QLCD for NZ readers
- Thread starter rodmcm
- Start date
a Web Search indicates there are some issues with documentation
There is some discusion about documentation issues here;
http://www.lcdstudio.com/forum/search.php?search_author=tronix&sid=6b00960334c028230998b38695f31de2
and that references a data sheet here;
http://sub.chipdoc.ru/pdf/lcd/lcm/Agena/graf/ag1286401.pdf
Hope this helps.
There is some discusion about documentation issues here;
http://www.lcdstudio.com/forum/search.php?search_author=tronix&sid=6b00960334c028230998b38695f31de2
and that references a data sheet here;
http://sub.chipdoc.ru/pdf/lcd/lcm/Agena/graf/ag1286401.pdf
Hope this helps.
westaust55
Moderator
Agena GLCD
Just in case some further documentation from across the Tasman (and the Nullabor) may be helpful:
Just in case some further documentation from across the Tasman (and the Nullabor) may be helpful:
Attachments
-
48.8 KB Views: 61
-
157.1 KB Views: 59
Thanks for that, but I had found all of that. Still nothing works, tested all pins and read correct..... thats the second one that doesn't want to work for me, maybe the programme?
Attachments
-
5.2 KB Views: 33
westaust55
Moderator
Help from across the Tasman
Firstly, good to see a reasonable level of formatting and comments in the listing.
Now a wild but educated guess here . . . .
Reset pins are typically high and are pulled low for a reset. In your initial setup you pull the resetpin low and wait 30 milliseconds but do not raise the reset again. You might want to check this.
A schematic diagram might help folks understand and check the program better.
The Agena LCD also needs an operating voltage (Vo) for the actual LCD that might need to be slightly negative. This topic has arisen in another recent thread.
Edit:
as a side issue (won't help in the short term till the LCD is up and running), I note that these gLCD's have the ability to read from the onboard memory which would make the need for a separate buffer obsolete and drawing of intersecting lines far easier.
Firstly, good to see a reasonable level of formatting and comments in the listing.
Now a wild but educated guess here . . . .
Reset pins are typically high and are pulled low for a reset. In your initial setup you pull the resetpin low and wait 30 milliseconds but do not raise the reset again. You might want to check this.
You have two CS (chip select ?) lines. Which is for what ? ?' set intial conditions
LOW Resetpin
pause 30
HIGH resetpin ; IS THIS REQUIRED ? ? ?
gosub DisplayOn
main:
A schematic diagram might help folks understand and check the program better.
The Agena LCD also needs an operating voltage (Vo) for the actual LCD that might need to be slightly negative. This topic has arisen in another recent thread.
Edit:
as a side issue (won't help in the short term till the LCD is up and running), I note that these gLCD's have the ability to read from the onboard memory which would make the need for a separate buffer obsolete and drawing of intersecting lines far easier.
Last edited:
Here is the circuit - Will retry with the reset pin
Attachments
-
9.6 KB Views: 56
The QLCD is almost identical to the one shown on this wonderful demo link which shows the actions of CS1/CS2, thought the low/high is different according to the spec sheet.
I have done the usual with a test program and forcing
http://www.geocities.com/dinceraydin/djgfxlcdsim/djgfxlcdsim.html
I have done the usual with a test program and forcing
http://www.geocities.com/dinceraydin/djgfxlcdsim/djgfxlcdsim.html
Rod,
Not sure where this is at.
You indicate that you need some help but it's not clear (to me anyway) what's working and what's not - "nothing works" is as much as we have to go on.
Do you have any output at all, are you not getting what you expect where you expect it, or does it appear 'dead'?
You could do some real basic testing without the 74LS595N and setting some data lines directly from the 18X ports just to sort out basic functionality.
I know you are working with an 18X but would consider a 28X1, an MCP23017 16 port i/o expander and i2c as a potentially simpler coding exercise to provide the 13 connections required to drive the GLCD.
Not sure where this is at.
You indicate that you need some help but it's not clear (to me anyway) what's working and what's not - "nothing works" is as much as we have to go on.
Do you have any output at all, are you not getting what you expect where you expect it, or does it appear 'dead'?
You could do some real basic testing without the 74LS595N and setting some data lines directly from the 18X ports just to sort out basic functionality.
I know you are working with an 18X but would consider a 28X1, an MCP23017 16 port i/o expander and i2c as a potentially simpler coding exercise to provide the 13 connections required to drive the GLCD.
