I have an esp32 nodemcu hooked up to a powerbank. My issue is that the power consumption is so low, that the powerbank thinks there's nothing connected to it and some electronics inside then tell it after a few minutes to stop delivering power to that port.

It's running a Bluetooth GATT server, so my next attempt is to make it try to constantly connect to an WiFi AP in parallel, but I'd like to know if there is another way, ie by using a variable resistor, but I'm not sure where to put it.

It should be as compact as possible, so no adapter between the USB plugs.

  • 1
    \$\begingroup\$ Another duplicate, electronics.stackexchange.com/q/286535/117785 \$\endgroup\$ Commented Aug 26, 2017 at 17:32
  • \$\begingroup\$ i would try turning on the dummy load periodically instead of constantly; usually power banks have a 5 second or so window, so if you pulsed 250ms of load every 4 seconds, you draw 16th the power of a constant load, and it still stays "awake" \$\endgroup\$
    – dandavis
    Commented Aug 27, 2017 at 22:11

1 Answer 1


Yes, you can just hook up use a resistor to generate some load.

However, I'd just get a power-bank that does not has this issue.

  • \$\begingroup\$ So where do I put it? Between the 5V and GND? Am I not risking something with the "max load on pins" issue, or is that only valid for data pins and the 3.3V? (I assume the 5V is directly fed from the USB voltage, while 3.3V has to run through a regulator, which might not like that extra load) \$\endgroup\$
    – Daniel F
    Commented Aug 26, 2017 at 17:20
  • \$\begingroup\$ At 5V a 100 Ohm resistor will draw 50mA of current which will turn 1/4 Watt of energy into heat. That should be enough for your power-bank to detect a load. So a 1/2 Watt 100 Ohm resistor will be fine. It will also not get hot enough to cause a fire. Still, the better solution is to get a less picky power-supply. \$\endgroup\$ Commented Aug 26, 2017 at 17:23
  • \$\begingroup\$ It's a power-bank which has solar charging capabilities (albeit a very cheap one, panels are integrated) so I think the power-off "feature" is related to that. Since this will power a device mounted on a bike, I kind of like the idea of having it getting charged by the sun. \$\endgroup\$
    – Daniel F
    Commented Aug 26, 2017 at 17:24
  • \$\begingroup\$ well, then you'd at the same time hate the idea of getting it discharged by a dummy load, wouldn't you? I'd really look into whether maybe shortly drawing a significant current and then continuing to draw whatever your device draws is sufficient for the powerbank to stay on. In any case, this powerbank is not behaving like it should. \$\endgroup\$ Commented Aug 26, 2017 at 17:32
  • \$\begingroup\$ @MarcusMüller, the pulsed load idea was suggested by Passerby here, electronics.stackexchange.com/a/286606/117785 The idea doesn't work all time because powerbanks seems to check the output load with certain periodicity, and if the pulsed load is not syncronized, the powerbank will eventually turn itself off. There was a heated discussion on this topic. \$\endgroup\$ Commented Aug 26, 2017 at 17:50

Not the answer you're looking for? Browse other questions tagged or ask your own question.