Missing line on dot matrix display

[AndreyS]

I have the same problem as on this image of guitar multi-effect processor.
There are 2 missing lines, I tried to apply force on both sides of the display in various places: I took thin pliers, sticked a paper on both arms, put the display between the pliers and gently squeezed them hard enough, that display crystals flowed. And lines never appeared. Then I took apart the display, unsticked rubber bands from the pcb (but left them sticked on the glass display) and put it together. And it didn't help neither.


My question is. Are there dead lines of pixels or is it display multiplexer's issue? Or is there still a hope?

[Quarlo Klobrigney]

Those sticky rubber bands are known as zebra strips. https://en.wikipedia.org/wiki/Elastomeric_connector
You need a tight weave cleaning cloth or tight cleaning swab in iso alcohol.
Clean everything, board and strips.
Do not touch the edges (dots or lines) of the strips after cleaning.
Reassemble. Fixes 99% of the problems.

Leave no residue. Check with magnification for fine or hairs debris.

[AndreyS]

Those sticky rubber bands are known as zebra strips. https://en.wikipedia.org/wiki/Elastomeric_connector
You need a tight weave cleaning cloth or tight cleaning swab in iso alcohol.
Clean everything, board and strips.
Do not touch the edges (dots or lines) of the strips after cleaning.
Reassemble. Fixes 99% of the problems.

Leave no residue. Check with magnification for fine or hairs debris.

I'm a bit of afraid to make the situation even worse. What if after cleaning with 98% alcohol and microfibre cloth, the more lines will be missing?
Do you have experience with zebra strips cleaning?

Does anybody else have experience with zebra strips cleaning?

[Quarlo Klobrigney]

Yes, I have done it many times with success (just yesterday in fact). That doesn't mean you may not make a mistake.
I ask what is the display number in case of accident? Is it a number you can get?
If you feel apprehensive, wait for others to chime in. I think you'll find you can do it.

[AndreyS]

Yes, I have done it many times with success (just yesterday in fact). That doesn't mean you may not make a mistake.
I ask what is the display number in case of accident? Is it a number you can get?
If you feel apprehensive, wait for others to chime in. I think you'll find you can do it.

So I took apart the display again and cleaned zebra strips on both sides, and also cleaned pads on glass display and gold plated pads on PCB.
On first try it only got worse (more missing vertical lines including those two).
On second try it returned back as it was (two missing lines).

On first try, when there were ~6 missing lines, when I squeeze the display four of them were appearing. But those two from original post did not.

[AndreyS]

Yes, I have done it many times with success (just yesterday in fact). That doesn't mean you may not make a mistake.
I ask what is the display number in case of accident? Is it a number you can get?
If you feel apprehensive, wait for others to chime in. I think you'll find you can do it.

after futher googling I found this topic. Quote:

I use denatured alcohol and cotton swabs. I clean both the zebra strip and the glass. The glass seems to collect a film from something that leeches from the elastomer, and it requires some effort to clean, with a generous amount of alcohol. I use a magnifier to make sure there are no cotton fibers or film left, and I'm careful not to touch the contact surfaces. Reassemble with everything in the same position it was before.

I took my IPA, cotton swabs. Prepared for cleaning. Finished cleanning. Turn on and nothing got fixed. Good old two vertical lines are still missing.
I think I give up.

[AndreyS]

Also worth to mention. I flipped all zebra connector, so that if the problem laid in zebra connectors, missing lines were in different places.
I cleaned hard gold plated pads 100% clean. And gently cleaned pads on glass, but still did it a lot. Everything didn't help

[AndreyS]

So what conclussion can be taken from this? Are IC damaged? or the problem is still somewhere in the pads?

[Quarlo Klobrigney]

From the link you provided:
"I have always thought that the problem was chemical breakdown reducing the elasticity of the connector, rather than a conductivity issue, contamination issues aside."

"If you disassemble these strips (by removing them from the display assembly) it makes no difference how you re-assemble them. The purpose of the strip is to carry data (digital signals) from the circuit board to the LCD display. Obviously, the signal from the PCB must be carried to the appropriate contact on the LCD display. The position of the connector strip makes no difference since it acts as a conductor with MANY POSSIBLE paths to the LCD. The conductors are spaced such that they cannot "short circuit" between signal paths. As a result, the proper connection is always made between the PCB and the DISPLAY."

In all honesty, I have been using this stuff to clean PCB's and yes the zebra strip 2 days ago https://www.walmart.com/ip/CRC-Industries-05110-Mass-Air-Flow-Sensor-Cleaner-Wt-11-oz/39950006 "Plastic safe-dries in seconds" I'm not saying to purchase this, only that I had some (a lot) left over getting my truck to run. Not cheap BTW, So I'm  using it up. It has zero residue.

You were wise to flip the strips. I'd do them like a 4 tire rotation on a car. It kind of assures that the wear is shifted about.

The conclusion is that either one of three things exist:
1: still some film dirt etc.
2: weak zebra strip
3: a damaged driver.
4: damaged glass

Can you tell me or show a picture of the LCD and part numbers? Maybe someone has seen it on the forums and can offer a replacement.

Quote from: AndreyS on Today at 05:06:05

So what conclusion can be taken from this? Are IC damaged? or the problem is still somewhere in the pads?

[AndreyS]

Can you tell me or show a picture of the LCD and part numbers? Maybe someone has seen it on the forums and can offer a replacement.

It's lgm122322.

If I use multimeter (in particular Ohm meter) on rows and columns, will the dot become black? I mean can I check with multimeter whether the glass display is faulty? And if it is safe not to damage the display?