Nothing is wrong with the 18X or the 595, getting correct inputs to the QLCD as measured with voltmeter. Have done many test regimes to test outputs from 18X and 595 etc. So, either i have a fundamental programming error (timing, initialisation etc) when I try to put the whole program together as a package or this is two out of two QLCDs from Surplus Electronics that are dud. The problem is I cannot personally discover which item is wrong hence the call for help before I spend any more money.
I can assure you that asking for assistance is only when I have exhausted my own resources.
I can assure you that asking for assistance is only when I have exhausted my own resources.
westaust55
Moderator
Until the reset signal is sorted out you are potentially holding the LCD in a form of OFF state and will not see anything.
Based on the LCD I have been using, when the Vo is correct, and the screen initialised, you may see a slightly darker region for the active LCD display area.
Let us know what the state is when you have added the reset High command to your program and we can look further for you. Have printed out your schematic but do not have much time right now to delve further.
Based on the LCD I have been using, when the Vo is correct, and the screen initialised, you may see a slightly darker region for the active LCD display area.
Let us know what the state is when you have added the reset High command to your program and we can look further for you. Have printed out your schematic but do not have much time right now to delve further.
"Nothing is wrong with the 18X or the 595, getting correct inputs to the QLCD as measured with voltmeter. Have done many test regimes to test outputs from 18X and 595 etc. So, either i have a fundamental programming error (timing, initialisation etc) when I try to put the whole program together as a package"
Since you feel you may have 'duds' I presume that means there has never been any output of any sort.
That's why I'm trying to suggest a real back to basics bit of testing so you just try to turn it on and light a single pixel by as direct a bit of code that you can, once that is sorted, then you can add all the other stuff back a step at a time.
There is another connection sheet here - http://www.vishay.com/docs/37233/12g064a.pdf - shows sense of some of the control pins.
Since you feel you may have 'duds' I presume that means there has never been any output of any sort.
That's why I'm trying to suggest a real back to basics bit of testing so you just try to turn it on and light a single pixel by as direct a bit of code that you can, once that is sorted, then you can add all the other stuff back a step at a time.
There is another connection sheet here - http://www.vishay.com/docs/37233/12g064a.pdf - shows sense of some of the control pins.
Last edited:
westaust55
Moderator
Thats in line with the info posted in posts 2 and 3 and the Reset is low for a reset but will need to be made high before anything else will happen.
westaust55
Moderator
That great to hear.
hirzel@xtra.co.nz
New Member
Hello Rod
If you get it going I'd be interested in the circuit and other information as well
Alfred
If you get it going I'd be interested in the circuit and other information as well
Alfred
As searches on the AG1286401 return this PicAxeForum posting I wish to post information pertaining to the LCD supplied by SURPLUSTRONICS (Auckland, New Zealand)
My examples have the following markings for reference.
AG1286401 Ver.2
17 Sep 2001
There is no onboard negative voltage generator as IC5 and the surrounding capacitors have not been populated.
Note: Solder jumper JP4 is also in place. This connects pins 3 and 18 of the module.
i.e Vo and Vee are connected on the module via JP4.
So, besides the +5Volts (Pin2) and Ground (Pin1) needed to drive the controller portion of the module, a second voltage needs to be provided to drive the LCD and establish the desired contrast. With the module configured as described above (JP4 in place) this voltage can be connected to either Vo (Pin3) or Vee(Pin18). It would be better practice to connect to Vo(Pin3).
This voltage, which must be negative, should be connected via a variable resistor (preset/trimmer or potentiometer) so that a contrast control can be affected.
Experiments revealed that the best configuration is:
Potentiometer / variable resistor = 10K Ohm
-10V to -12V (Bottom of potentiometer). (EDIT: 26/12/08)
Ground (Top of potentiometer)
Measured at the pot wiper...at -9V all pixels just start to turn on at full contrast setting. The effect is quite subtle and there is not enough contrast for those pixels that should be turned on. So a -9V supply is insufficent with any image very pale. With -10V the background pixels are obviously turning on and this is the best contrast setting for an image.With -12V all pixels are distinctly dark.
Use of the 10K potentiometer provides a control that is not too sensitive. I found 20K was to sensitive to set the desired contrast. (EDIT: 26/12/08)
As this display needs a healthy amount of negative voltage, connecting the top of the potentiometer to Ground (0Volts), as opposed to +5V, also makes the contrast control less sensitive.
