F5 lcd sobstitution

[Daxxin]

Okay now is clear Skippy , yes i have an oscilloscope , next saturday have lot of time then i check , just help me .. in a while i try to map what remain of lcd.

[Daxxin]

Skippy have scope and time what you suggest to check?

[Silver Dollar]

Here's A picture of a LCD we use at work and the schematic to it. It might help?

Segments


Schematic


Here's another one. I'm not sure if these are customized for our application?

Segments


Schematic

[Skippy]

With a scope, you should be able to see the serial data change, when the segments on the screen change. So if you set the machine up so it's picking up interference, and the 88 target ID value is going mad, this random change should be visible on the data coming fron the PIC. If you can identify these signals from the HC595's, then you have a starting point for trying a basic 88:88 LCD display.
I'm struggling to say what to look for. If I had it in front of me, it would be easier! Have you got an LCD to try on it? That would actually tell you more, initially, because it just displays black / silver, and doesn't change polarity every25 milliseconds (or whatever) like the digital signals do.
Your other way of controlling the display is using the adjustments. So if you vary the discrimination, you will change the 88 reading from 1 - 9, 10 - 65 (a guess). The fact that it displays 6 and not 06 could help sort out the digits, for example.
Really it's just a lot of educated trial and error. Once you have some display functions identified, the rest should be easer, for example if you set manual ground balance, those two 88 digits will change up/down.

[Daxxin]

Thanks Sinclairuser and Silver Dollar too , this the scheme .. what i draw ..hope there no errors not sure 100%

[Skippy]

Oh dear, that's really bad news. They have seemingly configured the internal tracking to make it simple, and then they sort out the mess in the microcontroller software..This is going to be harder to do by trial and error. I think some serious disassembly of the broken screen would help a lot.I've never cracked one open, but presumably there must be a number of tracks on the top glass sheet, which then make contact with tracks on the bottom sheet, plus the tracks that stay on the bottom glass. It is the crystals that make the dark areas, if they are cleaned out, it should be clear.
This also means that you cannot 'bodge' a solution using a standard 8888 display. Though you could use one to help decode any confusing/unknown problems during your reverse engineering.

[Daxxin]

very confused i am ..this the data coming to hc595 at pin 14 serial data in ..how to count the bits?

[Skippy]

You need to find the clock pin (pin 14). It looks like you can see the minimum clock period from your narrowest 1010 transitions.
I see that as:
1.....1001110111010110010.....0 but that is 18 bits, not 16 bits. Unless there is one data line, and all the 595's are cascaded, and fed with a 32-bit sequence. I had assumed they would have 2 data lines, and one clock, but that's a waste of a pin, really. And there will probably be one PIC pin used as the Latch contol, this will probably just be a short pulse every 20 millisecs or so.
Do a google (image) search for 74HC595 and PIC , you will find examples of circuits, it's a common technique to drive lots of things from a small number of PIC pins.
Here you go (Atmel, I know... but):
http://ba.protostack.com/2010/05/shi...ter_07_lrg.jpg
from:
http://www.protostack.com/blog/2010/...lling-16-leds/

Same thing, Arduino style:
http://www.geeetech.com/wiki/index.p...74HC595_module

[Daxxin]

wondering why have no 3-state waveform at 595 out Q's , example checking around the gnd balance ...outputs (lcd pins) controlling
have a square H or L waveform turning the knob suddendly they change from 30hz to 80hz nothing else , tought to found 3 state levels

[Skippy]

It's not normal to leave LCD pins 'floating', as they hold a charge, and will cause odd behaviour. If you don't want a segment black, just apply the same voltage to both electrodes.
I don't know why the 30 - 80 Hz change happens.

[Daxxin]

over 30 pins 27 goes to hc595 an with scope i see only 2 waveform changing frequency 30hz or 80hz by turning the knobs
first state
second state

over the 3 remaining pins i think reversing polarity to keep alive the lcd this the signal

[Skippy]

This means nothing to me, what/where are these 30 Hz waveforms?
Maybe these articles might help you understand what you are seeing:
http://www.pacificdisplay.com/lcd_multiplex_drive.htm
http://www.dkedisplay.com/knowledge/...plex_drive.htm