22 May
2000
22 May
'00
7:05 a.m.
It's that time of year again, when I think about designing an Altair/IMSAI
work-alike. Now that I have the ability to generate schematics/layouts, I
may actually try it.
Anyway, some parts have become scarce, so I would have to use substitutes.
For example, Jameco no longer carries the 8T97 buffer/driver. As I recall
the 8T97 is an LS373 without the inverter on the gate *and* is faster and
has a higher drive current. I've heard that the MC6887 is a sub, but I can't
find a datasheet on it.
The 8101 dual-port 256bx4 RAM is no longer available, but Jameco has a 5101
which looks like a likely sub. I would forego this type for a larger SRAM,
but I guess that it too would have to be dual-ported. I need to look at the
schematics for an S100 memory board.
Someone on-list did a design for his own clone, but I can't remember who it
is. I'm sure that it can be done cheaper than Tom Fischer's IMSAI-2 though.
Rich
22 May
22 May
7:21 a.m.
On Mon, 22 May 2000, Cini, Richard wrote:
It's that time of year again, when I think about
designing an Altair/IMSAI
work-alike. Now that I have the ability to generate schematics/layouts, I
may actually try it.
Anyway, some parts have become scarce, so I would have to use substitutes.
For example, Jameco no longer carries the 8T97 buffer/driver. As I recall
the 8T97 is an LS373 without the inverter on the gate *and* is faster and
has a higher drive current. I've heard that the MC6887 is a sub, but I can't
find a datasheet on it.
Ummm... has the T.I. 74LS367 gone extinct? We always used that as an
'exact match' sub. Same for the 74LS368 as a EM sub for the 8T98...
Many (most?) S-100 boards/kits/etc. used these parts interchangeably.
-jim
---
jimw(a)computergarage.org || jimw(a)agora.rdrop.com
The Computer Garage - http://www.computergarage.org
Computer Garage Fax - (503) 646-0174
8:35 a.m.
On Mon, May 22, 2000 at 12:05:55PM -0400, Cini, Richard wrote:
Someone on-list did a design for his own clone, but I can't remember who it
is. I'm sure that it can be done cheaper than Tom Fischer's IMSAI-2 though.
I want to say it was Tim Shoppa's "TIMSAI", a few years ago.
Am I right Tim?
I seem to remember that his projected price was going to be pretty
close to Fischer-Freitas's cost for the IMSAI Series 2. It's very
expensive to do small runs, and do them well.
Rich
-Seth
--
"As a general rule, the man in the habit of murdering | Seth Morabito
bookbinders, though he performs a distinct service | sethm(a)loomcom.com
to society, only wastes his own time and takes no |
personal advantage." -- Kenneth Grahame (1898) | Perth ==> *
10:07 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 1222 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/20000522/7d2e68ed/at…
4:44 p.m.
please see my remarks below.
Dick
----- Original Message -----
From: Dwight Elvey <elvey(a)hal.com>
To: <classiccmp(a)classiccmp.org>
Sent: Monday, May 22, 2000 7:27 PM
Subject: Re[2]: Altair parts substitutions
ard(a)p850ug1.demon.co.uk (Tony Duell) wrote:
IIRC, that's the difference between the 2101 and the 2112. The 2112 has the
bidirectional bus.
In any case, the 5101 is (AFAIK) just a single-ported 256*4 CMOS RAM.
It's not a dual-ported device.
Hi
No, it is not dual ported but it has separate
input and output ports and not a bidi bus.
Dwight
1:43 p.m.
Rich writes:
The 8101 dual-port 256bx4 RAM is no longer available,
but Jameco has a 5101
which looks like a likely sub.
Check the timing specs; the 5101 might be slower.
Also, 8101 is just a different marketing designation for 2101, just like
8702 for 1702, etc. The 2101 should be easier to find.
Apparently Intel's marketing people didn't think that engineers would be
able to figure out that their standard RAM, ROM, PROM, and EPROM
products would work with the 8080, so they introduced 8xxx-series part
numbers for them.
2:40 p.m.
Forgive me for jumping in here, but the '373 is in no way similar to the
8T97. The 373 is a transparent latch with tristate outputs, i.e. a storage
element, whereas the 8T97 is a tristate buffer, in this case two sharing one
enable and the remaining four sharing the other. It's output is heftier
than the otherwise identical 74367. Though it's somewhat of an
oversimplification, the Signetics (that's who first made 'em) 8T series had
inputs pretty much like the LS parts, yet had outputs like the schottky
series parts.
