7 Sep
2005
7 Sep
'05
1:53 p.m.
Does anyone have a datasheet for an 82S21? I checked all the obvious
places but can't find one.
82S21 - Signetics 32x2 SRAM
thanks!
-brad
7 Sep
7 Sep
2:27 p.m.
Does anyone have a datasheet for an 82S21? I checked all the obvious
places but can't find one.
82S21 - Signetics 32x2 SRAM
The best I can find is a pinout, and the following description : Open
Collector Outputs, 50ns access time, 130mA supply current. You already
know it's 32*2 bits, TTL.
Here's the pinout. I have no idea what some of the signals are for (does
it, for example, have separate write-enable inputs for the 2 bits?), but
it may be better than nothing.
1 : WE/
2 : WS(0)/
3 : DI(0)
4 : A(4)
5 : CE
6 : Strobe
7 : DO(0)
8 : Ground
9 : DO(1)
10 : A(3)
11 : A(2)
12 : A(1)
13 : A(0)
14 : DI(1)
15 : WS(1)
16 : Vcc
The NatSemi DM86S21 might be an equivalent, I can find no
cross-references to it in other memory databooks, though.
-tony
3:53 p.m.
Tony Duell wrote:
Here's the pinout. I have no idea what some of the signals are for (does
it, for example, have separate write-enable inputs for the 2 bits?), but
it may be better than nothing.
Thanks - I know the pinout. I think this ram would allow some sort of
simultaneous read/write (not dual port, just overlapped) but I don't
know how it works.
The NatSemi DM86S21 might be an equivalent, I can find
no
cross-references to it in other memory databooks, though.
Huh. I could not find anything on the 86S21, but I don't have national
books. Anyone else?
-brad
8 Sep
8 Sep
6:43 a.m.
The data sheet is in the 1976 Signetics Data Manual. I scanned the first
page (the second is signal timing diagrams):
http://www.jkearney.com/Misc/82s21-p1.pdf (255KB). Let me know if you want
the second page.
The book is hard-bound and I didn't have time to crop it, so it isn't a
great scan, sorry.
8:20 a.m.
1976 Signetics Data Manual (hardbound) - ahhh, if I
could have only one data manual for working on 70's
computers this would be it, a great resourse
--- Jim Kearney <jim at jkearney.com> wrote:
The data sheet is in the 1976 Signetics Data Manual.
I scanned the first
page (the second is signal timing diagrams):
http://www.jkearney.com/Misc/82s21-p1.pdf (255KB).
Let me know if you want
the second page.
The book is hard-bound and I didn't have time to
crop it, so it isn't a
great scan, sorry.
______________________________________________________
Click here to donate to the Hurricane Katrina relief effort.
http://store.yahoo.com/redcross-donate3/
9:40 a.m.
At 08:20 AM 9/8/05 -0700, Jim wrote:
1976 Signetics Data Manual (hardbound) - ahhh, if I
could have only one data manual for working on 70's
computers this would be it, a great resourse
I have a PB copy of about that vintage if someone wants to scan it. It's
at the warehouse so I don't have it in front of me but it has all of those
old PROMs in it. I also have a PB copy of the 1971 Signetics "Digital 8000
Series TTL/MSI" databook and 1972 Signetics MOS Silicon Gate 2500 Series /
Metal Gate 2000/2400 Series Data Book and 1978 Harris Bipolar CMOS Memory
Databook. I have more old memory data books but they're not here at the
moment.
Joe
--- Jim Kearney <jim at jkearney.com> wrote:
The data sheet is in the 1976 Signetics Data
Manual.
I scanned the first
page (the second is signal timing diagrams):
http://www.jkearney.com/Misc/82s21-p1.pdf (255KB).
Let me know if you want
the second page.
The book is hard-bound and I didn't have time to
crop it, so it isn't a
great scan, sorry.
______________________________________________________
Click here to donate to the Hurricane Katrina relief effort.
http://store.yahoo.com/redcross-donate3/
10:47 a.m.
Tony Duell wrote:
Quite likely, although the control signals look unconventional. I would
guess you can apply a 5 bit address and read the contents of that
location one the DO lines. You may have to pulse the strobe line, or
something like that. Presumably you can also provide new data on the DI
lines and write it into that location.
What device uses this RAM, and what are the control pins conencted to?
-tony
Here's the pinout. I have no idea what some of the signals are for (does
it, for example, have separate write-enable inputs for the 2 bits?), but
it may be better than nothing.
Thanks - I know the pinout. I think this ram would allow some sort of
simultaneous read/write (not dual port, just overlapped) but I don't
know how it works.
12:22 p.m.
Tony Duell wrote:
Quite likely, although the control signals look unconventional. I would
It turns out it's write-while-read ram (no doubt you've caught up on your
email by now and read the pdf which was posted).
The strobe signal clocks the ram data into a latch. This is available
for some time, even if you do a write.
What device uses this RAM, and what are the control
pins conencted to?
It's used in the MIT CADR lisp machine, for the m-memory, which is a
high speed scratch pad used by the microcode.