[coromonadalix]

The problem will be the connector,  all the lgm122322  i've seen have an one row idc connector   some fiddling must be done, and connector pins spacing is not the same

[Quarlo Klobrigney]

Wow! They're not giving that thing away! http://www.lcmdisplays.com/lgm122322.html $50 and they are out of stock. Even ePay and Alibooboo are up there in price.
Perhaps an email to Digitech or lcmdisplays explaining your problem, and if there is something else that will substitute it.

Regarding testing, if your good with PIC, AVR or Arduino, you could make a tester and do it that way. I wouldn't use a multimeter on it though..

[AndreyS]

Regarding testing, if your good with PIC, AVR or Arduino, you could make a tester and do it that way. I wouldn't use a multimeter on it though..

I have some PIC and atmega microcontrollers. But it's kind of waste of time: prototyping that tester. Also there is a challenge in connecting the display to the microcontroller.

I just want to turn on any pixel on any of two missing lines to see whether it is driver's fault or indead some film residue on pads

[AndreyS]

Bare LCDs, as this appears to be, need to be handled very carefully. DO NOT just connect a battery between segments and see what you get. LCDs are damaged by DC. A second or two may be OK, but its better not to push it in the first place. Always put AC with no net DC on LCD pins.

You can create a appropriate AC signal quite easily. Use a function generator or a microcontroller or something to create a digital square wave of a few 100 Hz. Use a resistor divider if necessary to get this down to the 2-3 Volt peak to peak range, then put a capacitor in series with it to block any DC. A 100 nF ceramic should do fine. LCD segments are quite high impedance.

Now you probe around looking for pairs of pins that cause segments to be solidly actuated. This is a small LCD so you probably only have one or two commons. It will take some experimenting, but eventually you should be able to identify the commons, and then map out what segments are actuated when each of the other wires are driven for each common.

This sounds a lot more complicated than it is.

Ok. I got it. Just got to make square wave generator. Gotta try that

[AndreyS]

So I used built-in square wave generator in m DMM. It is 50Hz, about 1.5V peak to peak. And, unfortunately, I couldn't make any pixel to work. When I touched row pad and column pad, no current would flow. Maybe should I have make it using rubber zebra connectors, but again they should be pressed, which isn't that easy to do with bare hands.

I think I leave the display as it is, with two missing lines

[AndreyS]

The problem will be the connector,  all the lgm122322  i've seen have an one row idc connector   some fiddling must be done, and connector pins spacing is not the same

As for connectors. I tested connectivity of all connectors and a cable with multimeter. Problem isn't in them.

I think the problem lies in driver or the display.
I took appart, assembled it back and cleaned (cotton swab, microfibre cloth, IPA, ethanol) so many times, that I am pretty sure it isn't even zebra connectors' problem nor bad connectivity between zebra and pads.

[Bashstreet]

The problem will be the connector,  all the lgm122322  i've seen have an one row idc connector   some fiddling must be done, and connector pins spacing is not the same

As for connectors. I tested connectivity of all connectors and a cable with multimeter. Problem isn't in them.

I think the problem lies in driver or the display.
I took appart, assembled it back and cleaned (cotton swab, microfibre cloth, IPA, ethanol) so many times, that I am pretty sure it isn't even zebra connectors' problem nor bad connectivity between zebra and pads.

The display portray all the classic signs of poorly connected zebra connector or damage in the connector.
Make absolutely sure you wiggle and test the connection thoroughly.
I know you done this many times but as it is so classic mode of error it is worth trying once more. 

I personally have never experienced this error resulting from anything else than poor connectivity hopefully someone has and can give some further pointers what to look for.

Best of luck ! 

[mikkkejickles]

Hey this sounds very like a problem I had with my gameboy - this works. You probably just need to reflow on just one line.

[AndreyS]

Hey this sounds very like a problem I had with my gameboy - this works. You probably just need to reflow on just one line.

That gameboy doesn't have zebra connectors. You missed the point

[mikkkejickles]

Hey this sounds very like a problem I had with my gameboy - this works. You probably just need to reflow on just one line.

That gameboy doesn't have zebra connectors. You missed the point

Hey, Sorry, I'll look more carefully next time. I play guitar and fix effects, I didn't read it was Zebra.

Heres the datasheet for the LCD (i'm pretty sure its the same one).

https://www.pacificdisplay.com/gdm/GDM-12232-26.pdf

and Driver

https://www.pacificdisplay.com/ics_app%20notes/epson/SED1520-21.pdf

Sorry but I couldn't find a service manual for that unit.

[coromonadalix]

I have seen  display problems like the op in the past, not always the zebra connector,  the matrix chips / display driver can go bad too.  I dismantled the lcd glass and zebra strips and do an surgery on another display to repair it  ... the problem was in the ic blob on the pcb ...

[Bashstreet]

Hey this sounds very like a problem I had with my gameboy - this works. You probably just need to reflow on just one line.

That gameboy doesn't have zebra connectors. You missed the point

Yes he missed the point but still nice to thank for effort right ?

[AndreyS]

Hey this sounds very like a problem I had with my gameboy - this works. You probably just need to reflow on just one line.

That gameboy doesn't have zebra connectors. You missed the point

Yes he missed the point but still nice to thank for effort right ?

I partially agree. If everybody wouldn't read original posts and put theirs 2 cents, the forum would be a mess with almost no usefull information.
On other hand, I appreciate the effort

[mikkkejickles]

No worries, I should have read it properly.

Were the data sheets of any use? I hope you can get it fixed.

Hey this sounds very like a problem I had with my gameboy - this works. You probably just need to reflow on just one line.

That gameboy doesn't have zebra connectors. You missed the point

Yes he missed the point but still nice to thank for effort right ?

I partially agree. If everybody wouldn't read original posts and put theirs 2 cents, the forum would be a mess with almost no usefull information.
On other hand, I appreciate the effort