Font corruption on GLK24064-25

Mike B · Post by **Mike B** » Mon Jun 17, 2002 8:29 am

Our customer is experiencing occasional font corruptions on some equipment that is built around one of your LCD units. These cause us major inconveniences when they occur as the unit has to be removed from the equipment and connected to a PC to reload the fonts.

Note, there are a number of these units in operation and they have been running for about one year.

The 3 instances reported were:

Character spacing on the display corrupted.
Failure of display, screen back lit but no characters output.
Screen blank.

What type of events are most likely to cause these font corruptions?

Note, a colleague has commented that the data output seems to gradually fade
with time (weeks as opposed to minutes!) after the fonts are first loaded.

Mike B

<font size=-1>[ This Message was edited by: Mike B on 2002-06-18 04:02 ]</font>

Post by **Cheese** » Wed Jun 19, 2002 1:56 am

Were these displays memory chip locked? http://www.matrixorbital.com/manuals/gl ... /index.htm - section 1.8? Might help...

r.

Mike B · Post by **Mike B** » Wed Jun 19, 2002 2:29 am

Thanks for the feedback.

I note that the version of the manual that was being used by my predecessors does not have the section on Memory Chip Lock Down.

Has this section been added specially to clear such faults?

Has anybody had experience of clearing such faults by doing this?

Also, has anybody got any comments on my colleague's observation that the display seems to fade with time?

<font size=-1>[ This Message was edited by: Mike B on 2002-06-19 02:57 ]</font>

Post by **Cheese** » Wed Jun 19, 2002 5:57 am

I've little idea about fading displays, I haven't used a big graphic display for long enough to comment.

In general locking helps because it is so easy for a small signal error to corrupt the memory on the display - it only has to mistake a character for #254 (&FE) then the following bytes will be taken as control codes which can do almost anything to the display - I can see this kind of corruption happening a few times a year easily.

Other things which might be the source of problems; how well regulated is the power supply? What temperature are the displays runs at? How much e/m noise are they exposed to?

I'm grabbing at straws, I don't really know enough to help you

r.

Post by **Henry** » Wed Jun 19, 2002 4:10 pm

Cheeseman covered most of the basics here... The memory chip lock down is there for units in the field. Can you please take a picture of the back of the display and post it?

Cammy · Post by **Cammy** » Wed Jun 19, 2002 4:39 pm

Hey Henry, can you do the same thing for a GLK12232-25 Display? I did'nt see anything in the manual about this.

Post by **Henry** » Wed Jun 19, 2002 5:03 pm

Yes you can, but it's much harder. The PCB does not have jumpers on it, it wasn't built to accomodate it... but it can be done, the 240x64 displays and 122x32 use the same memory chip, so you can follow the same procedure. New versions of the GLK12232-25 are coming in 6-8 weeks which will have this built on, just like the big guys do.

Mike B · Post by **Mike B** » Thu Jun 20, 2002 4:16 am

I'm afraid that I'm not in a position to be able to take a picture of the back of the unit at the moment but I can assure you that we have not incorporated the memory lock down.

Would you recommend that the memory lock down should be done for such units?

Post by **Henry** » Fri Jun 21, 2002 10:41 am

Yes we would. Especialy if development is done and the unit is in the field.

Mike B · Post by **Mike B** » Tue Jun 25, 2002 9:12 am

On inspection, it would appear that our unit is a GK 24064-25 rev. 1.01 (c) 1999. I can't seem to find a matching manual for this unit.
Is this the same unit as a GLK 24064-25?

Post by **Henry** » Tue Jun 25, 2002 11:00 am

Th eunit you have is a older one, it will not be able to do a memory lock down. You require PCB Rev 1.2 to do that at least. How old is this unit? from where?

Mike B · Post by **Mike B** » Wed Jun 26, 2002 3:12 am

I assume that the unit was bought in 1999 but I'm not sure where it was bought from.

We have been driving the unit with a higher message rate and have repeatedly managed to get display corruptions and then managed to corrupt the memory holding the font information. Would you expect this to happen?

My colleague has also observed that the display gradually dims during this process.
Do you have any idea as to why the contrast should dim?
Note: Our test program does NOT change the contrast.

Post by **Henry** » Thu Jun 27, 2002 1:57 pm

If there is too much data sent to the display, a buffer overrin can occure, over writing the memory. As for the contrast, I'm not sure there, can you discribe the component layout right about the RS232 connector?

Mike B · Post by **Mike B** » Mon Jul 01, 2002 9:38 am

It appears that the contrast on our display is lost if power is removed from the display. We can reset the contrast using mogd but this only lasts until the power is removed again.

Does this suggest that the unit has somehow lost its 'default' contrast? Could this default be overwritten by a buffer overflow? How can I reset the 'default', or set a value that is stored for use following power reset? Mogd does not appear to offer this facility.

Post by **Henry** » Tue Jul 02, 2002 10:15 am

If there is a buffer over run contrast settings can be over writen as well, just like anything else in the memory. THere is a contrast.rar files on the CD, extract it and run it to save a new value into memory. There is no way to "RESET" the module, just override.