I’ve breadboarded the debounce circuit. I get a nice, stable, debounced, enable pulse. (Here’s a nice page about switch bouncing and ways around the problem.) There is a strange, seemly random pulse that occurs on occasion when I apply power to the circuit, but it’s effects should be minimal.
I’ve also greatly expanded the description of my project in the z80 project page.
Now that the design of Tester has been successfully tested, work can continue on building the case for it.
To Do:
It’s pretty much self-explanatory, I have put labels all over it. I’ve also covered up the parts of the breadboard that arn’t relevant to this project.
I found this great website that helped me to getting this project working. http://www.epemag.wimborne.co.uk/resources.htm The first page of this website is very similar to what I’m doing. Thanks to Steve from Buffalo, NY for linking to this from his site on LCDs. http://home.iae.nl/users/pouweha/lcd/lcd_examp.shtml
As it turned out, the problem was the contrast of the screen. I had connected the pin directly to ground, thus giving me the most contrast. I thought this would be a good thing, but that made it far too dark to see. I think I was viewing the screen from a bad angle as well. After connecting the constrast to a potentiometer, and adjusting the constrast down, all worked out just fine.
As promissed, I have some pics. Unfortunately, the pics have bad resolution and they were taken in low light. The pic of the screen itself didn’t turn out at all. The first one is the breadboard of this project, and the second one are the switches I’m using to test it. I’ll post some better pics when I get them. (Yes, I’ll use a camera with a higher resolution than 100k pixels.)
Everything is wired and up ready to go. I turned it on, flipped the switches, followed the initialization procedure, and nothing happened. But thats to be expected since electronics almost never work correctly the first time.
I figure one of three things is the problem, either the enable line isn’t being used properly, the procedure is wrong, or the LCD is bad. Toggling the enable causes the module to execute the instruction that is on the data lines. Every other pin has been checked out and the voltages are at expected levels. The three possible problems are listed, in my opinion, in order of decreasing probability of being the cause.
Experience has taught me that the probable cause is the enable line. I unfortunately have a really rotten data sheet on this module, and I’m having to go off of the experiences of others. I don’t even know if I’m supposed to keep the enable high and bring it low or pulse it high and back low again. I have conflicting sources of data on this.
It would be much easier to get the LCD working if I had a working uP. It’s also much easier to get a uP working if you have a working LCD screen. Seems like a Catch 22 or the chicken vs egg problem to me. I’m going back to working on it, or doing homework that’s due tomorrow.
I’ve been designed and planning the building of this tester for weeks now. I have draw up diagrams and stuff for this. It isn’t really needed but it makes my life easier. Problems are easier to eliminate in the design stage.
So, I’m almost done with breadboarding, and can begin testing it to see if it will work. I have previously applied power to the module, and the top row was dark. From what DigiTan has written, this is a good sign.
If this test is succesful, Ill take a pic showing the result(I keep saying that, but I promise) and will begin building a case for it. It will just be a black Radioshack project box, but I think it will turn out quite good looking. For a project box case anyway.
I found an interesting bit of code that fetches values from a look-up table. It was located in “101 projects for the z80″ by Frank Tedeschi. The first data location has to be the relative address for the value the user wishes to get. This will be “added” to hl, forming the full address. I have preserved the original comments.
Input:
(data) contains the position of the value in the table.
ld hl, data ; set pointer
ld l, (hl) ; form address to look-up table.
ld a, (hl) ; get answer from table.data:
.db $01, $05, $00….
Output:
Register a would contain $05. This routine is 5 bytes long, and takes 24 TStates.
Limitations:
There are some limitations however, the table must use 1 byte data, and be no longer than 254 bytes. The lower byte of ‘data’ would have to be a zero. This would limit the locations the LUT can be stored.
I’ve created a stub page on this project. More will be added later.
I just added a nice long page explaining some of the first phase of my project. It’s turning out way longer than I thought it would. :-/ Anyhow, check it out here.
Last updated: 2/12/2007
It finally works. I need to get some pics up and show off my work. (yes, this is just a test message. )
Watch this spot.
