Shop OBEX P1 Docs P2 Docs Learn Events
P2 Eval RevB consuming 224mA on startup? — Parallax Forums

P2 Eval RevB consuming 224mA on startup?

samuellsamuell Posts: 554
edited 2020-01-02 18:53 in Propeller 2
Hi,

I've just caught my P2 Eval RevB doing something unusual: consuming 224mA on startup, without being programmed. This was just after enumeration, after being unused for a few days. The issue was solved after pressing the reset button. I've disconnected and reconnected the board after that, without being able to repeat the glitch. The P2 seems to work fine, though, so nothing got damaged.

Has anyone had encountered such an issue?

Kind regards, Samuel Lourenço
Screenshot_20200102_184929.png
1440 x 900 - 122K
Screenshot_20200102_184944.png
1440 x 900 - 113K

Comments

  • cgraceycgracey Posts: 14,133
    samuell wrote: »
    Hi,

    I've just caught my P2 Eval RevB doing something unusual: consuming 224mA on startup, without being programmed. This was just after enumeration, after being unused for a few days. The issue was solved after pressing the reset button. I've disconnected and reconnected the board after that, without being able to repeat the glitch. The P2 seems to work fine, though, so nothing got damaged.

    Has anyone had encountered such an issue?

    Kind regards, Samuel Lourenço

    So, after that, you can't get it to show high current again? It should be around 30 milliamps, right? Not sure why that would happen.
  • samuellsamuell Posts: 554
    cgracey wrote: »
    samuell wrote: »
    Hi,

    I've just caught my P2 Eval RevB doing something unusual: consuming 224mA on startup, without being programmed. This was just after enumeration, after being unused for a few days. The issue was solved after pressing the reset button. I've disconnected and reconnected the board after that, without being able to repeat the glitch. The P2 seems to work fine, though, so nothing got damaged.

    Has anyone had encountered such an issue?

    Kind regards, Samuel Lourenço

    So, after that, you can't get it to show high current again? It should be around 30 milliamps, right? Not sure why that would happen.
    Indeed, I can't reproduce the glitch.

    Note that doing "itusb1-detach" and "itusb1-attach" on the terminal is the equivalent of doing a re-connection of the physical cable, since the board is connected via the ITUSB1 test switch.

    Kind regards, Samuel Lourenço
  • JRoarkJRoark Posts: 1,215
    Was there an SD card connected? I had something similar happen right after I got the board. It pulled about 150 ma and wouldnt boot. Eventually I found it was the SD card. Nary a problem since.
  • msrobotsmsrobots Posts: 3,704
    Do you have a SD card installed?

    While developing my RAISD system I had lock up situations with not unmounted SD cards where a reset would not be enough and I had to power cycle or remove and reinsert the SD card.

    Mike
  • Cluso99Cluso99 Posts: 18,069
    There is an aweful lot of capacitance on the P2-EVAL so it is possible the inrush current could be this high although it would only be a power up spike.
    I have no way to measure the inrush spike though.
  • samuellsamuell Posts: 554
    edited 2020-01-02 23:42
    JRoark wrote: »
    Was there an SD card connected? I had something similar happen right after I got the board. It pulled about 150 ma and wouldnt boot. Eventually I found it was the SD card. Nary a problem since.
    msrobots wrote: »
    Do you have a SD card installed?
    While developing my RAISD system I had lock up situations with not unmounted SD cards where a reset would not be enough and I had to power cycle or remove and reinsert the SD card.

    Mike
    No SD card. Plus, all the switches were, and still are, in the "off" position, so the EEPROM was not being used.
    Cluso99 wrote: »
    There is an aweful lot of capacitance on the P2-EVAL so it is possible the inrush current could be this high although it would only be a power up spike.
    I have no way to measure the inrush spike though.
    But that spike would have to have a relatively long duration, in the order of ms. The program used to read the status takes five samples for the current measurement, and averages them. The one spike would need to be huge, or long in duration.

    Kind regards, Samuel Lourenço
  • JRoarkJRoark Posts: 1,215
    That realization has definitely moved this issue deep into “gremlin” territory. If it isnt the SD card, I got nothin...
  • evanhevanh Posts: 15,188
    It looks like the first reading after instrument initialisation. I wouldn't be treating that as an accurate reading without further evidence that it can't be screwy at that point.
  • TubularTubular Posts: 4,622
    You say its not programmed, but it would still run the startup code from Rom briefly, wouldn't it? The code that checks for a flash chip, listens on serial for a short while, etc?
  • SaucySolitonSaucySoliton Posts: 488
    samuell wrote: »
    JRoark wrote: »
    Was there an SD card connected? I had something similar happen right after I got the board. It pulled about 150 ma and wouldnt boot. Eventually I found it was the SD card. Nary a problem since.
    msrobots wrote: »
    Do you have a SD card installed?
    While developing my RAISD system I had lock up situations with not unmounted SD cards where a reset would not be enough and I had to power cycle or remove and reinsert the SD card.

    Mike
    No SD card. Plus, all the switches were, and still are, in the "off" position, so the EEPROM was not being used.
    Cluso99 wrote: »
    There is an aweful lot of capacitance on the P2-EVAL so it is possible the inrush current could be this high although it would only be a power up spike.
    I have no way to measure the inrush spike though.
    But that spike would have to have a relatively long duration, in the order of ms. The program used to read the status takes five samples for the current measurement, and averages them. The one spike would need to be huge, or long in duration.

    Kind regards, Samuel Lourenço

    All switches open means no brownout detection.

    Some OSes might probe serial devices looking for a modem.
  • samuellsamuell Posts: 554
    evanh wrote: »
    It looks like the first reading after instrument initialisation. I wouldn't be treating that as an accurate reading without further evidence that it can't be screwy at that point.
    The program always does one or two bogus reads (depending on the initial condition) and discards them unconditionally, just right before taking samples for the actual read. Thus, it cannot be screwy.
    Tubular wrote: »
    You say its not programmed, but it would still run the startup code from Rom briefly, wouldn't it? The code that checks for a flash chip, listens on serial for a short while, etc?
    Then I got "lucky", I guess.
    SaucySoliton wrote: »
    samuell wrote: »
    JRoark wrote: »
    Was there an SD card connected? I had something similar happen right after I got the board. It pulled about 150 ma and wouldnt boot. Eventually I found it was the SD card. Nary a problem since.
    msrobots wrote: »
    Do you have a SD card installed?
    While developing my RAISD system I had lock up situations with not unmounted SD cards where a reset would not be enough and I had to power cycle or remove and reinsert the SD card.

    Mike
    No SD card. Plus, all the switches were, and still are, in the "off" position, so the EEPROM was not being used.
    Cluso99 wrote: »
    There is an aweful lot of capacitance on the P2-EVAL so it is possible the inrush current could be this high although it would only be a power up spike.
    I have no way to measure the inrush spike though.
    But that spike would have to have a relatively long duration, in the order of ms. The program used to read the status takes five samples for the current measurement, and averages them. The one spike would need to be huge, or long in duration.

    Kind regards, Samuel Lourenço

    All switches open means no brownout detection.

    Some OSes might probe serial devices looking for a modem.
    No brownout detection just means that the P2 won't reset in case of a too low core voltage. Doubt that the glitch was caused by that. As for the OS trying to look for a modem, even if that happens, it shouldn't cause the glitch.

    Kind regards, Samuel Lourenço
Sign In or Register to comment.