Advise on hacking Epson LCD module

[rahandford]

Hello to alll,

I am trying to hack an LCD module form an old Epson printer (photograph attached).
I have identified a number of pins that are tied to either 5V or GND.
The remainder look like signal lines (3.3V LVTTL?), and there are about 3 of those.
There is one additional line that seems to drive a LED on the from panel, indicating a failure or paper jam.
The chip on the LCD module is E09A81AA, but I can't locate and datasheets on this chip.
I have also attached a copy of the signal analysis I have from starting the screen, but I am struggling to identify the protocol.
Any pointers of where to go from here would be much appreciated.

Thanks in advance for your assistance.

[stevenhoneyman]

I'd match the QFP/QFN package size up to one of these, and go from there:  http://erd.epson.com/index.php?option=com_docman&task=cat_view&gid=3&Itemid=435

Or maybe http://aitendo3.sakura.ne.jp/aitendo_data/product_img/lcd/tft/STM025QVT-001/STM025QVT001_S1D19122_rev0.8c.pdf

(your pic is too small to count the pins)
(also, info on the screen might help - color/b&w? physical size? resolution?)

[atferrari]

Hello to alll,

I am trying to hack an LCD module form an old Epson printer (photograph attached).
I have identified a number of pins that are tied to either 5V or GND.
The remainder look like signal lines (3.3V LVTTL?), and there are about 3 of those.
There is one additional line that seems to drive a LED on the from panel, indicating a failure or paper jam.
The chip on the LCD module is E09A81AA, but I can't locate and datasheets on this chip.
I have also attached a copy of the signal analysis I have from starting the screen, but I am struggling to identify the protocol.
Any pointers of where to go from here would be much appreciated.

Thanks in advance for your assistance.

The printer, is it available and able to be powered up? That could help a LOT.

Even if not, could you insert the board in place and check all Vcc and ground connections?

By mistake I've skipped most of your post.

Do you have access to the service manual (people usually say "yes" to this but almost all have just the user manual).

Read all technical data you can about the printer. That could help as well.

[rahandford]

The LCD display is a 2.5" TFT from an Epson Stylus printer.

I do have the full service manual but it only goes down to board level and unfortunately there are no schematics.

The E09A81AA chip on the back of the LCD is a 32 pin QFP. I can find no references to any datasheets via Google. If we can get a good idea of what the data lines are and what the protocol is I could then start throwing some code at it from a micro-controller or FPGA.

The screen shots that I included are all that I have so far, and I guess that they are generating the screen that is shown at startup, which is an error message and a graphic.

Any help or ideas on how to proceed from here would be very much appreciated.

Thanks to all that  have replied so far.

[stevenhoneyman]

A quick search says the pinout of the 9 pin ffc might be:

1. +V
2. +V
3. GND
4. SDATA1
5. SDATA2
6. PNLRST
7. SDATA0
8. SCLK
9. SCE


Can you get a higher resolution pic of it? Still kinda guessing blindly!

[amyk]

Google "PNLRST" and you'll find what looks like plenty of others puzzling over the same LCM. A 3-bit serial interface is unusual and doesn't look like anything I've seen before.

Seeing as you have the board which can drive it, for methods to reverse-engineer the protocol used.

[atferrari]

For the mechanical part, also in an Epson printer, I used with good results the auto test, activated by the user through the control panel; could you? Otherwise I would look into the manual for any "service" test.

Channels 2 and 3 in your analyzer seem to be "I am ready" (LCD) and "here we go" (micro).

If that comes true, you could manage to disconnect them, one at a time and you eventually could find out who's who.

Maybe a stupid question: what the error message says?

[scruffy.mogwai]

I don't have much of an idea about this but I managed to find a service manual for a CX9300F (which I just pulled apart) with REV. B schematics which include the 9 pin connector on the lcd, which should be labelled CN2. Looks exactly like the photo.
Most pins are just a direct connection through from CN1 on one of the front panel boards (with all the buttons). I will include labels from both connectors because it may help decipher the function of each pin.

CN1 PIN 1: 5V_VCC / CN2 PIN 8: VD5
CN1 PIN 3: SCLK / CN2 PIN 5: CLK
CN1 PIN 4: PNLRST / CN2 PIN 3: XRST (OVERLINED - Active low??)
CN1 PIN 2: SCE / CN2 PIN 7: A0
CN1 PIN 5: SDATA0 / CN2 PIN 2: DATA0

CN2 PIN 1: VD3 (connects to 3.3V supply)
CN2 PIN 9,6,4: GND

HTH.

[macboy]

My advice is don't bother. Small color TFTs with datasheets and lots of public sample code are very cheap.

[rahandford]

Hi scruffy.mogwai

That looks spot on!
Thanks very much, will have to have another look as soon as I get time.

Regards,

Rupert

[phooky]

I know this is a very old thread, but it was the only thing that came up when I was initially looking for resources. I recently got a somewhat similar panel from an oldish Epson inkjet working, so I figured I should share.
https://github.com/phooky/EpsonCP

In particular, the FFC pinout and a rough description of the protocol:
https://github.com/phooky/EpsonCP/blob/main/LCD-reversing.md

The panel takes a series of three byte long commands to configure itself, turn on the backlight, and accept image data. The image data is one byte per channel and is streamed in on three serial pins, one for each color. I'm just replaying the configuration commands that I sniffed from my panel; yours may be different. Since this is Epson silicon for an Epson product, the datasheet is probably a series of post-its in someone's cubicle, but some of the commands (turning on backlight, setting the resolution) seem reasonable.

Please respond through that github repo if you need to; this is a throwaway account that I don't plan to use. -a

[amyk]

That part number doesn't reveal any datasheets but Epson has a bunch of other LCD controllers with public documentation, which could be close if not an exact match --- they all tend to be very similar and their instruction sets can be separated out into large families.

Thanks for doing the work!

[Yusff]

Had a printer (RX610) with the same display.
Pinout: 1 Gnd, 2 +5V, 3 CE, 4 Gnd, 5 Clk, 6 Gnd, 7 Rst, 8 Data, 9 +3.3V (numeration on motherboard and LCD are reversed).
LCD has 128x160 resolution, interface: 9-bit SPI (3.3V level, 12 MHz), data sending in two words, format is RGB565 (RRRRR0GGG GGGBBBBB0). LCD controller and commands description is unknown, initialization routine and picture loading is captured from the working printer. Here is the simple code for STM8 (SPI through the synchronous UART):