CFAG320240CX-FMI-T on memory bus

[krudolf]

Hello,

I am using a CFAG320240CX-FMI-T LCD powered with 5V in 6800 mode and connected to a memory bus with CS# inverted = E, A0 = BA0, A1 = R/W# and D0-D7 connected to DB0-DB7. Further CS# is low and RESET# connected to a controller causing a low- to high edge after power up.
CS# is activated with address range starting at 0x40000000, so I get at address 0x40000000 data and parameter write, at 0x400000001 control write.
Timing is about 1100µs for E, BA0, R/W# and D0-D7 at every memory write or read cycle. Delay between E, addresses and data is less than 10ns, i.e. all signals change at the same time.
I am using a part of an initialization sequence from the data sheet (see below, mw.b means memory write byte, CS# is 8-bit):
mw.b 40000001 40
mw.b 40000000 30
mw.b 40000000 87
mw.b 40000000 07
mw.b 40000000 27
mw.b 40000000 42
mw.b 40000000 ef
mw.b 40000000 28
mw.b 40000000 00

mw.b 40000001 44
mw.b 40000000 00
mw.b 40000000 00
mw.b 40000000 ef
mw.b 40000000 60
mw.b 40000000 09
mw.b 40000000 ef
mw.b 40000000 00
mw.b 40000000 00
mw.b 40000000 00
mw.b 40000000 00

mw.b 40000001 5a
mw.b 40000000 00

mw.b 40000001 58
mw.b 40000000 56

mw.b 40000001 46
mw.b 40000000 00
mw.b 40000000 00

mw.b 40000001 5d
mw.b 40000000 04
mw.b 40000000 86

mw.b 40000001 59
The commands are fed into a terminal emulation (about 1-2 commands/s).
There is no reaction of the LCD at all (only back light on after power up).
Vo is -18.8V.

Any ideas (maybe setup time between address and E)?

Crystalfontz LCD Blog

Looking for additional LCD resources? Check out our LCD blog for the latest developments in LCD technology.

 

[cosmicvoid]

Is the DISPOFF (pin 19) tied high? Otherwise, I don't see an obvious problem. But, I would be suspicious of a timing issue, even though I can't identify one.

Can you get the screen to go dark by turning the contrast pot?

 

[krudolf]

Pin 19 is high (about 5V); but turning the contrast pot between -10v and -24v does not change the blue display.

 

[cosmicvoid]

So, the backlight works, but you have no evidence that the controller logic is actually functioning. The only thing I can suggest is to try using an indirect hookup, i.e. use I/O port connections to sequence the control and data signals, to verify the display is functional.

 

[krudolf]

found the solution now. setup and hold times >0 are important, but my hardware generated times of 0 or <0 ns (depending on processor timing). display is running now.

Thanks

 

[CF Tech]

Good to hear you got it going.

Please feel free to share schematics and or code samples for the other members of the forum if you are able to.