This code works (I can hear the relays clicking, and see 0x26 on the i2c scan) (note custom component module4relay is basically unit4relay adapter for a different register to enable the relays)

esphome:
  name: relay-test-s3
  friendly_name: relay-test-s3


esp32:
  board: esp32-s3-devkitc-1
  framework:
    type: arduino

# Enable logging
logger:

# Enable Home Assistant API
api:
  encryption:
    key: "Kh5g1ThAQ3SUqKw9GnyJUI5PyOkndNl4+ykbxHoUIjc="

ota:
  - platform: esphome
    password: "5f65f9ff9e4ab8b3c928a7216c551525"

wifi:
  ssid: !secret wifi_ssid
  password: !secret wifi_password

  # Enable fallback hotspot (captive portal) in case wifi connection fails
  ap:
    ssid: "Relay-Test Fallback Hotspot"
    password: "G1ypogSTYlkb"

captive_portal:

i2c:
  - id: bus_internal 
    sda: GPIO12
    scl: GPIO11
    scan: true 

  - id: bus_port_a 
    sda: GPIO2
    scl: GPIO1
    scan: true 

external_components:
  - source: github://ngolf/esphome-yaml/components
    components: module4relay

module4relay:
  i2c_id: bus_internal

switch:
  - platform: module4relay
    relay_1:
      name: "Module Relay Channel 1"
      restore_mode: RESTORE_DEFAULT_OFF
    relay_2:
      name: "Module Relay Channel 2"
      restore_mode: RESTORE_DEFAULT_OFF
    relay_3:
      name: "Module Relay Channel 3"
      restore_mode: RESTORE_DEFAULT_OFF
    relay_4:
      name: "Module Channel 4"
      restore_mode: RESTORE_DEFAULT_OFF

But adding the axp2101 component (which I need for the backlight) makes the relay stopping being found on i2c scan, and relays inoperable.

(external_components:)
  - source: github://ngolf/esphome-yaml/components
    components: module4relay

axp2101:
  id: axp2101_pmu
  i2c_id: bus_internal

I am trying to configure the Atom S3 Lite in order to toggle its integrated led on Button release.
This is what I have:

light:
  - platform: fastled_clockless
    chipset: WS2812B
    pin: GPIO35
    num_leds: 4
    rgb_order: GRB
    id: led_1
    name: led_1
    restore_mode: RESTORE_DEFAULT_OFF
    effects:
      - pulse:
          name: "Pulse"
          transition_length: 1s
          update_interval: 1s
      - strobe:
          name: "Strobe"
      - random:
          name: "Random"
          transition_length: 5s
          update_interval: 3s    

binary_sensor:
  - platform: status
    name: "Node Status"
    id: system_status
  - platform: gpio
    name: Button
    pin:
      number: GPIO41
      inverted: true
      mode:
        input: true
        pullup: true
    on_press:
      then:
        - homeassistant.service:
            service: ssh_command.exec_command
            data:
              host: "xxxx"
              port: "xx"
              user: "xxxxxxxx"
              pass: "xxxxxxxxx"
              command: "DISPLAY=:0 mumble rpc togglemute"
    on_release:
      then:
        - light.turn_on:
            id: led_1
            effect: "Random"

And this is the error that I am getting:

In file included from .piolibdeps/esphome-web-6670e8/FastLED/src/third_party/yves/I2SClockLessLedDriveresp32s3/driver.h:9,
                 from .piolibdeps/esphome-web-6670e8/FastLED/src/platforms/esp/32/clockless_i2s_esp32s3.cpp:19:
.piolibdeps/esphome-web-6670e8/FastLED/src/third_party/yves/I2SClockLessLedDriveresp32s3/src/I2SClockLessLedDriveresp32s3.h:38:10: fatal error: esp_lcd_panel_interface.h: No such file or directory

***************************************************************************************
* Looking for esp_lcd_panel_interface.h dependency? Check our library registry!
*
* CLI  > platformio lib search "header:esp_lcd_panel_interface.h"
* Web  > https://registry.platformio.org/search?q=header:%1B%5Bm%1B%5BKesp_lcd_panel_interface.h
*
***************************************************************************************

   38 | #include "esp_lcd_panel_interface.h"
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.
*** [.pioenvs/esphome-web-6670e8/lib6ad/FastLED/platforms/esp/32/clockless_i2s_esp32s3.cpp.o] Error 1
Compiling .pioenvs/esphome-web-6670e8/lib6ad/FastLED/platforms/wasm/ui/title.cpp.o
========================= [FAILED] Took 215.49 seconds =========================

Dou you have any idea of the problem?

