Shop OBEX P1 Docs P2 Docs Learn Events
XBee load during SLEEP more than expected — Parallax Forums

XBee load during SLEEP more than expected

FalconFalcon Posts: 191
edited 2012-04-09 08:07 in Accessories
I'm using the Series 1 XBee and a BS2 on a battery to send a few BYTES when a condition changes. Most of the time the XBEE is in Hibernate Sleep mode 1 which, according to the documentation, should draw <10 uA. My XBEE is drawing 3.4 mA during the Hibernate Sleep periods.

I am using the 32401 5V/3.3V Adapter Board. Would the circuitry in this Adapter Board increase the load that much?

Also, is there a way to disable the Blue Power LED on this Adapter Board. I don't need it and I want to minimize the power draw as much as possible

falcon

Comments

  • Mike GreenMike Green Posts: 23,101
    edited 2012-04-08 14:47
    Both the LED and the 3.3V regulator can draw enough current to account for what you're seeing. You can remove the LED from the adapter board or you can cut one of the traces going to it. The regulator is another issue. You'd have to remove the one on the board and substitute one with a lower quiescent current. You might need to change the filter capacitors for the regulator as well. This regulator is one with a particularly low quiescent current. It's quite limited in how much voltage can be applied to its input though. It's one example of what can be used.

    The datasheet for the regulator on the adapter board looks like it has a low quiescent current as well. You may find that disconnecting the LED will drop the minimum supply current enough for your needs ( <30uA)
  • Tracy AllenTracy Allen Posts: 6,658
    edited 2012-04-08 20:46
    Once you have the power led out of the circuit, another thing to watch out for is the RTSn input. Best to tie it high so that the gate doesn't float.

    Another thing to watch out for is the pullup resistors on the xbee, depending on what you have attached to the pins. Pullup resistors can be disabled or enabled using the ATPR command, and they are all enabled by default. For example, there is a pullup resistor on pin 16 of the XBee (RTS\), and if you were to connect the external input RTSn to ground, then level shifter would always be left sinking about 25µA current through that pullup resistor. Unused i/o pins should be left with their pullups enabled, just as with the BS2.

    XBee pin 9, sleep_rq, also has a pullup, so if you are interfacing to sleep_rq from the Stamp, you can do so through a transistor for level shifting, not via a driven connection from stamp pin to XBee pin.
  • FalconFalcon Posts: 191
    edited 2012-04-09 04:49
    Thank you Mike and Tracy,
    I appreciate the suggestions.

    Mike, Transmitter reaction time is not a critical factor so I am considering using a transistor and BS2 I/O pin to control power to the XBee/Adapter Board. This would avoid disabling the LED. Would this approach (one wake/transmit/sleep cycle every 15-30 minutes) cause excessive wear on the XBee? What other parameter changes would be helpful with this approach?


    Tracy, I'm only using the DIN, DOUT and sleep_rq pins...no I/O. I will check the pull-up resistors and see if I can work that transistor into the sleep_rq circuit.

    falcon
  • Mike GreenMike Green Posts: 23,101
    edited 2012-04-09 08:07
    There's not enough information in the documentation to judge whether cycling power would be similar to hibernation. The xBee has to go through some initialization and I don't know how long that takes. Hibernation would skip some of that initialization, but you'd be waking or powering up the xBee infrequently and it's probably close enough.

    Power cycling wouldn't cause wear. There are no moving parts and no power surges (like in a high power device).
Sign In or Register to comment.