Gap detection permanently disabled, not recovered by reset/recalibration

[ldericher]

I recently got my hands on one of these devices and got really excited by the prospect of using it without the smartphone app. Thanks!

First ran into issue #4 on my system (fixed that by adding "ControllerMode = le" under [General] in /etc/bluetooth/main.conf and restarting bluetooth), then could connect using Chrome, after manually setting the flags as described here: https://www.espruino.com/Quick+Start+BLE#linux.

Everything went great, until I fooled around for a bit (tried some different settings in the web UI) and somehow, now it doesn't work great anymore 😅

My device now seems to completely skip gap detection, instead just prints with no regard to position, and then spews out 5 or 6 empty labels.
I suspect it broke when I printed one using the "printer: D110M_V4" setting.

What I tried to no avail: Using the android app, now with the same issue. Pressing/Holding the reset button for a while, no change. Even the double-tap default print doesn't work properly anymore.

Maybe re-flashing the firmware might help? I can't seem to find any firmware images on the web though. Can @RenateUSB help? I've read their efforts in issue #7. If you've got something you can't send here: my contact info is on my website, found in my profile.

Maybe using the "D110M_V4" setting just f'd some "hidden" setting that can easily be reverted?

[ldericher]

btw, this is what happens now:

% uv run fichero text "Hello World"
Printing "Hello World"...
  Image: 96x240, 2880 bytes, 1 copies
  WARNING: no OK/0xAA from stop command
Done.

[RenateUSB]

I don't know.
I would have thought that maybe you put it into "continuous" mode.
I tried continuous mode (on 30 mm stickers) and it didn't seem to make a difference.

But maybe you switched it into some compatibility mode?

I would have thought that the reset button would do something, but I guess not.
It's probably a normal reset and my Fichero does that every time I unplug it (since no battery).

Are you getting a timeout?
After the "Printing..." how long does it pause?
Maybe the batch stuff is out of whack, I've never played with that.

I always got an aa if I didn't end the image transfer with the form-feed/stop.
If I used dot-feed/stop it would execute correctly but give me aa.
I think testing continuous mode with gapped stickers gets all wacky anyway.
I've got some 15 mm x 7 meters on order.

[RenateUSB]

You can try doing a power down reset. You'll have to pull the battery.
The front cover is held on with six tabs, two on each side and two on the bottom.
The front cover is the outside overlapping piece on the joint.
You want to try with a thin knife going from over the back and under the front.

[ldericher]

You can try doing a power down reset.

Done, still no progress 😢 However I noticed the gap detector connection seemed crooked. Realigned, also to no avail 🤷

I would have thought that maybe you put it into "continuous" mode. I tried continuous mode (on 30 mm stickers) and it didn't seem to make a difference.

But maybe you switched it into some compatibility mode?

I'm just as clueless. I'd also have thought that using the "double tap" for a default print would work regardless, but even that's borked.

Anyway. Any dice on getting a firmware dump? 😁

[RenateUSB]

The only datasheet I've seen for the processor is 80 pages of Chinese.
I don't know if factory program goes over USB or UART.
Yes, with a working system it goes over USB.

In short, nope, I don't know how to extract the firmware on this.
There are a couple of test points, especially T10.
It might put it in some other download mode.

[as0332162-cyber]

Make a website of different stickers

[CodeAsm]

I just bricked mine too... prints waaay to long and seems to not read the label ends anymore. maybe the build in label size is a buffer overblown or whatever, send tehm an email asking if there is a reset method.

opened my printer thanks to @RenateUSB and replugged the battery in and out, as if you presss the reset button. I guess its only powercycling the device. spottted testpads, intresting to test that with a serial con someday. the usb does not to appear at any point. when all but the battery is connected, the board flashes the leds (forgot which color).

Ill await customer support their email and when I got time, ill look at the serial port, and trace the usb. the usb seems to go near the battery charger circuit but also some traces go to the main ic. no external flash btw for those who havent had a look. only a battery charge ic, a steppermotor driver and a SoC thats like doing bluetooth and all the io aswell.

[RenateUSB]

It could be that the end-of-sticker detector has failed.
The small flex PCB in the upper left is for that.
You can try reseating it.

If you do a "form feed" with continuous stickers loaded it will spool out about 4" (100 mm) then give up.
You are using the stock 30 mm stickers?
Does this sound like your scenario?

