6 Oct
2005
6 Oct
'05
4:09 p.m.
As some of you know, I'm helping with an exhibition of classic machines
in the Department of Computer Science -- for Open Day tomorrow
(Friday), and running conducted tours on Wednesday as well.
Well, one of the supposedly-working exhibits is my KIM-1, but it died
this afternoon. I can't get it to do more than occasionally display a
single zero, on the leftmost 7-segment display. A cursory look with a
logic probe (and no docs to hand) shows the clock is running and at
least a few address lines are too. I bet it's a memory fault.
What are the common faults? I'm hoping it's one of the 2102 RAMs,
because I stand a chance of having a suitable replacement. However, I
fear it's more likely a 6530, and I assume they're unobtanium. Has
anyone ever come up with a kludge to effectively use a 6532 (and are
they any easier to find?) with an EPROM? Or anything else?
--
Pete Peter Turnbull
Network Manager
University of York
6 Oct
6 Oct
2:03 p.m.
fear it's more likely a 6530, and I assume
they're unobtanium. Has
anyone ever come up with a kludge to effectively use a 6532 (and are
they any easier to find?) with an EPROM? Or anything else?
My Commodore 8250LP disk drive has an official kludgeboard in what seems
to be a socket designed for an 6530. It contains a 6532, a ROM, and a
simple TTL chip ('04 or something like that).
IIRC, though, the chip select levels of the 6530 were mask-programmable,
so this might not work in all cases. And I don't think I can get any
details of this to you by tomorrow anyway.
-tony
4:43 p.m.
Well, one of the supposedly-working exhibits is my
KIM-1, but it died
this afternoon. I can't get it to do more than occasionally display a
single zero, on the leftmost 7-segment display. A cursory look with a
logic probe (and no docs to hand) shows the clock is running and at
least a few address lines are too. I bet it's a memory fault.
What are the common faults? I'm hoping it's one of the 2102 RAMs,
because I stand a chance of having a suitable replacement. However, I
fear it's more likely a 6530, and I assume they're unobtanium. Has
anyone ever come up with a kludge to effectively use a 6532 (and are
they any easier to find?) with an EPROM? Or anything else?
6532s are definitely not pin-compatible, and I've never seen any means of
converting them or integrating an EPROM for the mask ROM component. My own
KIM-1 is a little flaky and after awhile will do something similar, but
usually only after it's been running for some time or if there was a power
fault (and RS usually sets it right).
You could try replacing the RAM, but I don't think that explains your
symptoms.
--
---------------------------------- personal: http://www.armory.com/~spectre/ --
Cameron Kaiser, Floodgap Systems Ltd * So. Calif., USA * ckaiser at floodgap.com
-- "I'd love to go out with you, but my personalities each need therapy."
-----
7 Oct
7 Oct
11:25 a.m.
At http://www.6502.org/oldmicro/buildkim/buildkim.htm you can find
schematics of the original KIM and one that can be built with a 6532. There
might be enough info there in order to replace the 6530 with a 6532+EPROM
daughter board.
On 10/6/05, Cameron Kaiser <spectre at floodgap.com> wrote:
Well, one of the supposedly-working exhibits is
my KIM-1, but it died
this afternoon. I can't get it to do more than occasionally display a
single zero, on the leftmost 7-segment display. A cursory look with a
logic probe (and no docs to hand) shows the clock is running and at
least a few address lines are too. I bet it's a memory fault.
What are the common faults? I'm hoping it's one of the 2102 RAMs,
because I stand a chance of having a suitable replacement. However, I
fear it's more likely a 6530, and I assume they're unobtanium. Has
anyone ever come up with a kludge to effectively use a 6532 (and are
they any easier to find?) with an EPROM? Or anything else?
6532s are definitely not pin-compatible, and I've never seen any means of
converting them or integrating an EPROM for the mask ROM component. My own
KIM-1 is a little flaky and after awhile will do something similar, but
usually only after it's been running for some time or if there was a power
fault (and RS usually sets it right).
You could try replacing the RAM, but I don't think that explains your
symptoms.
--
---------------------------------- personal:
http://www.armory.com/~spectre/ --
Cameron Kaiser, Floodgap Systems Ltd * So. Calif., USA *
ckaiser at floodgap.com
-- "I'd love to go out with you, but my personalities each need therapy."
-----
5:49 p.m.
On 10/7/05, Eric J Korpela <korpela at ssl.berkeley.edu> wrote:
At http://www.6502.org/oldmicro/buildkim/buildkim.htm
you can find
schematics of the original KIM and one that can be built with a 6532. There
might be enough info there in order to replace the 6530 with a 6532+EPROM
daughter board.
