Unable to program the badge out of the box

[WUOTE]

Symptoms:

1. The display is stuck, the image does not change regardless of the programming method (USB, sound).
2. Powering the badge using a battery (3 new different batteries were tried) does not work reliably: the green LED either does not light up, or "blinking" as if the connection is intermittent. Connecting wires from the battery terminals to the badge's power pins seem to work (the green LED is constantly ON with constant brightness).

Things I've tried:

1. Installing and removing USB drivers: from pi, from arduino core, and using Zadig.
2. Flashing Micropython, CircuitPython, pimoroni's badger specific config using Thonny.
3. Changing USB cables and ports.
4. Flashing the board with MicroPython_RenameMainDotPy.uf2, flash_nuke.uf2, ggtag-v1.1.1.uf2, ggtag-v1.1.0.uf2, ggtag.uf2.
5. Changing the port's baude rate from 9600 to 115200.
6. Building the firmware from the sources and uploading it.
7. Building Frontend from the sources and trying to use it to program the device.
8. Using Chromium, Edge, Chrome to flash the device.

All the above were tried on Windows 11, Ubuntu 23, Ubuntu 24.

None of this worked. Is there anything else I can try? Is my badge defective?

[rgerganov]

2. Powering the badge using a battery (3 new different batteries were tried) does not work reliably: the green LED either does not light up, or "blinking" as if the connection is intermittent.

This problem has been reported several times, it seems the battery holder doesn't make good contact on some PCBs. We are trying to reproduce and provide a workaround.

2. Connecting wires from the battery terminals to the badge's power pins seem to work (the green LED is constantly ON with constant brightness).

Can you try sound programming and provide javascript logs when you connect wires and supply 3V?

I really don't know what is the issue with USB programming though. The javascript log looks fine, the data is successfully sent to the device. If same happens with sound, then maybe there's something wrong with the display.

[WUOTE]

Programming using sound from desktop PC, from a phone, and playing the demo video results in no change.

Console output:

main.js:424 Programming over sound
main.js:426 Encoded data length: 56
main.js:427 Encoded data: 0,54,0,80,82,23,35,46,205,155,208,119,223,203,102,64,29,168,161,114,50,236,217,189,7,125,252,182,100,0,184,2,23,35,46,205,155,208,119,223,203,102,64,20,31,161,114,50,236,217,189,7,125,252,182,100
ggwave.js:9 n = 98304, nActual = 98304
ggwave.js:9 n = 98304, nActual = 98304
ggwave.js:9 n = 98304, nActual = 98304
ggwave.js:9 n = 98304, nActual = 98304
main.js:473 All done

[gerenski]

Hi,

I have just tried this on 20+ PCBs trying to see why would it happen, but it didnt. If I see correctly - when you press the battery with the multimeter green LED comes on? If so - try to clean all- the PCB pad, battery and the metal battery holder. Try changing the battery too, just in case. This is a very simple schematic - not many things could be wrong.

As for the not programming issue - during the manufacturing process they are all individually tested both ways - USB and sound. What comes to my mind are two things: check and see if the cable is plugged in all the way into the connector and check and see if there is any damage on the cable itself. Finally - may be (although very small possibility) if you are able check and see of there is something wrong with the connector itself.

BR,
Hristo

[WUOTE]

I've cleaned the pad and the whole board using rubbing alcohol.
Adjusting the battery + contact tension allowed creating an electrically stable connection (the battery slips out if the badge is turned tho).
Programming with sound with both battery inserted directly and using extension wires (as shown on the photos above -- those are not multimeter probes) doesn't change the screen. Tested with three different brand new batteries each measuring 3V.

After spending 2+ days trying to fix the issue the only visual change that happened was the appearance of horizontal stripes across the screen during the initial programming attempt via USB.

I've detached the display's ribbon and inspected it and the connector for defects: no apparent visible damage (I did scratch the dark part of the connector -- the latch: I've assumed that it's a "pull" and not "flip" type). The badge itself was never dropped or mishandled.

[WUOTE]

What else I can try/test?
What are the USB drivers that should be used with the device, can you provide a link?
Am I doing the "start from zero" correctly: upload flash_nuke.uf2, upload ggtag-v1.1.1.uf2, try programming the badge?

[rgerganov]

What else I can try/test?

From the logs it appears that the software part is working and I cannot suggest other things to try for the hardware part. You can try to return your ggtag to CrowdSupply as defective.

What are the USB drivers that should be used with the device, can you provide a link?

The device is represented as USB serial and doesn't require any special drivers. We have tested on Windows, Linux and MacOS.

Am I doing the "start from zero" correctly: upload flash_nuke.uf2, upload ggtag-v1.1.1.uf2, try programming the badge?

I don't know what is flash_nuke.uf2 but it is not needed. You only need ggtag-v1.1.1.uf2

[WUOTE]

Update: the replacement item provided to me functions correctly,-- both sound and serial programming methods work. Thank you for your answers.