Fipy & Expansion board 3.1 - deepsleep power consumption

[czlowiek]

Hi,

I am trying to make FIPY & Expansion board 3.1 consume as minimum energy as possible. The machine deep sleep works fine when only on FIPY

machine.deepsleep(10*1000) 

It makes only FIPY go down to 30uA (power connected to Vin 5V, actual Voltage is 3.3).

However, if connected to Expansion Board 3.1 it draws 290uA. So I can not figure out how to decrease it. According to this
https://forum.pycom.io/topic/4336/new-v3-1-expansion-board-firmware-v0-0-11/6

"No, the Expansion Board doesn't know that the Wipy module is in sleep, but it detects no USB, so puts itself in low power mode. So the ~30uA is obtained when Expansion+Wipy are on battery."

Could you suggest what to research? Is there maybe some pull ups on EB3.1, or something. I have removed almost all gates (leaving TX,RX). Maybe FIPY is doing something on connection with EB3.1.

I guess it has nothing to do with WIFI/BLE/CAT-M1/LORA, because I am not using any of this, disabling wifi on boot, and the same program just on FIPY gives satisfying results (30uA).

[Gijs]

@czlowiek said in Fipy & Expansion board 3.1 - deepsleep power consumption:

"No, the Expansion Board doesn't know that the Wipy module is in sleep, but it detects no USB, so puts itself in low power mode. So the ~30uA is obtained when Expansion+Wipy are on battery."
Could you suggest what to research? Is there maybe some pull ups on EB3.1, or something. I have removed almost all gates (leaving TX,RX). Maybe FIPY is doing something on connection with EB3.1.

Hi,
I believe the additional current is in the orange LED. You can also turn that off through removing the jumper. Other than that, I am not exactly sure where the additional power is lost. There are of course the battery voltage divider resistors for the ADC that lose some current (Depending on the expansion board you have, that is 1M 1M or 115K and 56K.

[rcolistete]

My tests with Expansion Board v3.1 with WiPy3 in deep sleep mode, power from battery (Vbat = 3,61-3,66 V) :

• 34 uA with only TX/RX mini-jumpers;
• 102 uA with all mini-jumpers;
• 183 uA with 8GB Sandisk UHS-I uSD and all mini-jumpers;
• 160 uA with 16GB Toshiba M203 UHS-I uSD and all mini-jumpers;
• 692 uA with 32GB Samsung Evo UHS-I uSD and all mini-jumpers.

So :

• the LED mini-jumpers adds 68 uA;
• microSD card, just inserted, adds 58-590 uA, depending on the uSD card (not only the capacity).

So the question is : do you have a microSD card inserted in your Expansion Board v3.1 ?

[czlowiek]

@rcolistete your answer pointed me to good direction. So I used 3.3V voltage for my device and then the sleep current is 289uA. I moved the voltage to 3.6V (which is limit of power profiler kit I am using) and that actually dropped in deep sleep to 30uA. So I guess that might not trigger EB3.1 to go to deep sleep.

I should have checked more closely in spec - the correct levels are from 3.5 to 5.5 in VIN. I guess I confused it with logic level.

[rcolistete]

About active state, Expansion Board 3.1 has a 'feature' (or bug ?) when using TX/RX mini-jumpers and battery supply (3.61-3.66 V) :

• 34.5-35.0 ma without mini-jumpers;
• 68.8-69.4 ma with only TX/RX mini-jumpers
• 69.6-70.2 ma will all mini-jumpers.

This is a 33.8-34.9 mA increase.

But when using USB power (Vusb = 5.06-5.08 V), this issue doesn't happen, i = 43-44 mA with any mini-jumpers configuration.

[jcaron]

@rcolistete I suppose you mean uA (microamperes) and not mA (milliamperes)...

[czlowiek]

@jcaron It's actually mA because meant in active state (not deep sleep). And this is true, it made me happy it actually consumes less energy. Thanks for the tip @rcolistete

[rcolistete]

@jcaron No, mA as it is active state (not in deep or light sleep).

[robert-hh]

@rcolistete said in Fipy & Expansion board 3.1 - deepsleep power consumption:

34.5-35.0 ma without mini-jumpers;
68.8-69.4 ma with only TX/RX mini-jumpers

That sounds like a critical design in the circuitry. Maybe it's only the TX jumper, where the high output level would source the PIC, which is only powered as intended when supplied via USB. Not good for either chip. A series resistor would help to prevent damage. Too bad that Pycom refuses to publish schematics.
For testing, the jumper could be replaced by a two pin plug with a resistor.

[rcolistete]

@robert-hh Yes, I'd like confimation from Pycom if this is a feature or bug. If a bug, if it can be fixed by new firmware for Expansion Board 3.1.

[robert-hh]

@rcolistete I do not know if that can be fixed by firmware. I have the impression that it may happen, because the PIC is not powered when the board is supplied by battery. Then the current out of the TX pin would run into the ESD protection diode of the PIC pin to the PIC's internal power lines.

[rcolistete]

Posted a new topic, "Guide/Tutorial/Reference about Expansion Board", which includes the measurements cited above by me and many other current, voltage and power measurements about all Expansion Board models.