Has anyone had anyluck impersonating a SED133x 4 bit lcd controller?
I have been trying to get a controller-less LCD running right now, and it says it should be connected to an SED133x controller, but I figured it could be too hard to fake. Anyone done this before?
Jimmy
Jimmy
Comments
http://forums.parallax.com/forums/default.aspx?f=25&m=261658
Just some notes:
There are several kinds of controller-less LCDs with different requirements in terms of control lines:
- those who need Data, x shift clock, frame and load.
- those who need Data, x shift clock, frame, load and M (alternate signal, around 400 Hz).
- those who need Data, x shift clock, frame, load, y shift clock and y start line (may or may not need M).
Basically are all the same, but the last one needs a bit more signals. y shift clock is like load but starts before it and ends after it.
Do not forget that the high-voltage (negative or positive) should only be applied after the LCD was powered and better after you sent a full frame.
If you knew all this... well ignore it.
Feel free to ask
Edit: I forgot: If it is a 5V display, I'd recommend you level shift the propeller's outputs with some HCT buffers (I used OR gates because... they where laying around) with a resistor of small value in series from the prop to the buffer. My displays did not like the 3.3 V signals...
Post Edited (Ale) : 6/27/2008 7:31:43 AM GMT
I have also attached the sheet I found on this LCD
The ECL signal has to be toggled every 16 XCL... does not seem very difficult, the driver sends bursts of 32 pixels (reads longs from HUB ram), after 4 groups of 4 pixels you can toggle this signal. luckily 320 is a multiple of 16
From what I read, LAT is wider than LP, starts before and ends after LP, but it is something you can test.
Well put hands to work and tell us about your project (do you have a logic analyzer ?) (A simple one can be done with a propeller, up to 100 Msps @ 100 MHz) That helped me a lot with my driver.
You will have to play a bit with the pointers to use a linear frame buffer to the way the display expects data, but for a simple proof, use it as it is.