AT/PS2 Keybord - PIC Interface?

[Kamikaze47]

Ok cool. I spotted a couple of them already and changed them to suit a 16F84A, but i missed a couple. Thanks.

[mytekcontrols]

Kamikaze47,

Just out of curiousity, what are you doing that requires a full keyboard instead of something like a keypad?

Oh, and also just to let you know there are 2 keys that currently will not be decoded properly. The first is the PrntScrn/SysReq key which will put out a "*", and the other is the Pause/Break key which will mimic the NumLock key. I plan to do some more decoding for these, but simply haven't got to it yet. Any improvements you make and/or features you add, please post them here for all to see.

Thanks,
Michael

[Kamikaze47]

At the moment im going to be transmitting the decoded ascii characters over an RF link to another pic which will write them to an LCD screen. I've got the RF part working, so now just need to get the keyboard interface working.

Another project im thinking of doing with this is a device that goes inbetween your keyboard and your pc which will convert the keyboard to DVORAK. A friend of mine wants to have 2 keyboards on his PC - 1 QWERTY and 1 DVORAK, but the keyboard settings in windows are global so you cant have 2 different keyboard layouts.

Thanks for your help mytekcontrols. I'm sure this info will be usefull to a number of ppl on these forums.

[mytekcontrols]

Kamikaze47:

At the moment im going to be transmitting the decoded ascii characters over an RF link to another pic which will write them to an LCD screen. I've got the RF part working, so now just need to get the keyboard interface working.

Sounds like a very interesting project. Is this something you are doing for fun, or do you see it having commercial possiblities? If so, I hope it does well for you.

Well I put together all the pieces into one file and tested it out stand-alone. And then I pushed the compile button, while remembering that famous Clint Eastwood line "do you feel lucky?". Well I think you can guess the answer to that one. Of course there were a few errors , but nothing too terrible.

Here is what was missing:

<img src="http://www.picbasic.co.uk/forum/images/icons/icon2.gif" align="absmiddle"> from the equates:

init var bit            ' flag to use timeout test alternate return path 
ktable var word         ' translation table index

<img src="http://www.picbasic.co.uk/forum/images/icons/icon2.gif" align="absmiddle"> and from the PIC18F aliases:

INT2IP  VAR INTCON3.7   ' INT2 external interrupt priority bit

(I screwed up on this one, and accidentally cut & pasted the wroung item)

Anyway once I fixed it, everything compiled and ran just fine. Total bytes required = 2626 (wow! pretty interesting result).

Download the complete and tested file below.

[Kamikaze47]

cool. thanks.

i havnt had time to get this running yet, i will work on it over the weekend and see how i go.

yea im basically just doing it for fun. but there are a couple of projects that i have done in the past that i think would have commercial potential, but i wouldnt know how to go about doing anything about them.

[mytekcontrols]

I tweaked the code in order to better mimic the actions of the keypad on a PC. Specifically I now have shifted functions in both NumLock and non-NumLock modes as outlined below:

NumLock Mode:
Shift-Up --- numbers are generated 0,1,2,3... (ascii $30 - $39)
Shift-Dn --- standard navigation INSERT, END, DOWN... (ascii $90 - $99)

non-NumLock Mode:
Shift-Up --- standard navigation INSERT, END, DOWN... (ascii $90 - $99)
Shift-Dn --- numbers are generated 0,1,2,3... (ascii $30 - $39)

Note: when in a number output mode, the delete key on the numpad = ".", else it generates the ascii $7F code for delete. Also please notice that all ascii assignments above $7F are of my choosing, and not a part of the standard ascii code set.

Code space requirements have gone up to 2666 with this new version.