Hi there,

I noticed that the 6" e-paper displays are getting really affordable (25-35 Euro).

Datasheet /media/uploads/GerritPathuis/lb060s01-rd02.pdf

Unfortunately they need a positive and negative power supply, there are many interface connections and the thin copper foil is a pain.

The questions are is e-paper display worth pursuing and or is there a affordable quick fix for the given issues.

Thanks,

Gerrit

Is the interface also specced somewhere? So which data you need to send to it?

In general the idea I get is that it is far from impossible to do it yourself, but also not exactly trivial. You will need a bunch of DC-DC converters which start up in the correct order. In the end the Chinese will most likely start to offer cheap interface boards, but the question is how long that is going to take.

Out of interest I looked a bit deeper into it, to see how much work it would be if you would want to make some kind of interface PCB. But without having looked extremely deep into it, there doesn't seem to be a user manual (with the digital protocol) available. So then you would either need to check if it is available for another display with same connections (and assume it is equal), or you would need to buy a complete e-reader (since these displays are from e-readers), and monitor the data traffic.

I agree, this seems to be a dead end. At this moment in time there is not sufficient information available to make this work.

There are some smaller but more user friendly e-Ink modules available with documented interfaces and example software see here. Embedded Artist had some available that came with mbed example code. Seems to be sold out now.

I am running an 2.1" EPD based project on mbed now, which covers firmware of mbed API, software of Python/Android, hardware shields, 3D enclosure design, as well as application guideline for QRcode, barcode and font.

It may help you to find some useful information.

Because iPad kills E-book reader market, there are many 6" EPD panels available in the market as maintenance. In order to make it work, you can try to find a second hand device to start your development.

I have checked the Chinese market, SONY PRS-600 has used similiar panel (LB060S01) from LG display.

You have to contact LG display to sign NDA, get demo board to start your develop. But this product is out of date, obviously.

EPD's application is unique and limited. Since 6" is mainly designed for e-book reader, what else applications you are developing for ?

For LB060S01/02 panel, your processor must have a frame buffer of 800*600/2 = 240000 (234.375KB) . If using double buffering and RTOS and overhead of mbed, you need 512KB RAM and 1MB ROM.

So far only FRDM-K64F can cover this requirement. As I know, Kindle uses i.MX from Freescale. So this panel is only for Linux hackers, not for microcontroller developers.

I can find 2.1", 2.7", 6", 9" from different vendors. But I think smaller panel can find its market.

That it uses greyscale indeed increases your frame buffer alot, although if you want you could always do it in B/W. Also while your RAM point is true, I don't see why you would need 1MB ROM (sure external storage if you want to show pictures, but an external flash chip would do fine too).

Regarding mbed chips, the LPC4088 easily covers this, 512kB flash, another 8MB QSPI flash which can also run program code according to the description, and a 32MB RAM (including the external RAM), makes it have more than enough RAM.

I recommended 1MB ROM because people (usually clients) prefer loading their logo as splash screen. Additionally, bigger RAM micro always comes with bigger internal flash.

You are right, we can use external memory, anyway. Especially for ROM images and etc.

Actually my project was successfully running on MSP430F+FM24CL256. because MSP430 has very limited RAM, I was enforced to use external I2C FeRAM as framebuffer. The performance is good since EPD is slow enough, but I don't prefer external frame buffer, since it increases software development effort. Unless there is a commercial requirement.

I haven't recommended LPC4088 because its cost, since LPC4088QSB + Base = 80USD. Even expensive than other Linux boards (about 30USD?). It would be great if NXP can send me one for evaluation.

With 8MB QSPI and 32MB DRAM, it should be able to run embedded Linux. (just like uClinux on LPC1788)

Happy to see your feedback, by the way.

Yours sincerely

Allan

Yep the 4088 should also be able to run embedded Linux, but then it comes down at a choice what you exactly want it to do, if you want to run on the bare metal, with an RTOS, or embedded linux. Cost wise it then also depends on your overal setup, it is a quite expensive chip though, considering it goes for 13 euros at farnell for just the chip.

I think ideally for a simple microcontroller project you would have an interface PCB with not only the generation of supply voltages, but also with a microcontroller which has fast access to a memory IC as framebuffer, and then your host microcontroller can just send commands to the interface microcontroller, similar to how you can send just for example a logo to the average SPI LCD controller, and that is added to the screen, without requiring a complete frame buffer on your host uC. But for now that isn't available.

We can probably agree that direct interfacing with such a screen currently will give quite some issues with your random microcontroller.