Thank you!

I am starting a project and already made some progress with my software. I am using SPI and UART (among other stuff).

HardwareSerial provides an OnReceiveCallback. Good. But what about Transmitting? HardwareSerial.Write transmits data. Or better, it buffers it in the background and transmits it. When will this be finished? How can I detect an ongoing/finished transmission? Why is there no OnTransmitFinishedCallback?

Similar with SPI:
this enables CS line, transmits data and disables CS line.

But this is blocking. Even if SPI frequencies are high and such transmissions are only taking a few 100microseconds. Why is this not done similiar like UART with background transmission and OnReceivedCallback or similar?

In HardwareSerial I was able to find this OnReceiveCallback after digging a lot in the provided libs. But what about SPI?

logger:
  baud_rate: 0
  hardware_uart: UART0
  level: VERBOSE
  initial_level: DEBUG
  logs: 
    usb_uart: VERBOSE
    usb_host: VERBOSE
    uart_debug: VERBOSE

Outputs

output:
  - platform: axp2101
    type: range
    channel: DLDO1
    id: lcd_backlight_output
    min_voltage: 2600
    max_voltage: 3300

  - platform: axp2101
    channel: ALDO1
    voltage: 1800

  - platform: axp2101
    channel: ALDO2
    voltage: 3300
  
  - platform: axp2101
    channel: BLDO1
    voltage: 2800
  
  - platform: axp2101
    channel: BLDO2
    voltage: 1500

Then I set these switches:

switch:
  - platform: gpio
    name: "bus_out_en_sw"
    internal: true
    pin:
      aw9523b: aw9523b_hub
      number: 1
    restore_mode: ALWAYS_ON

  - platform: gpio
    name: "usb_otg_en_sw"
    internal: true
    pin:
      aw9523b: aw9523b_hub
      number: 5
    restore_mode: ALWAYS_ON

  - platform: gpio
    name: "boost_en_sw"
    internal: true
    pin:
      aw9523b: aw9523b_hub
      number: 15
    restore_mode: ALWAYS_ON

  - platform: gpio
    name: "AW RST P0_2"
    internal: true
    pin:
      aw9523b: aw9523b_hub
      number: 2
    restore_mode: ALWAYS_ON

And configure the usb_uart

usb_uart:  
  - type: cdc_acm
    id: scanner_bus
    vid: 0x152A
    pid: 0x880F
    channels:
      - id: uch_1
        debug: true

The device is a barcode scanner Digitus DA-8002
DC 3.3 - 5V, 120 mA, idle 1uA
The scanner is set as "USB serial interface output".
Adapter USB-C to USB-A Greencell 2.4A 480 Mbps
Power supply CoreS3: Yingjiao YS35V-1203000, 12V, 3A, 36W
Baudrate scanner: 9600
The scanner is powered. When I plug it in, it beeps, and it also beeps when I scan.
It has a USB-A interface, and I’m using a USB-C to USB-A OTG adapter.
The scanner works without any problem with this exact adapter on a Windows PC and on an Android smartphone.
This is the log when I plug in the scanner.

