![]() Probably I'll use AI Thinker instead of Timer Camera X to save power, I'd prefer to use Timer Camera X as it has a plastic case and pcb seems to haver better quality but it needs 7 times more power to work. I tried to put RESET_GPIO_NUM (GPIO 15 in Timer Camera X) to 0 to disable OV3660, but power consumption is the same. Unfortunately according with schematics, in Timer Camera X I cannot change PWDN_GPIO_NUM as it's not connected to any pin in ESP32 (PWDN pin is connected to GND through a 10K resistor). Those 22-24 mA are similar values that I got with Timer Camera X, I thing that the real "culprit" of power consumption is OV3660, not USB-TTL. If I put PWDN_GPIO_NUM to 0, power cosumption jumps to 22-24 mA. I cannot reach the deep sleep power consumption probably due voltage regulators and another electronics in device, but is a far better number that the 18 mA from Timer Camera X. Connected to 5V, AI Thinker device only uses 2.7 mA in deep sleep (4.7 mA in light sleep). I saw that with AI Thinker device, if I put PWDN_GPIO_NUM to 1 (GPIO32 in AI Thinker), OV3660 is powered down. I attached the OV3660 from Timer Camera X to AI Thinker (Devices have same socket, but connected to different pins, I only needed to modify pin defines). ![]() My_ (34.13 KiB) Viewed 2910 times ESP32_S3_DEV2.jpg (1.9 MiB) Viewed 2911 times ESP_DEV_BOARD1.jpg (2.This weekend I made some tests with an ESP32-Cam (The AI Thinker model) that doesn't have USB - ttl converter and I compared results with Timer Camera X Is it possible that they leak so much current through the voltage dividers? ![]() These 3 above are main components that should draw current, we dont use much else apart from 3 ADC inputs. MYRGP330100W21RATR Max Quiescent Current - 800uA According to the ESP32 documentation, it is possible to achieve even lower than 100uA consumption which is would be more than ideal.Īccording to the components we use and their datasheets, the current consumption we expect: Anything above 1-2mA is not acceptable for us. We are trying to understand the reason why our custom device is drawing so much current in the deep sleep mode. I think the reason why ESP32 Devkitv1 is using so much current is because they use bad voltage regular that is leaking a lot of current. Measured current in deep sleep mode of our device : 5.1mA We have checked the datasheets of the SPI EEPROM memory, I2C expander and their current consumption in standy modes are very small (in uA range). The output of this IC is connected in series with Ampmeter and powers the ESP32 device as well as all other peripherals that we use (SPI EEPROM memory, I2C gpio expander and etc). To measure our custom board power consumption we supply the 4.2V battery input to the voltage regulator IC (MYRGP330100W21RATR). Our custom ESP32 board with ESP32-WROOM-32UE chip. To measure the current consumption of the ESP32 S3 dev board we have also used the same method as above. To measure the current consumption of the ESP32 Devkitv1 we used 3.3 Power supply in series with ampermeter (UNI-T UT70A). We have tested power consumption of 3 different boards: (chip_info.features & CHIP_FEATURE_EMB_FLASH) ? "embedded" : "external") Įsp_sleep_enable_ext0_wakeup(GPIO_NUM_0, 0) Printf("silicon revision %d, ", chip_info.revision) (chip_info.features & CHIP_FEATURE_BLE) ? "/BLE" : "") (chip_info.features & CHIP_FEATURE_BT) ? "/BT" : "", Printf("This is %s chip with %d CPU core(s), WiFi%s%s, ", Code: Select all CONDITIONS OF ANY KIND, either express or implied.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |