[support] New Muse Lux device, dead on arrival (unstable)

Support request for a new device that is dead on arrival of sorts. I did get one (or two) successful power-up magically while I was typing this post.

tl;dr device appears DOA, thought I finally got some logs, its unstable. I cannot seem to get it functioning.

I received the device the other day and tried using it but didn’t get anything from it.

I turned it on, no effect (no sound, no led). Perhaps battery is discharged for transportation, so I plugged it in to charge. Red charging light came on.

I waited a few minutes and tried turning it on, no effect. Checked for Bluetooth, nothing appears.

Left it charging over night, tried again, no effect. Though, I could subtle hear the speaker loosing power when powered off. So I know that’s working.

Took it to my desktop and plugged in it with provided USB cable, opened up ESP Web^1 to view device logs. Device connects but there’s nothing on the terminal. Power cycled, no effect. Tried resetting the device, nothing shows up either. I also attempted flashing, but it just gets stuck preparing.

I did look at trying to pry it open and get to the ESP32 itself, but it wasn’t clear to me how to open it. It’s rather slick/professional looking, I’d rather not break it trying to get to the board.

At some point, after hitting Reset and disconnecting it from my computer it randomly started. Literally while I was typing this is turned on and TTS’s “Bluetooth”, led on.

I’m not sure what’s going on there, it feels like power-cycling leaves the device hung. I’ve been trying to repeat the last steps at getting the device starting to see if its reproducible. But I haven’t had much success. I was able to pull some logs at some point, I’m unable^2 to attach them to the post, so I’ve uploaded logs elsewhere.

Observations:

  • After power cycling, it appears the device may have to be rest. (unclear)
  • Device arrives in a DOA-like appearance
  • Power switch has a mushy feeling. As a user, it makes me think its not working. (hearing the subtle speaker crackle on power-off tells me it is).

I’d rather not return it, so I’m posting this to request support. Thank you!

Thank you very much!

^1: Tried both from raspiaudio.github.io and then also to the recommended sle118/squeezelite-esp32.
^2: Only image file type extensions are currently enabled on the site. It could be beneficial to permit txt/log files too. I believe admins can configurable discourse via “… site setting is authorized extensions under the files section of site settings.”

Of course, having posted this I just spotted instructions for resetting the device’s auto-bootloader. I will give that a try later today or tomorrow and see if it address the problem. I’ll update this post when I’ve done that.

Update:

Hidden-reset button (inside the 3.5mm headphone) had no effect. I tried with esptools v4.7 as well, and could never get a connection, simply:

> .\esptool.exe --port=COM4 chip_id
esptool.py v4.7.0
Serial port COM4
Connecting......................................
A fatal error occurred: Failed to connect to Espressif device: No serial data received.
For troubleshooting steps visit: https://docs.espressif.com/projects/esptool/en/latest/troubleshooting.html

Reading the thread further, I found someone reporting an issue with the Windows drivers.

Looking at my device manager, I too have the 2023 drivers:

Silicon Labs CP210x USB to UART Bridge
Silicon Laboratories Inc.
4/13/2023
11.3.0.176

So I’ve downloaded:

CP210x Windows Drivers
v6.7.6
9/3/2020

Ran the installer CP210xVCPInstaller_x64.exe, went to the device in Device Manager and opted to select from files on my computer to switch to this driver version. Replugged the device, still no effect. Nothing will connect, can’t read the chip or flash anything at all with esptool.

Trying all of that, I turned off the device, loaded up PuTTY, connected it and turned it on.
For awhile there is nothing, then after power cycling/reset/reconnecting it I did at one point see output on the terminal, but only after switching the device off:

ets Jun  8 2016 00:22:57

rst:0x10 (RTCWDT_RTC_RESET),boot:0x3f (SPI_FAST_FLASH_BOOT)
flash read err, 1000
ets_main.c 371
ets Jun  8 2016 00:22:57

rst:0x10 (RTCWDT_RTC_RESET),boot:0x2f (UART_BOOT(UART0))

I’m rather new to esp32s, but I’m honestly unsure what else to do at this point.

Update 2:

tl;dr, case design is blocking power switch from moving fully into “on” and latching properly, probably causing bouncing or unreliable electrical connection.

