Does anyone have any experience with HD44780 type displays?

I am trying to use one that is based on a S6B0108 controller and it seems to be 100% HD44780 compatible right down to the timing. Initialization appears to be working, I can move the cursor, set it to flash or not, blank the display, clear it, etc… However, when I send it a character to display, all that happens is the cursor moves. No character is displayed.

What could I be doing wrong?

Are there known timing issues with 650xs and these devices? This does not seem likely because I can get the thing to anything I want, except display a character. I’m using a 6507 at 1.8432 Mhz and the data is holding for 100ns after the enable strobe, which seems to be well within spec.

Any thoughts appreciated.

_________________
Bill

Have you tried reading back the characters ?

I've used those LCDs tons. It sounds like you're going too fast for it. 1MHz is a good limit to stick to. In the first design where I used one of these displays, the 65c02 switched between 1MHz and 180kHz (to save power when it didn't have much to do), so I put the LCD right on the bus. After that I always interfaced to the display through a VIA.

I'm going to try that when I get a chance.

_________________
Bill

I'm writing to it pretty slowly (time between writes) to begin with and given that it responds correctly to every command, I'd say it's capable of receiving the data sent to it.

I'll try to read a character back when I get a chance. It may just be a dud.

_________________
Bill

I'm not talking about the amount of time you give it between writes, but of course you do have to pay attention to that too. The ones I've used needed a write time of something like 450ns which is almost the entire phase-2-high time at 1MHz, and too long any way you work your glue logic at 1.8432MHz.

It is a standard display sold at Adafruit and tutorials / datasheet are available here:

http://www.adafruit.com/index.php?main_ ... cts_id=181

You might be able to get some help and translate it to the project / microprocessor you are working with.

Control data is getting in but not display data. For some reason there is a double enable occurring when I go through the Tiny Basic call.

The data is not asserted when the first one occurs and the display may be busy when the second one comes. However, this does not explain why it will respond to control commands.

I have to put this project down or a while but will try a ML routine to test it next time to avoid this weird double enable.


Thanks the for input folks.

_________________
Bill

Could the problem be the 6502's extra accesses?

Or even the Basic combined with the address decoding. (For example, I think I saw that BBC Basic implements a byte-sized peek using a word-sized peek, which means it does four reads to adjacent addresses)

Do you have the source for your Tiny Basic? We could see what it does for PEEK and POKE.

Cheers
Ed

Similar writes to the I/O via ML routines do not cause the same double enable.

The Tiny Basic is a disassembly of Tom Pitman’s version written for 6502. It was originally written or the KIM-1 and I extracted it from a hex file he had on his site. Tom put a very early version of what I did up on his site. You can see that version there, but it will not be revealing in this case. There is no real PEEK or POKE. There a small bits of code that are called via the USR() call that perform these functions. I have a newer version I’ve not yet passed over to Tom that is a bit better formatted and documented, but even then, following the code is difficult.

If anyone is interested I guess I could send it to them via e-mail, but this site does not allow attachments (??!?).

_________________
Bill

I found another tutorial on this LCD. How is the project going, by the way?

How to Use LCD Module
HD44780 Character LCD Controller
http://elm-chan.org/docs/lcd/hd44780_e.html