Announcement

Collapse
No announcement yet.

questions about power

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • questions about power

    I have a battery-powered application where I'm using a NiMH battery pack. I have about 6 hours of battery life if operation is continuous (i.e., display on) and 28 hours of battery life if the display is fully dimmed. I switch into dimmed mode whenever the device is still for 20 seconds or so, and I wake back up if the device is disturbed. When the device is hooked to the charger, it is set up to supply the necessary current to run the dimmed display plus trickle-charge the battery (so it can stay on the charger indefinitely).

    All of the behavior is the previous paragraph works as advertised. I have a couple of issues I'd like to address, though. One, it would be nice if there was a way to go into sleep mode with a uOLED-160 and wake up somehow, but there's no touchscreen and nothing is attached to the serial port in this application. Two, I cannot figure out what to do to power the system back up if the user forgets to place the device in the charger.

    When the user places my dead device back onto the charger, the uOLED-160 apparently tries to boot, and draws more amperage than my charger is willing to supply. It sits in this state, consuming more power than is available, and does not allow the battery to recover to the point where it can boot and run the display. Even if I increase the amperage to the point where the batteries eventually charge, the display module seems to never get to a point where it'll boot, if it's seen (and tried to operate on) that lower voltage. I don't really have room for much additional circuitry to alleviate this problem, nor is it immediately obvious to me how I would handle this issue.

    One way I could at least delay the issue would be if I could go into sleep mode when the device is quiescent. Because the display-off operation of the device still draws 25ma, I eat through the batteries in a little over a day. If I could really let the device sleep, I could increase battery life to where I could maybe safely assume that the device would be placed on its charging base before the batteries went too flat.

    Do I need to hide a reset switch (which I guess would disconnect the power momentarily) on this thing somewhere? Do I need to somehow disconnect the module while the batteries are charging (the voltage never exceeds the spec, so I was intending to leave it connected and operating). I can possibly get my accelerometer to generate an interrupt when it senses movement, but I don't see anywhere to attach that to wake up my uOLED-160.

    Ideas? Am I missing something obvious, here?

    Thanks,

    Scott

  • #2


    I think maybe the HV regulator on the display is trying to 'keep going' as the voltage gets way to low.

    Could you perhaps wire up the display so that power is cut off below a certain voltage and see how that goes?
    Mark

    Comment


    • #3


      Yeah, I've been toying with ways to do that...

      On the other tactic...where is the sleep command in the documentation for this module?

      How much of a serial character does the module need to see in order to wake from sleep? Does this work in GFX mode? If I just twiddle that line a little, will things wake up?

      The device doesn't need to work while it's being charged, so perhaps there's something I can do to keep it asleep when it's on the charger. I can obviously sense that fairly easily, but what's the expedient, simple, and compact way to act on it?

      Is there something I could do with the reset line? Not finding a clear description of that anywhere, either. Can you point me to the correct section of whichever document?

      Cheers,

      Scott

      Comment


      • #4


        The 'sleep' for Goldelox is ProgramExit() the only way to wakeup is to toggle the Reset line.

        This will probably give you longer battery life, but you still need to solve the 'overconsumption' issue if the batter gets too flat.
        Mark

        Comment


        • #5


          Hmmm...I don't really have time to dink with this issue right now, so I've switched to a smart charger which provides sufficient power to get the module running when it's placed on the charger, even from a fully-discharged state.

          I'll revisit the root cause later. Perhaps there's some trickery I can use to get a simpler system to behave the way I want, but it's not obvious right now. I suspect what I'm doing as a work-around will beat up the battery pretty badly over the long haul (because the available smart charger pumps more current into the battery than I'd really like).

          Comment


          • #6


            The smart charger only changed the issue...once the charger shifted over to trickle, the screen ever so slowly drained that battery, and the smart charger wasn't smart enough to detect this and turn back on.

            So, I implemented a hardware power-down solution for when the thing is on the charger. All's fine from a battery point of view, but as the screen powers back up, it has an ugly flash of static (or some such).

            If I eliminate the clearscreen call at the outset of my program, the flashed screen is that last thing it showed. A fraction of a second later, it starts doing what I believe the program is telling it to.

            Any clue what's going on here? How do I eliminate this flash?

            Cheers,

            Scott

            Comment


            • #7


              Unfortunately there isn't much you can do about that.

              It's caused by the commands that initialize the display.

              They are supplied by the manufacturer of the OLED and often cause flashes and such.
              Mark

              Comment

              Working...
              X