Electron Lipo current in deep sleep mode


#1

When my Electron is in deep sleep [System.sleep(SLEEP_MODE_DEEP)] and waiting to be woken by the WKP pin the Lipo battery connection draws about 170uA. The current waveform consists of high pulses of current about 180mA (yes mA) that are very fast (about 6uS) and are spaced about 1.6ms apart. This accounts for the DC ammeter measurement of 170uA.

What is the cause of these pulses and how can I eliminate them? I’d like a sleep current below 5uA for my application.

I assume I need configure the PMIC to eliminate the pulses. I don’t see much information about how to do this. Any tips?


#2

What system firmware version have you got on your device?


#3

0.6.0


#4

It’s always best to try with most recent versions (stable 0.6.4, pre-release 0.7.0-rc.6 or 0.8.0-rc.1).
IIRC there have been some tweak to the sleep modes and general power consumption since 0.6.0.

Look here for more detail (starting with 0.6.1-rc.1 down to now)


#5

I tried 0.6.4 and get similar current numbers. I’m not seeing anything in the pre-release versions that shows changes to power.

I keep thinking it has to be related to PMIC because the micro sleep current should be nearly nothing.


#6

What type of meter are you taking the current measurements with?


#7

To see the current pulse I used a Tektronix TCP0030A probe on the wire to the Lipo battery. The USB port was not connected.

For equivalent DC current I used a Keysight 34461A meter in series with a 3.7V power supply. I placed four 470u caps on at the JST connector and measured the current in to the caps. The caps were important because even though the DC supply can source 8A there is enough lead inductance to keep the module from operating correctly. Plus the meter adds to the ESR. The result is probably not exactly the same as a Lipo battery but it still shows me the basics.

The area under the pulse when spread out would be about the same as the DC value.