[CodeAsm]

I saw the small flex pcb, and I had everything disconnected and reconnected, same behavior. I might try dig into the top half or the bottom of the paperrol holder and see if i can reach the sensor and see whats going on there (reseat if possible again). In the meantime I got a response from Fichero (was Dutch, used google translate):

It is unfortunate to hear that your label printer is currently not maintaining the correct size. From your description, we gather that the printer no longer recognizes the gap between the labels, causing it to keep rolling.

You can usually easily resolve this by recalibrating the printer. To do so, follow these steps:

1. Manual calibration (Sensor reset)

Ensure the printer is turned on.

Open the cover and ensure the label roller is taut. Let the first label protrude slightly from the opening and click the cover shut firmly.

Press and hold the power button (approximately 3-5 seconds) until the light flashes or you hear a beep. Then release the button.

The printer will now feed one or two labels through to determine the correct length and should stop exactly at the tear line.

2. Check settings in the app

Open the app and check that the paper size is set to exactly 30 x 14 mm. Check that the paper type is set to "Label" or "Gap", and not "Continuous".

3. Cleaning the sensor

If the printer still keeps rolling, there may be some dust or glue residue on the sensor. You can carefully clean the inside (the groove over which the paper runs) with a dry cotton swab.

These dint work and they now ask if I can record a video of the issue and send that to them. also they want model number and receipt... i .... sadly misplaced that particular receipt so hope they still wanna help. if that ends their help, ill try ask the store them selves or worse... get some coin and buy another printer and dump its firmware (if possible. might try this one to see if the serial port allows some trickery. havent looked at the main IC if there are debugging tools or something.

the main IC seems to be a YiCHip YC3121 (BT). the https://jlcpcb.com/partdetail/YICHIP-YC3121D/C2916800 seems to indicate it does have USB + and -. and serial TX and RX. ome OTP registers tho... datasheet is chinese tho, ugh. ow well, none breaking attempts to dump whatever i wonder.

I took a video, need to write that email still, probably monday they can reply, busy weekend and week ahead tho. bought it for fun, so I got no worries. but this "soft brick" isnt cool

[RenateUSB]

Hey! That was interesting.
I hadn't really dived into the gap detector.

The twin black little rectangles on the body are IR LEDs.
If you use your phone's camera in a dark place you'll see a red/purple glow from one.
The round photodiode in the cap makes its way back over the big flex.
It probably only "twiddles" between the two IR LEDs when it's actually feeding.

If the cap is open you should get blinking status LED red.
The complete opposite case is the cap closed with no stickers should also get blinking status LED red.

There is T16 test point next to the big flex J4.
Measure the voltage with a DVM, black on the USB connector body, red on T16.
Your values should be somewhere here:

0.0 cap open
0.1 cap closed, sticker in position
0.3 cap closed, carrier tape in position
0.9 cap closed, nothing in the way

The photodiode is not sensitive to even bright (non-IR) light..

[CodeAsm]

0.010 Open cap, no sticker.
0.079 Closed cap, Sticker over sensor
0.392 Carrier tape (no sticker)
0.905 empty printer, closed cap

first i was a bit confused where to find T16, the big flex as you said. my DVM meter claims to do 5 digits (4 after the dot). those move to much for my taste, but within margin

the IR led to the carrier tape (middle, center) did light up.
The values did appear to change when I exposed the opened top cap to either a lamp or sunlight.

There have been soms delays due to study and being me, sorry bout that
And 2 images to give a clue for those in the future

[RenateUSB]

Those voltages all look correct.
I'm thinking that you have it in continuous mode somehow.
Are you using the app?
Did you try my Windows software (you have to put stuff in the registry).

[CodeAsm]

Those voltages all look correct. I'm thinking that you have it in continuous mode somehow. Are you using the app? Did you try my Windows software (you have to put stuff in the registry).

I used https://0xmh.github.io/fichero-printer/ first, and may have clicked some option maybe that might have triggered it. After this, i used the official app from the app store (google, android) but no option over there offered a fix.

Where can I find your windows software and can this maybe toggle the continuous mode? I may be able to get a windows machine with BT going, im mostly a Linux user.

[RenateUSB]

https://www.temblast.com/fichero.htm
This works with USB, not BT. Read the instructions.
fichero /v