Skip to main content

Current draw of a D1 Mini in deep sleep

To measure the deep sleep current draw of the D1 Mini, I hooked up 4 NiMH AA batteries in series (~5.2V) to its 5V and GND pins.

The sketch uploaded to the D1 Mini was a nominal:

void setup() {
  ESP.deepSleep(60)*60*1000000UL, WAKE_RF_DEFAULT);

void loop() {

The current draw was a pretty steady 0.8mA, or 800uA. That's a obviously a far cry from the sub-100uA reportedly achievable with the barebones ESP-01 due to all the extra components on the D1 Mini.

Reported deep sleep current draw for the D1 Mini is all over the place, from 0.21mA (5V), to 0.3mA (3.3V), to 6mA (USB)!


1. Connecting a 18650 battery (~4V) to the 5V pin did not work. In theory, the MC6211 LDO used by the D1 Mini means anything higher than 3.56V should work. But when connected, the onboard LED started to flash in a slow but erratic fashion, I suspect it is randomly resetting (because each time the D1 Mini powers up, the onboard LED flashes briefly).

2. Connecting a 18650 battery to the 3.3V pin did work, though in theory it shouldn't be done because the 3.3V is not connected to any voltage regulator. So the 4V goes directly to the ESP8266, which has a theoretical upper limit of 3.6V. But it did work for me (others have reportedly connected up to 5V to the ESP8266 with no problems, but don't push your luck!), and the deep sleep current draw was ~1.5mA. I suspect if I could hook up a regulated 3.3V source to the pin, the deep sleep current draw should fall closer to the 0.8mA range.

3. Connecting 2 x alkaline AA (3.2V) to the 3.3V pin did not work. The onboard LED started to flash erratically again. In theory, this should work, since the ESP8266 has a minimum operating voltage of 2.5V, but there have been many reports of the chip being finicky with the input voltage, so who knows.


Popular posts from this blog

Update: Line adapter for Ozito Blade Trimmer

Update (Dec 2021): If you access to a 3D printer, I would now recommend this solution , which makes it super easy to replace the trimmer line. I have been using it for a few months now with zero issue.

3D Printer Filament Joiner

I have been looking at various ways of joining 3D printing filaments. One method involves running one end of a filament through a short PTFE tubing, melting it with a lighter or candle, retracting it back into the tubing and immediately plunging the filament to be fused into the tubing: One problem with this method is that you can't really control the temperature at which you melt the filament, so you frequently end up with a brittle joint that breaks upon the slightest bend. Aliexpress even sells a contraption that works along the same line. As it uses a lighter or candle as well, it suffers from the same weakness. I am not even sure why you need a special contraption when a short PTFE tubing will work just as well. Another method involves using shrink tubing/aluminium foil, and a heat gun: But a heat gun is rather expensive, so I wanted to explore other alternatives. The candle + PTFE tubing method actually works quite well when you happen to melt it at the rig

Attiny85 timer programming using Timer1

This Arduino sketch uses Timer1 to drive the LED blinker: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 /* * Program ATTiny85 to blink LED connected to PB1 at 1s interval. * Assumes ATTiny85 is running at 1MHz internal clock speed. */ #include <avr/io.h> #include <avr/wdt.h> #include <avr/sleep.h> #include <avr/interrupt.h> bool timer1 = false , led = true ; // Interrupt service routine for timer1 ISR(TIMER1_COMPA_vect) { timer1 = true ; } void setup() { // Setup output pins pinMode( 1 , OUTPUT); digitalWrite( 1 , led); set_sleep_mode(SLEEP_MODE_IDLE); // Setup timer1 to interrupt every second TCCR1 = 0 ; // Stop timer TCNT1 = 0 ; // Zero timer GTCCR = _BV(PSR1); // Reset prescaler OCR1A = 243 ; // T = prescaler / 1MHz = 0.004096s; OCR1A = (1s/T) - 1 = 243 OCR1C = 243 ; // Set to same value to reset timer1 to