E-Paper display has problems under sunshine

Other Hardware Displays
1

Hello, everyone,

I use a 4.2" E-Paper display from WaveShare V2.1 for a multifunctional display in the marine sector. The project is described here:

https://open-boat-projects.org/en/diy-multifunktionsdisplay-obp-60/

As hardware I use:

  • GxGDEW042T2 (V2.1)
  • ESP32 Wroom
  • Library: zinggjm/GxEPD@3.1.0

To display the display content, I use full refreshes when changing pages and partial refreshes when updating display values. This works quite well as expected as long as I don't expose the display to direct sunlight. As soon as sunlight hits the display, the contrast of the black image content fades over time. In particular, after a full refresh when changing pages, display problems arise in the form of incorrectly displayed characters with a lack of contrast. Sometimes there are also pixelated images. Partial refreshes can partially compensate for the lack of contrast, but only affect the areas that are refreshed by the partial refresh. According to the data sheet, each display is subjected to a UV test of 168 hours at 40°C before delivery. Some sources online claim that e-paper displays cannot tolerate UV light. Theoretically, UV light can neutralize charge carriers. No improvement could be achieved in extensive tests with UV filters in front of the display (blocking filter: 250...450 nm). The problems can also be verified using a 100W halogen lamp as the sun. The spectrum roughly corresponds to that of sunlight. The cause of the problems could be determined. The IR component in sunlight is annoying and causes display problems with the contrast. If IR filters are used in front of the display, the display works correctly again without loss of contrast. Apparently, the IR light heats up the black parts of the display so much that the viscosity of the gel layer becomes so great that the charged color balls can approach and mix (that's my theory). The problems are more serious with older e-paper displays (< V2.0) than with newer models (> V2.1)

Here are some videos showing the problems:
https://www.youtube.com/channel/UC2jOz-8Ag1Ig2XpOJ2Td6ug/ (see the last 4 videos)

The data sheet contains settings for the voltages used to align the color balls (VDH, VDL, VCOMM). What influence do these tensions have on the contrast of the image content? As it looks, alternating voltages are used for the full refresh and static voltages for the partial refresh, depending on the image color. What influence does the timing of the voltages have on the contrast? It seems to me that this might solve the problem.

Norbert

2

@norbert-walter, Hi Norbert, welcome to the forum!

These e-paper panels are susceptible to UV light. There have been 1.54" panels with UV protection.
There was a post that reported the controller being the cause of this. But I don't know details.

It is important to call powerDown() with the library GxEPD, to switch off the charge pump for the panel driving voltages, before removing power supply, or before a longer idle time. This may help for this issue. GxEPD2 internally calls powerOff() after any full refresh.

There are different generations of the 4.2" b/w panels, with the same panel name. Different waveform timings are optimal with the different panels. With GxEPD2 I tried to adapt, and have provided parameters for the waveform tables that can be easily changed by users wanting to optimize on their own. Also available with GxGDEW042T2.cpp

Jean-Marc

3

Thank you for your quick comments. I will follow the hint with powerDown(). The problem could at least be eliminated with an IR sun protection film.

4

Hi Norbert,
Did you find any solution to the updating the epaper in direct sunlight?
What exactly is the IR protection? Does it really help?
I am using the Gxepd2 library which supposedly uses the powerOff() automatically but still the black pixels stay almost white if updating in direct sunlight...
The weird thing is that during the update when the whole screen flashes from black to white a few times the black pixels are perfectly black even in direct sunlight. Only in the very last moment of the update tge black pixels fade to almost white...

5

Hello Fred,
That's exactly how my display behaves. I have not been able to achieve any improvements in sunlight using software. The problem is the sun's IR radiation, which heats the gel layer under the black pixels, making the gel layer more viscous. Then the differently charged colored balls can attract each other. The black pixels then fade and sink into the gel layer. The only solution is to use an IR protection film like those used for windows. With that I solved the problem. However, this makes the display a little darker.

6

This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.