If you're building an Altair work-alike, yet not bent on using the same
parts, I'd certainly recommend you use the 74ACT or even AC series if you
can get them. They have completely symmetrical outputs, i.e. drive upward
as hard as they pull downward, hence have more easily controllable
impedance, hence will make nicer waveforms on the backplane.
What's more, I'd recommend you look into the more up-to-date AC terminations
for the bus, as opposed to the rather poor active DC termination that was
promoted after people got the idea that there was trouble on the bus.
Almost any AC termination together with symmetrical bus drivers as I've
suggested, will improve the behavior of the bus much more than the passive
or active DC terminations that were popular in the '70's. You may find it's
easier to use 8-bit-wide parts for bus drivers/receivers, e.g. 74AC244's
instead of 8T97's. You get two 4-bit buffers per package, but they're
really better than the 8T97's. The fact that they're built in multiples of
4 bits wide makes them handier than the 6-bit 8T97's.
Good luck!
Dick
----- Original Message -----
From: Cini, Richard <RCini(a)congressfinancial.com>
To: 'ClassCompList' <classiccmp(a)classiccmp.org>
Sent: Monday, May 22, 2000 10:05 AM
Subject: Altair parts substitutions
It's that time of year again, when I think about
designing an Altair/IMSAI
work-alike. Now that I have the ability to generate schematics/layouts, I
may actually try it.
Anyway, some parts have become scarce, so I would have to use substitutes.
For example, Jameco no longer carries the 8T97 buffer/driver. As I recall
the 8T97 is an LS373 without the inverter on the gate *and* is faster and
has a higher drive current. I've heard that the MC6887 is a sub, but I
can't
find a datasheet on it.
The 8101 dual-port 256bx4 RAM is no longer available, but Jameco has a
5101
which looks like a likely sub. I would forego this
type for a larger SRAM,
but I guess that it too would have to be dual-ported. I need to look at
the
schematics for an S100 memory board.
Someone on-list did a design for his own clone, but I can't remember who
it
is. I'm sure that it can be done cheaper than Tom
Fischer's IMSAI-2
though.
Rich
3:07 p.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 768 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/20000523/69f37bf1/at…
4:06 p.m.
Please see embedded remarks below.
Dick
----- Original Message -----
From: Tony Duell <ard(a)p850ug1.demon.co.uk>
To: <classiccmp(a)classiccmp.org>
Sent: Monday, May 22, 2000 6:07 PM
Subject: Re: Altair parts substitutions
>
> Forgive me for jumping in here, but the '373 is in no way similar to the
> 8T97. The 373 is a transparent latch with tristate outputs, i.e. a
storage
True...
> element, whereas the 8T97 is a tristate buffer, in this case two sharing
one
However, it is _possible_ to use a '373 as a 3-state buffer. You tie the
enable to whichever state (high?) makes it transparent (rather than
holding the last value), apply the inputs to the D pins, put Q to the bus
and control it with the OE/ line.
Yes, but given that there are parts available that are more suitable, as
he's building a new device this time, I'd say it is probably preferable that
a tristate buffer be used instead of a transparent latch. I've seen this
sort of thing, but the advantage in using a '244, for example, is that it,
when used as a receiver, has schmidt-trigger inputs, which yields better
noise immunity, and when used as a driver, sources and sinks, in the CMOS
case, 24mA, while the LS has widely varying parameters, depending on the
maker, but you can rely on the output low (sink) current being at least
twice the output high (source) current.
I don't recomend this (the output drive capability is not that good for
one thing), but I did use it one in an emergency when I had to make an
input port for a microcomputer (a TRS-80 model 1 IIRC) in a hurry and I
didn't have any 3-state buffers in stock.
-tony
with CMOS bus buffers you needn't worry so much about the source and sink
current. The CMOS doesn't dc-load the bus, though it does put a bit of
AC-load in the form of capacitance on it. The great thing about AC
terminations is that they put a capcitance in the transient state in
parallel with the load capacitance. I'll let you work out the details, but
it does improve things, as anyone who's hung a 'scope on and watched a few
million transitions will tell you. They're not always textbook perfect, but
VERY close to it!
Dick
23 May
23 May
9:45 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 815 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/20000523/73048bd8/at…
8821
days inactive
8822
days old
10 comments
7 participants
participants (7)
-
ard＠p850ug1.demon.co.uk -
elvey＠hal.com -
eric＠brouhaha.com -
jimw＠agora.rdrop.com -
RCini＠congressfinancial.com -
richard＠idcomm.com -
sethm＠loomcom.com