Okay, well after sitting here working on prying the face off it just suddenly turned on and announced “bluetooth”. So that made me suspicious. That’s happened twice in the last day or three.

Finally. I ended up prying it open and while examining it accidentally touched the power button. It’s mushiness was the root cause. I can now reproducibly and successfully reboot the device with the PCB out of the case. Moving the switch + extender has a satisfying ‘click’. The slot on the case has insufficient clearance!

ets Jun  8 2016 00:22:57

rst:0x1 (POWERON_RESET),boot:0x17 (SPI_FAST_FLASH_BOOT)
configsip: 0, SPIWP:0xee
clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00
mode:DIO, clock div:1
load:0x3fff0018,len:4
load:0x3fff001c,len:1216
ho 0 tail 12 room 4
load:0x40078000,len:10944
load:0x40080400,len:6388
entry 0x400806b4
 SPIFFS used bytes  ====> 408628 of 896321
volume ==> 50
volume ==> 50
0  ====> /bluetooth.wav
 rate = 48000
volume ==> 50
Battery : 1073471928

Brownoutets Jun  8 2016 00:22:57

rst:0x10 (RTCWDT_RTC_RESET),boot:0x3f (SPI_FAST_FLASH_BOOT)
flash read err, 1000
ets_main.c 371
ets Jun  8 2016 00:22:57

rst:0x10 (RTCWDT_RTC_RESET),boot:0x17 (SPI_FAST_FLASH_BOOT)
configsip: 0, SPIWP:0xee
clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00
mode:DIO, clock div:1
load:0x3fff0018,len:4
load:0x3fff001c,len:1216
ho 0 tail 12 room 4
load:0x40078000,len:10944
load:0x40080400,len:6388
entry 0x400806b4
 SPIFFS used bytes  ====> 408628 of 896321
volume ==> 50
volume ==> 50
0  ====> /bluetooth.wav
 rate = 48000
volume ==> 50
Battery : 1073471928
Battery : 1073471928
Battery : 1073471928
Battery : 1073471928
Battery : 1073471928
Battery : 1073471928
Battery : 1073471928
Battery : 1073471928
Battery : 1073471928

Brownoutets Jun  8 2016 00:22:57

rst:0x10 (RTCWDT_RTC_RESET),boot:0x3f (SPI_FAST_FLASH_BOOT)
flash read err, 1000
ets_main.c 371
ets Jun  8 2016 00:22:57

rst:0x10 (RTCWDT_RTC_RESET),boot:0x17 (SPI_FAST_FLASH_BOOT)
configsip: 0, SPIWP:0xee
clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00
mode:DIO, clock div:1
load:0x3fff0018,len:4
load:0x3fff001c,len:1216
ho 0 tail 12 room 4
load:0x40078000,len:10944
load:0x40080400,len:6388
entry 0x400806b4
 SPIFFS used bytes  ====> 408628 of 896321
volume ==> 50
volume ==> 50
0  ====> /bluetooth.wav
 rate = 48000
volume ==> 50
Battery : 1073471928
Battery : 1073471928

Brownoutets Jun  8 2016 00:22:57

rst:0x10 (RTCWDT_RTC_RESET),boot:0x2f (UART_BOOT(UART0))

Which brings me to trying to flash. I think I’ve finally got a sequence that boots the device properly. All others left me in a UART_BOOT(UART0) mode which doesn’t appear to be correct.

Update 3:
I did finally get it right, but it required something like: connecting to the device which resets it, and you have to be holding the boot button. But then, also navigating to the console and resetting it again while holding the boot button. At which point, backing out and flashing screen. I don’t recall this being the process on other esp32’s but, I’m new to the devices. So shrug. Just leaving this here for anyone that runs into similar problems.

Thanks for signaling that the on/off switch is too tight with the case. We will fix this in the new batch this month. Not all Luxe units have this issue, and you are still eligible for a refund. I will also publish a DIY solution with a small modification of the casing.

1 Like

I think the switch issues is why nothing looked like it was working correctly. I should be able to enlarge the hole myself, but I’ll wait for the DIY solution guide :). I had read elsewhere this was an issue on prior (usb-c?) models, so I didn’t really pay much attention to the feeling.

Right now I’m trying to figure out why neither squeezelite-esp32 nor my docker-based test are happy with LMS nor hass-music-assistant. The fun never ends!