LED BACKLIGHTING
NOTE: This display has no current limiting resistors in place for the LED backlighting.
You should insert a series resistor to limit the current. The back light configuration has 76 LEDs. 38 paralleled sets of 2 LEDs in series. The datasheet indicates up to 390mA for the back light so this indicates 10mA per LED
When supplying the LED back light from +5V, a series resistor of 4.7 Ohms will give a LED current typically around 200mA and with a spread 150mA to 300ma going by the Forward Voltage Max/Min figures in the datasheet.
EDIT: 12/12/08 With +5V the following LED currents were measured.
4.7 Ohm series resistor gave 195mA (At a brightness level I preferred)
5.1 Ohm series resistor gave 186mA (Possibly a brightness level a little low)
EDIT 6/1/09
MISSING NEGATIVE VOLTAGE GENERATOR CIRCUITRY
For those interested the missing IC is an Epson SCI7661
R7 =1Meg, and C1,C2 and C7 would be 10uF 16V
Check the SCI7661 datasheet for the configuration needed to attain the -10Volts
As I write this there is a SCI7661 source on eBay US$9 plus freight of US$5 to NZ. Ex Hong Kong
The jumpers will need addressing. Un-solder JP4 and deduce what should be done with JP1 thru JP3 using the datasheet.
I hope you find this information useful.
My examples have the following markings for reference.
AG1286401 Ver.2
17 Sep 2001
There is no onboard negative voltage generator as IC5 and the surrounding capacitors have not been populated.
Note: Solder jumper JP4 is also in place. This connects pins 3 and 18 of the module.
i.e Vo and Vee are connected on the module via JP4.
So, besides the +5Volts (Pin2) and Ground (Pin1) needed to drive the controller portion of the module, a second voltage needs to be provided to drive the LCD and establish the desired contrast. With the module configured as described above (JP4 in place) this voltage can be connected to either Vo (Pin3) or Vee(Pin18). It would be better practice to connect to Vo(Pin3).
This voltage, which must be negative, should be connected via a variable resistor (preset/trimmer or potentiometer) so that a contrast control can be affected.
Experiments revealed that the best configuration is:
Potentiometer / variable resistor = 10K Ohm
-10V to -12V (Bottom of potentiometer). (EDIT: 26/12/08)
Ground (Top of potentiometer)
Measured at the pot wiper...at -9V all pixels just start to turn on at full contrast setting. The effect is quite subtle and there is not enough contrast for those pixels that should be turned on. So a -9V supply is insufficent with any image very pale. With -10V the background pixels are obviously turning on and this is the best contrast setting for an image.With -12V all pixels are distinctly dark.
Use of the 10K potentiometer provides a control that is not too sensitive. I found 20K was to sensitive to set the desired contrast. (EDIT: 26/12/08)
As this display needs a healthy amount of negative voltage, connecting the top of the potentiometer to Ground (0Volts), as opposed to +5V, also makes the contrast control less sensitive.
LED BACKLIGHTING
NOTE: This display has no current limiting resistors in place for the LED backlighting.
You should insert a series resistor to limit the current. The back light configuration has 76 LEDs. 38 paralleled sets of 2 LEDs in series. The datasheet indicates up to 390mA for the back light so this indicates 10mA per LED
When supplying the LED back light from +5V, a series resistor of 4.7 Ohms will give a LED current typically around 200mA and with a spread 150mA to 300ma going by the Forward Voltage Max/Min figures in the datasheet.
EDIT: 12/12/08 With +5V the following LED currents were measured.
4.7 Ohm series resistor gave 195mA (At a brightness level I preferred)
5.1 Ohm series resistor gave 186mA (Possibly a brightness level a little low)
EDIT 6/1/09
MISSING NEGATIVE VOLTAGE GENERATOR CIRCUITRY
For those interested the missing IC is an Epson SCI7661
R7 =1Meg, and C1,C2 and C7 would be 10uF 16V
Check the SCI7661 datasheet for the configuration needed to attain the -10Volts
As I write this there is a SCI7661 source on eBay US$9 plus freight of US$5 to NZ. Ex Hong Kong
The jumpers will need addressing. Un-solder JP4 and deduce what should be done with JP1 thru JP3 using the datasheet.
I hope you find this information useful.
Last edited: