HD 44780 display question

What's the maker and muddle number of the synth? I did a display transplant on my Korg DSS-1, which fortunately didn't require any hacking.

What inspired the transplant was the original electro-luminescent back light slowly dying from old age. I initially took apart the original EL display, replaced the EL back light with a sheet of plastic, hot melt glued some LED's along the edges, and ended up with a rather ugly back light. Oh well.

Here's some drivel I found on addressing 1601 displays: This looks interesting: See table 2.3 Instruction Set. The DL bit seems to set the word size to or 4/8 bits. If it default to 4 bits, which methinks might be the culprit.

I also suggest trying a 2 line 1602 display, using only one line of the display. However, I don't know if it will work, or what it will look like.

yep. if you don't want to do a firmware hack your only option would be to build a converter (a uP would probably work best) to collect the 8 bit data writes and output them as the appropriate 16 bit signals. I would suggest an arduino and some support hardware. the actual task is not difficult - latch the 2 8-bit signals and then output as one 16-bit.

I think I have some ATMega328s in stock around here. I'd really rather leave the gear stock, and frankly it seems somewhat easier to do it that way than try to compile a hacked binary, flash an appropriate DIP EEPROM (which I don't have, with a programmer I don't have) and pull the original (non-socketed) ROM and replace it.

Maybe.

Hacked source code I should say, rather. Though I think there may be a binary image of the hacked code available on that site, somewhere.

try this:

the pinout is different - you have to check if it is software compatible - it seems it is but YMMV

bitrex wrote in news:1yAIy.182694$Ia.62645 @fx08.iad:

Ever had a problem with those displays so I digged out the datasheet to find out what was going wrong. These displays can contain a 128 printable characters though only 16 of them are visible. Like you, I found out where they were stored and I wrote PIC-code to be able to handle both types. The choice could be made by setting or removing a jumper. I even planned to rewrite the code to make that jumper superfluous but I did not come so far.

To handle the 2x8 display you need to keep track of the write cursor and add that 40H to it when a character needs to be written between position 8 and 15. (Off course back when you enter a new line.) The interface itself can be used in 8 bits or 4 bits mode, not 16.

You can build some circuit that catches the code sent to the display and filter out that forth and backward moves of the cursor. Which requires some knowlegde of the (display) chip and the implemantation of the hardware. You also have to realize there can be cursor movements that should not be suppressed. IMHO hacking the firmware is a more efficient way to go. Unless that firmware code is written very badly, there will be one routine that sents characters to the display. That routine wil handle the sursor movement and the only thing you need to do is changing the 40H to 0 in one or two places.

petrus bitbyter

OLED displays are great, but you need to add a microcontroller to run it. The displays are inherently graphical. Text display involves font libraries. I've made instruments that use three different fonts at different times. The font libraries are open source and were created by one of my associates, Alan Stern. I ran the display with I2C with one of the Arduino family MCUs.

It's apparently written very badly. ; ) I think the author of the article in the link I posted notes there are "write" routines scattered all over the code.

Instead of replacing the whole display with an OLED display, I'm thinking about suggesting to this customer to just replace the EL backlight with an LED backlight and leave the LCD intact. I don't really want to get into the problems of removing the ROM chip and replacing it with hacked firmware unless he desperately wants that fancy display.

Using a nice LED backlight would take care of the dual issues of EL fading over time, and the fact that the high-voltage inverter on some of these devices are pretty prone to failure.