[18:14:15.614][D][usb_host:265]: Open device 1
[18:14:15.615][D][usb_host:272]: Get descriptor device 1
[18:14:15.615][D][usb_host:279]: Device descriptor: vid 152A pid 880F
[18:14:15.616][D][usb_host:289]: Device connected: Manuf: MINJCODE; Prod: MINJCODE MJ2818A; Serial: 00000000050C
[18:14:15.618][V][usb_host:089]: *** Device descriptor ***
[18:14:15.618]bLength 18
[18:14:15.618]bDescriptorType 1
[18:14:15.618]bcdUSB 1.10
[18:14:15.618]bDeviceClass 0x2
[18:14:15.618]bDeviceSubClass 0x0
[18:14:15.618]bDeviceProtocol 0x0
[18:14:15.618]bMaxPacketSize0 64
[18:14:15.618]idVendor 0x152a
[18:14:15.618]idProduct 0x880f
[18:14:15.618]bcdDevice 1.00
[18:14:15.618]iManufacturer 1
[18:14:15.618]iProduct 2
[18:14:15.618]iSerialNumber 3
[18:14:15.618]bNumConfigurations 1
[18:14:15.627][V][usb_host:070]: *** Configuration descriptor ***
[18:14:15.627]bLength 9
[18:14:15.627]bDescriptorType 2
[18:14:15.627]wTotalLength 67
[18:14:15.627]bNumInterfaces 2
[18:14:15.627]bConfigurationValue 1
[18:14:15.627]iConfiguration 0
[18:14:15.627]bmAttributes 0x80
[18:14:15.627]bMaxPower 500mA
[18:14:15.652][V][usb_host:056]: 	*** Interface descriptor ***
[18:14:15.652][V][usb_host:056]: 	bLength 9
[18:14:15.652][V][usb_host:056]: 	bDescriptorType 4
[18:14:15.652][V][usb_host:056]: 	bInterfaceNumber 0
[18:14:15.652][V][usb_host:056]: 	bAlternateSetting 0
[18:14:15.652][V][usb_host:056]: 	bNumEndpoints 1
[18:14:15.652][V][usb_host:056]: 	bInterfaceClass 0x2
[18:14:15.652][V][usb_host:056]: 	iInterface 0
[18:14:15.669][V][usb_host:042]: 		*** Endpoint descriptor ***
[18:14:15.669][V][usb_host:042]: 		bLength 7
[18:14:15.669][V][usb_host:042]: 		bDescriptorType 5
[18:14:15.669][V][usb_host:042]: 		bEndpointAddress 0x82	EP 2 IN
[18:14:15.669][V][usb_host:042]: 		bmAttributes 0x3	INT
[18:14:15.669][V][usb_host:042]: 		wMaxPacketSize 8
[18:14:15.669][V][usb_host:042]: 		bInterval 10
[18:14:15.679][V][usb_host:056]: 	*** Interface descriptor ***
[18:14:15.679][V][usb_host:056]: 	bLength 9
[18:14:15.679][V][usb_host:056]: 	bDescriptorType 4
[18:14:15.679][V][usb_host:056]: 	bInterfaceNumber 1
[18:14:15.679][V][usb_host:056]: 	bAlternateSetting 0
[18:14:15.679][V][usb_host:056]: 	bNumEndpoints 2
[18:14:15.679][V][usb_host:056]: 	bInterfaceClass 0xa
[18:14:15.679][V][usb_host:056]: 	iInterface 0
[18:14:15.692][V][usb_host:042]: 		*** Endpoint descriptor ***
[18:14:15.692][V][usb_host:042]: 		bLength 7
[18:14:15.692][V][usb_host:042]: 		bDescriptorType 5
[18:14:15.692][V][usb_host:042]: 		bEndpointAddress 0x1	EP 1 OUT
[18:14:15.692][V][usb_host:042]: 		bmAttributes 0x2	BULK
[18:14:15.692][V][usb_host:042]: 		wMaxPacketSize 64
[18:14:15.692][V][usb_host:042]: 		bInterval 0
[18:14:15.699][V][usb_host:042]: 		*** Endpoint descriptor ***
[18:14:15.699][V][usb_host:042]: 		bLength 7
[18:14:15.699][V][usb_host:042]: 		bDescriptorType 5
[18:14:15.699][V][usb_host:042]: 		bEndpointAddress 0x81	EP 1 IN
[18:14:15.699][V][usb_host:042]: 		bmAttributes 0x2	BULK
[18:14:15.699][V][usb_host:042]: 		wMaxPacketSize 64
[18:14:15.699][V][usb_host:042]: 		bInterval 0
[18:14:15.719][D][usb_host:304]: Call on_connected
[18:14:15.721][D][usb_uart:331]: device connected
[18:14:15.730][D][usb_uart:031]: intf_desc: bInterfaceClass=02, bInterfaceSubClass=02, bInterfaceProtocol=01, bNumEndpoints=1
[18:14:15.740][D][usb_uart:041]: ep: bEndpointAddress=82, bmAttributes=03
[18:14:15.750][D][usb_uart:031]: intf_desc: bInterfaceClass=0A, bInterfaceSubClass=00, bInterfaceProtocol=00, bNumEndpoints=2
[18:14:15.760][D][usb_uart:041]: ep: bEndpointAddress=01, bmAttributes=02
[18:14:15.771][D][usb_uart:041]: ep: bEndpointAddress=81, bmAttributes=02
[18:14:15.771][V][usb_uart:081]: Found CDC-ACM device
[18:14:15.781][D][usb_uart:338]: Found 1 CDC-ACM devices
[18:14:15.783][D][usb_uart:390]: enter enable_channels
[18:14:15.791][D][usb_uart:398]: start_input for channel
[18:14:15.793][D][usb_uart:278]: call transfer_in epAddress: 0x81, MaxPacketSize: 64
[18:14:15.802][D][usb_uart:283]: transfer_in returns true
[18:14:15.802][W][component:547]: usb_host took a long time for an operation (189 ms)
[18:14:15.812][W][component:550]: Components should block for at most 30 ms
[18:14:15.826][D][usb_host:174][usb_task]: New device 1

