Good morning,
I’ve been struggling to flash my new Emporia Vue 3 since last Friday. I purchased the USB TTL recommended in Digiblur’s YT video and corresponding write up, and I first tried to do this using the 3d printed template, and I could not get a connection. I then broke down and bought the BDM Frame that digiblur used, and I’ve set it up multiple times, and I’m getting no where.
I’m running MacOS, and I’ve installed the esptool python 3 library so I can run the flash-id command.
tamorgen@Mac-Studio ~ % python3 -m esptool flash-id
esptool v5.2.0
Found 2 serial ports...
Serial port /dev/cu.usbserial-0001:
Connecting......................................
/dev/cu.usbserial-0001 failed to connect: Failed to connect to Espressif device: No serial data received.
For troubleshooting steps visit: https://docs.espressif.com/projects/esptool/en/latest/troubleshooting.html
I’m using the USB on/off switch that Digiblur recommended as well, and I can tell it does actually work and connect to the USB TTL device.
From HAAS and ESPHome Builder, when I go to add a device, I can see the CP2102 device show up, and disappear if I turn the switch off.
Maybe, but it is the one that Digiblur recommended in his post, as in directly linked to on Amazon, and used in his video.
I’m wondering if there might be a permissions issue with MacOS. The only windows machine I have is my work laptop, but they lock down the USB ports. I’m thinking of trying from Parallels or see if anything changes.
I doubt digiblur is some guru you should blindly follow.
75mA should be sufficient for powering esp32 board for flashing but there’s not much margin.
Unfortunately I have no experience with mac os.
And everything on the Internet is true and accurate?
I will say, at best, the 3V3 supply is marginal, so it might work some of the time, but probably not all of the time.
I use an independent 3V3 supply and connect to a powered USB hub whenever I program something that is not direct USB connected.
Your log says it didn’t get anything back from the device. That could be because you got TX and RX backwards. I know I have done that more than once. To prevent that, I usually use a terminal emulator to verify that the device can output data, since the bootloader always outputs something. When I see that I know that I have the wires the right way and the device is powered at least somewhat well.
Not sure about the Mac, but on Windows many of the drivers have been modified to not work of fake controller chips. Pretty much all of the cheap USB serial converters use fake chips. I used to prefer the CH340 based ones, because they were cheap enough that they were usually real. But it seems that is no longer the case and the newer versions of the drivers don’t work with many of my devices that should have real chips in them (since I bought them from reputable suppliers and they are not generic devices).
So, verify your computer can hear your Vue talk with your setup. Only then are you likely to have success flashing it.
No, of course not. Again, like I mentioned, every post I’ve found about integrating the Vue 3 (and for the Vue 2) seems to reference his work.
I’ve tried everything I can come up with. I’ve tried through Windows as a Parallels desktop from my Mac. I’ve installed the lastest drivers. I’ve tried running screen on the port to see if it reads anything, both on the vue 3 and a stock ESP32 board. I did manage to get some feedback on screen one time on the ESP32 board, but I haven’t been able to repeat it.
I think it’s possible I have a bad controller. I’ve put in for an RMA with Amazon and I’ll have a replacement tomorrow.
I don’t know anything about Vue3, but electrically you are in gray zone where things might work or not. If you draw more current from CP2102 than it can comfortably provide, the voltage drops. Then you have high contact resistance from your cheap probes causing more drop.
So use external 3.3V supply (with common gnd), clean the pads well and keep wires short.
Update. Got it working. Wasn’t the USB TTL device, lack of amperage, or MacOS. I think the board has some sort of coating on it that was preventing a good connection.
I had tried both non-invasive methods recommended. 1), a 3D printed template with leads through holes leading to the pads on the board and 2) a BDM Frame with probe pins. Both of these failed to make a good connection.
I tried a last ditch effort after the second USB adapter didn’t work, and I soldered leads directly to the pads. Not what I wanted to do, but it worked without issue.