eInk 1.54" refresh

EL Wire/Tape/Panels, LEDs, pixels and strips, LCDs and TFTs, etc products from Adafruit

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
Locked
User avatar
PBPZ22
 
Posts: 2
Joined: Fri Mar 03, 2023 9:21 am

eInk 1.54" refresh

Post by PBPZ22 »

Hello,
started to work with the monochrom 200x200 eInk display. In the description I saw the remark

Do not update more than once every 180 seconds or you may permanently damage the display

If I load the Adruino Excample Adafruit EPD ThinkInk_mono onto the boardthe display is refreshed nearlly every 5 seconds (Frist text, then shapes, etc.). Is that a problem of the Excample, or are I'm missinterpretating the remark to not update once every 180 seconds. My understanding would be that bewteen every change of whats shown on the display I would at least have to wait 180 seconds.

Could you please go a little more into detail?

Regards
Patrick

User avatar
adafruit_support_mike
 
Posts: 67446
Joined: Thu Feb 11, 2010 2:51 pm

Re: eInk 1.54" refresh

Post by adafruit_support_mike »

PBPZ22 wrote: Fri Mar 03, 2023 9:28 am the display is refreshed nearlly every 5 seconds (Frist text, then shapes, etc.). Is that a problem of the Excample
The 5 second delay in the example is much less than the 180 second delay you should use long-term. If you left the example running for several days, there's a good chance it would damage the display. We had to make a tradeoff between what's best for the display over the long term and making people wait 3 minutes to see the display work the first time they use it.
PBPZ22 wrote: Fri Mar 03, 2023 9:28 am Could you please go a little more into detail?
epaper displays contain microscopic spheres floating in oil, each in a spherical hole slightly larger than the sphere itself. The spheres are mixed into a gel that gets cast into a flat sheet, then is exposed to oil. The gel absorbs some of the oil, and in the process, it expands. The expansion pulls the gel away from the surface of the spheres and fills the space around each sphere with oil. Once they're fully separated from the gel, the spheres can rotate freely in their little bubbles of oil.

The spheres are made from two different kinds of plastic, one white, and one with a color. The plastics are 'electrets', which means they have a built-in electric charge.. for the sake of argument, let's say the white side holds a positive charge. That means the white side of the sphere will rotate to point toward the negative pole of any external electric field, and will also move toward the negative pole if possible.

epaper displays take advantage of both kinds of motion.

You can make a sphere rotate so its white side points toward the front or back of the display by putting the appropriate electric field across the sheet. Continuing the assumption above, negative voltage on the front of the screen will make the sphere rotate to show its white side. Positive voltage on the front of the screen will make the sphere rotate to show its color side.

Once the sphere has rotated toward its correct orientation in the electric field, it will move toward the edge of the hole it floats in. The oil makes a barrier between the sphere and the hole, but eventually the oil film will get thin enough for Van der Walls forces to make a point on the sphere stick to a point on the surrounding hole.

That's how epaper displays can hold the same image for months without power: the Van der Walls linkage is strong enough to keep the sphere from rotating to another orientation.

That adds another step to the update process: when you apply an opposing electric field across the display, it pushes the sphere away from the edge of the hole hard enough to break the Van der Walls connection and leaves the sphere free to rotate again.

The word 'opposing' is important though. If you apply a field that pulls the sphere even closer to the edge of the hole, it makes the Van der Walls connection stronger. If you keep doing that long enough, the Van der Walls connection will eventually be too strong for the external electric field to break.

At that point the sphere is stuck, and the display looks like the previous image has burned in.

The only way to guarantee reversal of all the Van der Walls connections is to follow every image with its inverse. Any bitmap update that doesn't invert the previous image has pixels the same color as the ones in the previous bitmap. Those contribute to making the Van der Walls connections stronger.

Over time, the pixels that aren't balanced 50-50 between black and white will eventually get stuck.

Timing also plays a role though.

The Van der Walls connection between the sphere and the edge of the hole is strongest immediately after you apply the external electric field. Thermal vibration will gradually push the sphere away from the edge of the hole, making the Van der Walls connection weaker. That's why epaper can hold an image for months, but the image will decay eventually.

Letting the display sit for 180 seconds between updates gives the Van der Walls connections time to weaken a bit. The strength they lose is a little more than the strength they gain from a pixel being updated to the same color again, so you don't have to invert the bitmap between updates to keep the spheres from getting stuck.


Making the Van der Walls connections strong enough to create stuck spheres is a long-term process. A few fast updates now and then are unlikely to do any lasting damage. Doing fast updates continuously for days or weeks will eventually give you lots of stuck spheres.

User avatar
PBPZ22
 
Posts: 2
Joined: Fri Mar 03, 2023 9:21 am

Re: eInk 1.54" refresh

Post by PBPZ22 »

Okay thanks for explaining. As I am useing this display to static status messages, then I should be fine even if the status changed from time to time in <180 seconds.

Another thing I realized is, that sometime I can still see very slightly the prior message in the display. This effect vanishes if i use the message twice completly clearing the display and then printing the message. So if that is done directly in <180 seconds I still should be fine, right?

User avatar
adafruit_support_mike
 
Posts: 67446
Joined: Thu Feb 11, 2010 2:51 pm

Re: eInk 1.54" refresh

Post by adafruit_support_mike »

Ghosting would be the result of pixels not rotating fully to the new position. A double update will make the pixels rotate more completely.

Don't do both updates quickly though.. that increases the chance of getting stuck pixels. Ideally you'd do the second update at 180 seconds. If you want to do fast updates, set the whole screen red, then black, then white, then draw the next image.

Locked
Please be positive and constructive with your questions and comments.

Return to “Glowy things (LCD, LED, TFT, EL) purchased at Adafruit”