20 Aug
2001
20 Aug
'01
3:06 p.m.
It's a George-Risk keyboard, made by the company in Nebraska, IIRC, and the
encoder is a custom-coded version of a standard GI product. There's a fix
inherently possible, but it involves wire and solder ... BEWARE!
see below, plz.
Dick
----- Original Message -----
From: "Foust, Daniel" <DanielF(a)McLean.gov>
To: <classiccmp(a)classiccmp.org>
Sent: Monday, August 20, 2001 12:40 PM
Subject: Apple II+ Keyboard Encoder
My Apple II+ Keyboard Encoder card is faulty and I
would like to know if
it's fixable. I've tried several different keyboards and 2 motherboards so
I'm certain it's the encoder. I've run a keyboard diagnostic and it is
acting as though the control key is always pushed but it is not. It does it
with 3 different keyboards so it's not a "stuck key" problem. Any ideas
or
help much appreciated. Thank you.
If you know precisely what the code produced for each key is, which shouldn't
be
a problem, you need simply program an MCU to produce that code set and then
haywire it in place of the existing and apparently defective encoder. IIRC, the
encoder was a General Instruments model, albeit custom coded, but one can take
the output generated by another version of the same device and translated it via
a lookup table in EPROM. This has to be done externally to the Apple][+, of
course, but its not difficult to manage.
You could also put a "band-aid" on it by programming a PIC or similar MCU to
look only at the existing incorrect key codes and implement a firmware repair
based on which key combinations are apparently wrong.
I had to wrestle with this problem once, admittedly a long time ago, but the fix
is still valid. If your MCU has enough ROM to look up a value for each scanned
key in combination with the normally isolated keys, e.g. shift, ctl, alt, cap's
lock (if there is one), then it's quite straightforward.