Xiao ESP32-C3 and ESP32-S3 deep sleep current high

[gugray]

I’ve been experimenting and taking measurements for two days and I cannot wrap my head around the current readings I’m getting for my Xiao ESP32-C3 and ESP32-S3 boards.

I am getting a consistent 80uA for the C3 and 300uA for the S3. I have tried two boards for each.

• Pasting my code below
• I am getting the same reading when I’m powering the board via the battery pins at the back, and via the GND and 3V3 pins
• For the measurement USB is disconnected and the power source is 3V from a pair of fresh AA batteries
• No pins of the board are connected to anything (no peripherals)

I have been scouring the internet for info, including these links:
forum.seeedstudio .com/t/xiao-esp32s3-deep-sleep-current-higher-than-expected/276916
forum.seeedstudio .com/t/xiao-esp32s3-deep-sleep/276152
esp32 .com/viewtopic.php?t=32177
esp32 .com/viewtopic.php?t=22622
forum .seeedstudio.com/t/charging-characteristics-of-xiao-nrf52840-and-xiao-esp32c3/267686
forum .seeedstudio.com/t/comparison-of-sleep-currents-for-xiao-esp32c6-s3-and-c3/276444

I am running out of ideas, and hope. Any suggestions what else to try, or if I should even hope to get close to the datasheet values (47uA for the C3 and 14uA for the S3)? Is it possible I got boards from a buggy/unlucky batch?

#include <Arduino.h>
#include <esp_sleep.h>

void gotobed()
{
    esp_sleep_pd_config(ESP_PD_DOMAIN_RTC_PERIPH, ESP_PD_OPTION_OFF);
    esp_sleep_pd_config(ESP_PD_DOMAIN_XTAL, ESP_PD_OPTION_OFF);
    esp_sleep_pd_config(ESP_PD_DOMAIN_CPU, ESP_PD_OPTION_OFF);
    esp_sleep_pd_config(ESP_PD_DOMAIN_RTC8M, ESP_PD_OPTION_OFF);
    esp_sleep_pd_config(ESP_PD_DOMAIN_VDDSDIO, ESP_PD_OPTION_OFF);
    esp_sleep_pd_config(ESP_PD_DOMAIN_RTC_SLOW_MEM, ESP_PD_OPTION_OFF);
    esp_sleep_pd_config(ESP_PD_DOMAIN_RTC_FAST_MEM, ESP_PD_OPTION_OFF);
    esp_sleep_disable_wakeup_source(ESP_SLEEP_WAKEUP_ALL);
    esp_sleep_enable_timer_wakeup(5 * 1000 * 1000);
    esp_deep_sleep_start();
}

void firstTimeInit()
{
    delay(2000);
    gotobed();
}

void wakeUp()
{
    delay(1000);
    gotobed();
}

void setup()
{
    int cause = esp_sleep_get_wakeup_cause();
    if (cause == 0)
        firstTimeInit();
    else
        wakeUp();
}

void loop()
{
}

[cgwaltney]

Welcome! I am sure you will get alot of responses on this one!

[msfujino]

I think the low battery voltage is causing the built-in 3.3V regulator to consume more current.
I tried it with C3 and it was about 50uA when 3.8V was applied from the battery pad and 80uA when 3.3V or lower. With S3 10uA and 300uA.

Please also refer to the following links.

Comparison of Sleep Currents for XIAO ESP32C6, S3, and C3

Now that I have three different XIAO ESP32s (C3, S3, and C6), compared their sleep currents. I hope this will be helpful when considering low-power applications. First, I analyzed the schematic and datasheet as a reference. A 3.3V analog regulator XC6210 is used in C3, and this regulator’s quiescent current of 35uA accounts for most of the C3 sleep current. However, a diode NSR0520 is used, which has low reverse current, giving it an advantage in this area compared to S3 and C6. The power sup…

Although not related to sleep current, note the following when using C3 and S3.

1. It is dangerous to supply power from USB-C while a non-LiPo battery is connected to the battery pad.
2. At least 3.3V is required from the battery batts, as it passes through a regulator.
3. When powering from the 3V3 pin, the minimum operating voltage of 3.0V or higher is required for C3 and S3.
4. When using a radio, a battery capable of supplying a peak current of more than 200mA is required.

[gugray]

Thanks for the response @msfujino! I tested the S3 with three AA batteries at 4.5V connected to the battery pads and I got the nominal 14uA current. This solves my problem!

Some thoughts, partly for my own record:

• This only works via the battery pads. Connecting to the 5V pin brings on the battery charging LED, just like when I power the board via USB.
• I didn’t expect the voltage regulator’s consumption to jump like this due to low voltage. The regulators I used in solar low-power projects in the past did not exhibit this behavior.
• I didn’t check if the battery input officially supports 4.5V, but as it’s only 0.3V above the fully charged LiPo voltage of 4.2V, I’m ready to just wing it.

Also thanks for the additional notes! They are all duly noted.

I forgot to include in my original post but here is the platformio.ini file I was using:

[env:seeed_xiao_esp32s3]
platform = platformio/espressif32@^6.9.0
board = seeed_xiao_esp32s3
board_build.mcu = esp32s3
framework = arduino
build_flags =
  -DARDUINO_USB_MODE=1
  -DARDUINO_USB_CDC_ON_BOOT=1
  ;-DCONFIG_PM_ENABLE=1
  ;-DCONFIG_ESP_SLEEP_FLASH_LEAKAGE_WORKAROUND=1
  ;-DCONFIG_ESP_SLEEP_PSRAM_LEAKAGE_WORKAROUND=1
  ;-DCONFIG_ESP_SLEEP_POWER_DOWN_FLASH=1
monitor_speed = 115200

This may be interesting to the extent that:

• Removing the USB related build options did not change anything. The content was not related to the USB chip.
• The build options I commented out above were all things I tried that made no difference. They came up in other posts I found about ESP32 deep sleep current.