn't switch to VAC because it was never detected. You could check by checking ACPWR bit in STATUS register. Unfortunately, the TPS65217 is not well-suited to battery-less operation and this is not one of the suggested methods to support it. The best way I am seeing is to avoid this by ORing the two prior to the PMIC and just using one input (as described in datasheet section on battery-less operation). This can be done with a couple of Schottky diodes.

    For (2), does the Ethernet connect to the PMIC in some way? Are the PMIC rails going low or is it just the processor resetting? Is this only with AC unplugged? If implementing above solution, you may want to use the AC input if no other over-voltage is being used on external rails for extended input range.
  • 0 in reply to Kevin LaRosa
    TI__Intellectual 2740 points
    Hey Kevin,

    The Vbat is a full volt below the Vac. Vbat = 4V and Vac comes up at 5V. What else could be the issue?

    -Amanda