As it looks, the transfer is submitted with usb_host_transfer_submit. There are absolutely no error messages and everything looks correct. The callback is just never called—no matter what, how much, or how fast I scan.

What did I forget? Is this even possible at all?

indeed on the pictures the back of both SE and Lite are marked SE - strange.

Anyways, have you tied to burn and run the M5CoreS3 demo firmware? This should give you a good indication about whether the camera works or not.

Please note: I do not have an M5CoreS3 Lite to test this myself.

Thanks
Felix

the battery connector powers everything, including RTC. But, yes, when the external battery is fully drained RTC time is lost as there is no built-in battery for the RTC.

You can find the schematics here.

Thanks
Felix

I see your config puts the reset pin definitions directly in the correct driver sections for the touchscreen and LCD so that those driver handle them correctly. I think this is better than doing it separately the way I suggested :-)

The device boots, connects to WiFi, and the Bluetooth Proxy works perfectly. However, the display refuses to initialize. The backlight stays on, but the screen remains white (no pixels).

The logs show a clear memory allocation failure:

[E][display:017]: Could not allocate buffer for display!

[E][component:119]: Component display was marked as failed.

I suspect I am hitting a hard hardware ceiling because this board lacks PSRAM (it only has ~320KB internal RAM). It seems like the combination of these three components exceeds the available memory:

1. WiFi/API: Baseline load.
2. Bluetooth Proxy: Reserves a massive contiguous block.
3. Display (320x240): The ili9xxx driver seems to need ~153KB for the frame buffer. Since the display initializes last, it finds zero contiguous RAM left and fails.

To test this theory, I crippled the display configuration to use less memory. By setting dimensions: 320x60, I forced the buffer size down to ~38KB.

This immediately fixed the crash. The device booted, the text appeared (in the top 60px strip), and Bluetooth continued working.

Is this a known hardware limitation of the M5Stack Basic? Has anyone managed to run a Bluetooth Proxy AND a Full Resolution (320x240) Display on an ESP32 without PSRAM?

I am wondering if there is a way to optimize the Bluetooth stack size or the display buffer (maybe partial updates?) to make them fit, or if I am simply asking for too much from this hardware.

https://github.com/RikerZhu/ESPHome-M5Unit-Miniscale.git