Optrex F-51553 Graphic LCD: How fragile are they?

Charlie U · Posted: Sun Dec 26, 2004 9:34 am

Here's the problem. It's off the main purpose of this forum, but I am using the ICD-U40 with 3.212 PCWH. I think I may have damaged my glcd and I was wondering if anyone out there has experience with this type of display and can confirm that I have toasted the poor little thing.

I am using a processor board that we designed for test fixtures, etc. It normally uses 2 or 4 line character lcds. We had an adapter board built with all of the required circuitry for interfacing the glcd to the board. All of the control lines for the glcd come directly from the processor. I started the project with the ICE2000 emulator and got the display up and running with a modified version of glcd driver that was posted here (thanks to all who post here!!!). I was able to display characters anywhere on the display and with various sizes. I even have a picture, so I know that I am not dreaming!

I wanted to demo it for my customer, so I shutdown the whole mess as normal, pulled out the emulator pod and installed a new processor (PIC18F452 PLCC). I then plugged in my trusty ICDU40 and attempted to program the processor. Here we had a bit of trouble, because I had changed a few things since the last time that I had used the ICDU40. Once I sorted that out, the device was programmed, the board was turned off, the ICD was unplugged, and the board repowered. Then, nothing. Since that time, I have done everything to try to get data on the display, including changing back to the emulator and running the same code. The display just sits there. I have looked at the voltages generated by the display, and they usually appear to be correct, but occasionally are too low.

My question is "Could I have damaged the display be leaving it powered up with the processor unprogrammed and/or in reset?" Here is why I ask, I was reading the data sheet and found this little gem:

Will Reeve · Joined: 30 Oct 2003 Posts: 209 Location: Norfolk, England

Try the simple things first! Make sure your standalone processor is working! Get a scope of the osc. pins and also scope your LCD pins to make sure it is being sent something! I've had LCD display just die due to static (I believe) before, we had a batch of 1 x 8 displays which were very susceptible.

Will

Charlie U · Posted: Sun Dec 26, 2004 10:29 am

Thanks Will for the reply!

I have gone through and verified that the processor is running and that the signals are being generated. The frustrating part is that I go back to the configuration that worked originally, and it doesn't do anything. I have stepped through the ininitalization routine ( I know it should be done within 5 ms of reset release, but I am desparate at this point), and at this point I don't remember exactly where, the display does change state to all pixels on, but then nothing. If someone has ideas, I'll try them.

drh · Posted: Sun Dec 26, 2004 12:07 pm

If the ICD40 pulls the MCLR line to +13V to cause the uP to enter the program mode, and I think it does, AND if a glcd pin was connected to the MCLR line when this occured, you may have damaged the display.
_________________
David

Charlie U · Posted: Sun Dec 26, 2004 1:03 pm

Good thought David, and thanks for the reply. Unfortunately, all of the pins on the display go to i/o pins on the processor. The MCLR of the processor is isolated from the RES of the display. This has made me think though. I'll recheck the hardware and see if there was an assembly problem. We have used this same hardware (different physical board) for several products without incident, but that doesn't mean that isn't a problem with this physical board. Thanks again.

Charlie

Will Reeve · Joined: 30 Oct 2003 Posts: 209 Location: Norfolk, England

When we had a dead display it would behave exactly the same, it would look like a normal power-up, all pixels lit (you could even adjust the contrast etc) but no digital communications were possible at all. It doesn’t look good!

Will

Charlie U · Posted: Sun Dec 26, 2004 2:29 pm

Thanks Will,

I'll have to wait until tomorrow, but I think my client has another one that I can try. We can also try my suspect one on a different working system, I hope. In the mean time, I'll add pullup/down resistors to all lines, as required, just as a precaution to help prevent this in the future.

Thanks again
Charlie

PCM programmer · Joined: 06 Sep 2003 Posts: 21708

My basic rule when building a prototype is to buy two of everything,
in case something goes wrong. Whenever I don't do this, I end up
kicking myself for it. I then have to do an expensive UPS Red delivery
from Digikey.

Charlie U · Posted: Sun Dec 26, 2004 4:21 pm

Thanks for the moral support PCM,

I, too, generally buy more than I need, but this one was a freeby from my client. They are just a few miles away, and I had such great, early success that I didn't grab another one. Anyway, I wish I had all of the money I spent on UPS Red service, I could pay for a mighty fine new PC with all of the bells and whistles, and more!?!?

Thanks again,
Charlie

Charlie U · Posted: Mon Dec 27, 2004 12:51 pm

@#$%^ ^%$#@ @#$%^ *&%^@#,

Can you say "broken solder joint"?????

I checked the continuity of all glcd io and everything was fine. It appears that I was pressing on a contact that opened as soon as I removed the meter probe. I can only surmise that I must of broken the joint when I removed the display adapter board to install the uC the first time.

I have resoldered the connection and the display appears to work fine. I suppose the one thing good about this exercise is that I now know alot more about these displays than I had ever imagined I would.

Thanks again to all who helped,
Charlie