Deepsleep current

[gedue]

I want to use the Wipy for a data logger which is most of the time in Deepsleep-Mode. In the Spec's of the Wipy you find that the Hiberante current is only 5µA, but actualy I measure 450µA which is way higher. Regarding the schematics and the data sheet of the voltage regulator, I see, that the ground current of the used TPS79601 is already 265 - 365µA. The voltage divider at the output of the voltage regulator (R15/16) which is a resitor of 51k+30k=81k, consumes 41µA more. I don't see any possibility to avois this consumption in the current design. Are there any plans to use another voltage regulator for the future?

[danicampora]

Do you have any floating pins?

We do have plans to upgrade the regulator design, but in the meantime you can always use your own 3V3 regulator a connect it's output to both VIN and 3V3.

Cheers,
Daniel

[gedue]

What do you mean with floating pins? I have an external AD-Converter and a SD-Card connected but the other pins are unconnected.
I tried your suggestion with the external voltage regulator already, but it did't change much. The problems remains. The internal voltage regulator stays connected to ground and the R15/16, too. Unfortunately these components are under the cover which makes it almost impossible to simply remove them. When do you think, the modified Wipy will be available?

[danicampora]

OK I understand. We'll investigate...

Next revision will take some months...

[kfricke]

There was a discussion regarding SD-card power requirements... http://forum.micropython.org/viewtopic. ... card#p5501

[gedue]

Thanks for the hint regarding the SD Card consumption. I checked, but the measured values are without SD-Card. The one I am using, just consumes 100µA. And here I have the possibility to add a transistor to supply it only when the Wipy wakes up.
To the internal current consumption I have no influence with the given design.

[Damien]

Any unused IO pins should be configured as inputs with pull ups (or pull downs) to get the least possible leakage of current through them when sleeping. On some MCUs this has more of an effect than others so you'll need to test it out.

[TinheadNed]

I get 480uA sleep current (without the SD card, mine is another 150uA). As this isn't going to be fixable, could you please update the documentation to be accurate? I don't mind hitting this problem, but I'm annoyed to hit in in Feb 2016 and it's known in Oct 2015 but only mentioned here.

And that the SD card isn't disconnected either.

FWIW, I have also set all pins to inputs, pulled low with the following code:

Code: Select all

def lo_power():
    from machine import Pin, deepsleep
    for pin in dir(Pin.board):
        p = Pin(pin)
        p.init(Pin.IN, pull=Pin.PULL_DOWN, alt=-1)
    deepsleep()

This has made no measurable difference. I haven't tried off the expansion board yet.

[TinheadNed]

WiPy on its own gets me about 350uA FYI, so still out by a factor of 70.

[mbevilacqua]

Well... that's quite disappointing to say the least.