3 May
2002
3 May
'02
7:51 a.m.
I just came across a large number of loose 2102 chips in a junkbox, and
wondered if anyone has a schematic of a test circuit for these. I'm not
sure how many of them there are yet, but they all appear to be from the
same batch; AM9102BPC/P2102A-4 7632QD [32nd week, 1976]. An couple
interesting facts I dug up on these chips is that the AM9102 was AMD's
first RAM product, and it was first produced in quantity in 1975.
I'm thinking these chips might be a good source of ram for a 6800 based
homebrew system, especially since I already have the chips and lots of
wirewrap (and patience)...
-Toth
3 May
3 May
9:05 a.m.
It would probably be easy to program a PIC or UBICOM
to GO/NOGO them. Maybe even look for bit pattern
sensitivity or check for speed. I'm building old
8080/6800/8008 stuff, and I'm thinking of using these,
but that's a lot of socket, wire wrap, board real
estate, and time! Wow, do they use power also, even
the low power ones. I prefer using 2114(L)'s, but
those seem to have become a little scarcer and more
expensive recently. OTOH, 2102's are very reliable. If
you have a good #, I'd trade you a 6871 or 6875 - the
6800 clock chips, surprisingly hard to find/expensive
- for some.
--- Tothwolf <tothwolf(a)concentric.net> wrote:
I just came across a large number of loose 2102
chips in a junkbox, and
wondered if anyone has a schematic of a test circuit
for these. I'm not
sure how many of them there are yet, but they all
appear to be from the
same batch; AM9102BPC/P2102A-4 7632QD [32nd week,
1976]. An couple
interesting facts I dug up on these chips is that
the AM9102 was AMD's
first RAM product, and it was first produced in
quantity in 1975.
I'm thinking these chips might be a good source of
ram for a 6800 based
homebrew system, especially since I already have the
chips and lots of
wirewrap (and patience)...
-Toth
__________________________________________________
Do You Yahoo!?
Yahoo! Health - your guide to health and wellness
http://health.yahoo.com
2:25 p.m.
Probably the most important test will be reduntdant mapping (where the same
region of the memory appears twice or more, meaning some part isn't working at
all) which indicates some sort of address/data interaction. Once you write a
test procedure, try letting an address line float and see what happens. Does
your test find it? Extend the test so it tests the device as a 4kx1. Does it
pass? There are lots of tests that do!
Dick
----- Original Message -----
From: "Loboyko Steve" <sloboyko(a)yahoo.com>
To: <classiccmp(a)classiccmp.org>
Sent: Friday, May 03, 2002 3:05 PM
Subject: Re: 2102 SRAM test circuit?
It would probably be easy to program a PIC or UBICOM
to GO/NOGO them. Maybe even look for bit pattern
sensitivity or check for speed. I'm building old
8080/6800/8008 stuff, and I'm thinking of using these,
but that's a lot of socket, wire wrap, board real
estate, and time! Wow, do they use power also, even
the low power ones. I prefer using 2114(L)'s, but
those seem to have become a little scarcer and more
expensive recently. OTOH, 2102's are very reliable. If
you have a good #, I'd trade you a 6871 or 6875 - the
6800 clock chips, surprisingly hard to find/expensive
- for some.
--- Tothwolf <tothwolf(a)concentric.net> wrote:
I just came across a large number of loose 2102
chips in a junkbox, and
wondered if anyone has a schematic of a test circuit
for these. I'm not
sure how many of them there are yet, but they all
appear to be from the
same batch; AM9102BPC/P2102A-4 7632QD [32nd week,
1976]. An couple
interesting facts I dug up on these chips is that
the AM9102 was AMD's
first RAM product, and it was first produced in
quantity in 1975.
I'm thinking these chips might be a good source of
ram for a 6800 based
homebrew system, especially since I already have the
chips and lots of
wirewrap (and patience)...
-Toth
__________________________________________________
Do You Yahoo!?
Yahoo! Health - your guide to health and wellness
http://health.yahoo.com
11:09 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 864 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/20020504/93fce4a3/at…
10:30 a.m.
These days you can get a 32K*8 SRAM chip in a 28 pin
DIL package for very
little money. And they're easy to find. There's little wiring up to do
(wiring up rows of wire-wrap sockets pin-for-pin is not my idea of fun...)
One of those, and a suitable EPROM would be enough memory for most small
homebrew 8 bit systems. 32K RAM + 32K EPROM is a full memory map. If you
need more RAM and less EPROM, use 2 of the 32K*8 RAM chips and just
disable them for the part of the memory map where you want to put the
EPROM. It's easier to do that than to mess about with several smaller
RAMs IMHO.
-tony
Agreed, KISS is the idea and these 32K srams are very reliable and
plentiful.
Dig into discards or osolate mid to late 486 and early pentium
boards, these boards used 32K or 64K x 8 bit DIPs in sockets.
Ditto to EPROM's in ceramic package w/ quartz windows from those.
Usually 64K x 8 or 128K. Better yet, flash chip. Still can be hot
swap flashed on "disposeable" late model 486 or mid period pentium
boards w/ your own file and bios flash program that doesn't care
what file it is of your own creation and use flash IC on your
homebrew 'puter instead of eprom as an "eprom". This beats having
eprom programmer. :-)
These common flash chips (DIP package) is 32pin and are usually 128K
or 256K x 8.
I dug up many boards this way for free or for a price of a decent
meal.
Cheers,
Wizard
2:16 p.m.
You're right about what they are but I think you may be making a mistake
attempting to use 'em when there are probably 64kx8, or at least 32kx8 srams
on a board lying in the closet. 64 of these babies would take a huge amount
of power, yet only amount to 8Kx8. One of the others, or maybe two, would use
1% of the power, and of the effort, and of the socket cost, yet do the whole
job.
If you really want to use 'em, find an application they fit. That'll be a
real classic and worth the trouble.
Thorough testing is something that ought to happen in a board for your PCB.
If you wire 'em on an ISA ww card, you can test 'em 8 at a time. Don't solder
the sockets to anything, so you can use 'em for something useful later on. I
do this sort of thing all the time and find it easy to reuse ww sockets.
A memory test that doesn't let the occasional failure slip by will be an
interesting software challenge, too.
Dick
----- Original Message -----
From: "Tothwolf" <tothwolf(a)concentric.net>
To: <classiccmp(a)classiccmp.org>
Sent: Friday, May 03, 2002 1:51 PM
Subject: 2102 SRAM test circuit?
I just came across a large number of loose 2102 chips
in a junkbox, and
wondered if anyone has a schematic of a test circuit for these. I'm not
sure how many of them there are yet, but they all appear to be from the
same batch; AM9102BPC/P2102A-4 7632QD [32nd week, 1976]. An couple
interesting facts I dug up on these chips is that the AM9102 was AMD's
first RAM product, and it was first produced in quantity in 1975.
I'm thinking these chips might be a good source of ram for a 6800 based
homebrew system, especially since I already have the chips and lots of
wirewrap (and patience)...
-Toth
8586
days inactive
8586
days old
5 comments
5 participants
participants (5)
-
ard＠p850ug1.demon.co.uk -
edick＠idcomm.com -
jpero＠sympatico.ca -
sloboyko＠yahoo.com -
tothwolf＠concentric.net