I've seen that before - nice writeup on what has to be done to make a
KIM from easily-available parts. I think the hardest thing is to find
a suitable keypad. If one were to hand-wire one, one could even use
Radio-Shack protoboards with the 22-position 0.154" spacing edge
fingers if one had external accessories to attach (or use 0.1" headers
and ribbon cable for "modern" expansions). Easy to max out the memory
these days... one chip!
-ethan
3:51 a.m.
Am 7 Oct 2005 0:09 meinte Pete Turnbull:
As some of you know, I'm helping with an
exhibition of classic machines
in the Department of Computer Science -- for Open Day tomorrow
(Friday), and running conducted tours on Wednesday as well.
That's unusual, standard procedure is that a machine dies at
the opening day at the first real presentation - no matter
how often you testet before.
Well, one of the supposedly-working exhibits is my
KIM-1, but it died
this afternoon. I can't get it to do more than occasionally display a
single zero, on the leftmost 7-segment display. A cursory look with a
logic probe (and no docs to hand) shows the clock is running and at
least a few address lines are too. I bet it's a memory fault.
Not necersarry. Memory faults are rather rare on KIMS (in fact,
I never had one). First I'd look for dirt at the connectors and
arround the LED drivers, since a shortcut there might just screw
everything ... I found more than one KIM where the one or more
of the port lines where fried.
What are the common faults? I'm hoping it's
one of the 2102 RAMs,
because I stand a chance of having a suitable replacement. However, I
fear it's more likely a 6530, and I assume they're unobtanium. Has
anyone ever come up with a kludge to effectively use a 6532 (and are
they any easier to find?) with an EPROM? Or anything else?
No, not made of unobtainium, KIMs are still common, just the
process of separation is rather painful :) If it's one of the
6532, a replacement needs to be build. First thing would be
to connect a logic analyzer to see if the CPU is still running
a programm in ROM or not. If yes, all you need is either to
replace the internal RAM or the I/O Ports of the 6530. If you
have to hurry, adding a 6530 via a piggy pack PCB might be a
doable way - or put both onto a PCB. if there's more time,
I'd go ahead and put the I/O Part of the 6530 into a CPLD,
unsolder the 6530, bend the I/O pins below the chip (maybe
add also some insulator) and glue and wire the CPLD onto the
back of the board. Perfect disguise, as long as nobody turns
the board arround.
Same way if the ROM is fried (extrem unlikely): just add a
some (SMD) EPROM (1K used) onto the back.
Hans
--
VCF Europa 7.0 am 29/30.April und 01.Mai 2006 in Muenchen
http://www.vcfe.org/
12:48 p.m.
Am 7 Oct 2005 0:09 meinte Pete Turnbull:
I've found a fairly effective way to prevent this is to have a good
toolkit and a pile of test gear at the demonstration too, and let the
machine know you intend to use it if necessary :-)
-tony
As some of you know, I'm helping with an
exhibition of classic machines
in the Department of Computer Science -- for Open Day tomorrow
(Friday), and running conducted tours on Wednesday as well.
That's unusual, standard procedure is that a machine dies at
the opening day at the first real presentation - no matter
how often you testet before. 
7:33 p.m.
On Fri, 7 Oct 2005 20:48:23 +0100 (BST)
ard at p850ug1.demon.co.uk (Tony Duell) wrote:
At work this past week I was given a nice small Telequipment 'scope that
is perfect for that sort of mobile use. It's an S54A and seems nice and
bright and works great. Single trace, which is probably why I got it
for free. It's my first Telequipment 'scope, though some of my Tek
scopes are British made so probably essentially 'Telequiment' scopes as
well. It's not a painful 'luggable' like my Tek 465M.
Am 7 Oct 2005 0:09 meinte Pete Turnbull:
I've found a fairly effective way to prevent this is to have a good
toolkit and a pile of test gear at the demonstration too, and let the
machine know you intend to use it if necessary :-)
As some of you know, I'm helping with an
exhibition of classic
machines in the Department of Computer Science -- for Open Day
tomorrow(Friday), and running conducted tours on Wednesday as
well.
That's unusual, standard procedure is that a machine dies at
the opening day at the first real presentation - no matter
how often you testet before. -tony
7015
days inactive
7017
days old
7 comments
7 participants
participants (7)
-
ard＠p850ug1.demon.co.uk -
chenmel＠earthlink.net -
ethan.dicks＠gmail.com -
Hans.Franke＠siemens.com -
korpela＠ssl.berkeley.edu -
pete＠dunnington.plus.com -
spectre＠floodgap.com