LCD screen matrix not working as intended

[RBDGT]

Hey everyone,

I qualify as a beginner, but I've never set anything on fire or blown anything up, so I guess I'm okay. Right?

Anyway... The screen on my Kurzweil PC3K88 synthesizer died, it was one of those cathode tube affairs and I figured I would replace it with a LED-screen. After some research I orderded the following on AliExpress: https://nl.aliexpress.com/item/32813587013.html?gatewayAdapt=glo2nld&spm=a2g0o.order_list.0.0.21ef79d2BplA7A

I soldered a connector, made a ribbon cable, made sure the LED was effectively on, etc...

But as you can see from my attachement, there seems to be an issue with how things are presented on the screen. Nothing is missing, the screen is just skipping two pixels every time. Two blank pixels, six working pixels, etc... Needless to say this sucks.

(The datasheet says something about font choice 6*8 (H) or 8*8 (L) on pin 19, and it's H as it is, but maybe that has nothing to do with this issue.)

Long story short: Did I buy the wrong screen? Or is there some way to get this to work correctly?
If you need any more information to point me in the right direction, ask away.

Many thanks in advance

[jpanhalt]

Welcome to EEVBlog.

It looks like you have a graphical LCD.  Can you read the labels on the controllers?  If so, the datasheet might help.  Attached is an example of text with a ST7567(A) controller.  The whole screen is 64x132.

I would look at your font table.  There is probably autoadvance for the X-direction (columns), which might be contributing to the problem.  In any event,  I don't think it is a controller/LCD problem but rather a coding problem.

[RBDGT]

Hey, thank you very much for your answer.

I've included pictures of the modules' controllers and they seem to be the same. (rainbow cable is the newer one)
What I can't seem to get my head around is that the connection is basically the same as for the original module but this one has the screen turning out bad. To be clear, the screen on the older module still more or less works how it's supposed to, so the issue must be on the newer modules side (or a bad solder by me for all I know)

I don't know exactly what you mean by "font table". There's also no coding involved, I'm just connecting the module to the synthesizer motherboard in the way the original module was connected. I guess I could pull up or pull down some of the pins, but that's it. Hence my mentioning of pin 19. My guess is it's high and thus configured for 6*8 which, in my head (which is no reference btw), explains what I'm seeing. Now, to fix it is something else.

The datasheets for the controllers are beyond my capacities at the moment, but I'll sit down and go over them slowly. Try and understand how exactly the image on the lcd is built.

There's not much I could find on the internet concerning replacing the screen in this synth, so I'm kinda winging it. I just thought it would JUST WORK, since it was only a matter of getting it connected. Alas, no...
If anyone has any other ideas or tips, please shoot them my way.

[jpanhalt]

I looked up that Toshiba T69630 controller.  That's a neat controller.  It seems to allow OR, AND,  and XOR writes to the screen as well as single bit manipulations (really neat).  I only have experience with brands of GLCD controllers that don't allow that.  Wish I had it. I have to keep a mirror (sort of) and write entire bytes each time or read the screen with a parallel interface and do the logical operations in the MCU.  Unfortunately, I am not familiar with the Toshiba.  The "font table" is described as being in ROM (maybe a separate chip -- not sure). 

There is code needed to manipulate the screen.  I didn't realize you were doing a like-on-like plug-in replacement.  If that's the case, then it is unlikely code is the issue.  Sorry I couldn't be more help.

John

[free_electron]

i would not be surprised the old lcd was skipping column drivers ...
you are not missing columns ( you can line up the columns and have the correct text. )

[RBDGT]

Yes, you're right, that's what I noticed too. Nothing is actually missing, it just skips two pixels after every six pixels.
The screen also responds to using the controlls on the synth as expected, it's just that column issue.

Any ideas? 

[RBDGT]

I just wanted to follow up on this thread because I managed to solve my problem.

After measuring the connections on the controllerchip itself I realized some pins were connected to vss and others to vdd where it should be the other way around. I then took a close look at the board and noticed tiny bridges connecting stuff. I removed one (for the font selection) and switched the ones for screen size round and BAM my screen works as intended.

For full reference (and because there's not a lot to be found about this specific repair online), this was a 240*64 screen ordered via AliExpress for €22 to replace a bad screen in a Kurzweil PC3K8. I'm guessing that the expensive (€100 and up) ones you can buy from "sellers" already have the bridges configured correctly. It took me a while to figure out, because I thought "it would just work". It doesn't, it requires a little extra work, but it's doable.