I wrote some programs to convert the original CAD files (in SUDS) into
verilog and I've been (slowly) working on getting them to simulate. I
want to boot one under simulation and eventually put one in a simple
fpga (hah, by the time I'm 75 at this rate). I figure they'll make nice
christmas gifts :-)
I've got it fetching microcode from the prom. Most of the problems have
been with 74 style 'parts' I have made - in some cases I had to guess
because I have no data sheet but I'm slowly fixing that.
Also iverilog and cver both seem to have some small issues but mostly
work. I've used modelsim too but I perfer to run on linux.
-brad
1:57 p.m.
What device
uses this RAM, and what are the control pins conencted to?
It's used in the MIT CADR lisp machine, for the m-memory, which is a
high speed scratch pad used by the microcode.
I wrote some programs to convert the original CAD files (in SUDS) into
verilog and I've been (slowly) working on getting them to simulate. I
want to boot one under simulation and eventually put one in a simple
fpga (hah, by the time I'm 75 at this rate). I figure they'll make nice
My experience with simulators suggests that's a massive _underestimate_
of the time it would take.
christmas gifts :-)
I've got it fetching microcode from the prom. Most of the problems have
been with 74 style 'parts' I have made - in some cases I had to guess
because I have no data sheet but I'm slowly fixing that.
FWIW< I have a reasonably complete set of TTL databooks from several
manufactueres. Let me know if there's any 74-series part you want me to
look up.
-tony
3:03 p.m.
Tony Duell wrote:
FWIW< I have a reasonably complete set of TTL
databooks from several
manufactueres. Let me know if there's any 74-series part you want me
to look up.
I have a brace of CDROMs from TI on the shelf which
have datasheets for many (possibly all) of their TTL
parts (and a bunch of their analog stuff too). I
expect that, unless you want some really obscure
stuff, it's all on there and, I would expect, on
their website too.
Right now, the two CDROMs are easier for me to reach
than the half-dozen or so TI databooks too :-)
Antonio
--
Antonio carlini
arcarlini at iee.org
3:30 p.m.
a.carlini at ntlworld.com wrote:
I have a brace of CDROMs from TI on the shelf which
have datasheets for many (possibly all) of their TTL
parts (and a bunch of their analog stuff too). I
expect that, unless you want some really obscure
stuff, it's all on there and, I would expect, on
their website too.
Ummm what about a 7480 :)
Right now, the two CDROMs are easier for me to reach
than the half-dozen or so TI databooks too :-)
I find most of the time I need a data sheet I have to
surf the web.
Antonio
Too bad I find out about all the neat
parts that
are going bye-bye. Now is the last chance to get
any of the old TTL before they go *POOF*!
3:29 p.m.
Ummm what about a 7480 :)
A gated full adder? (I had to look it up...). I have a data sheet which
even includes a transistor-level schematic.
Right now, the two CDROMs are easier for me to
reach
than the half-dozen or so TI databooks too :-)
I find most of the time I need a data sheet I have to
surf the web.
4:23 p.m.
New subject: datasheets for 82S21 - Signetics 32x2 SRAM? : pdp2
Tony Duell wrote:
Hmm that is not web ... drat .
I am guessing that is about 15 ns delay.
I am using CPLD's rather than TTL since I don't have get BIG
PCB's made up, as well as the CPLD language is simple logic
equations rather than * Your Favorite big $$$ hardware design
language * as with FPGA's, but it nice to compare what real
TTL speeds are.
Ummm what about
a 7480 :)
A gated full adder? (I had to look it up...). I have a data sheet which
even includes a transistor-level schematic.
Prolbems are that I can't access the web or read CD-ROMs at my workbench
where I do most of my designing (I prefer to design with real components
and a soldering iron), and anyway, it's a lot easier to flip through a
databook looking for interesting chips than to display a pdf (every
machine I've ever used, including fast PCs owned by friends of mine, take
a noticeable time to display a page from a pdf file. A lot longer than it
takes me to glance at a page and see if it could possibly include the
information I am looking for).
While surfing the web for a better way to do a ALU I found a
hardware design for a Pdp 2. http://www.locl.net/homes/pdp2/
While I think using 74XX's is better for vintage hardware so you
have more a idea of realistic timing I wish him good luck.
-tony
9 Sep
9 Sep
5:03 p.m.
New subject: datasheets for 82S21 - Signetics 32x2 SRAM? : pdp2
Tony Duell wrote:
Hmm that is not web ... drat .
I am guessing that is about 15 ns delay.
It would seem to be a lot slower than that...
I have (typical/max in ns)
t(PLH) (Cn to Cn+1/) 13/17
t(PHL) (Cn to Cn+1/) 8/12
t(PLH) (Bc to Cn+1/) 18/25
t(PHL) (Bc to Cn+1/) 38/55
t(PLH) (Ac to S) 52/70
t(PHL) (Ac to S) 62/80
t(PLH) (Bc to S/) 38/55
t(PHL) (Bc to S/) 56/75
t(PLH) (A1 to A*) 48/65
t(PHL) (A1 to A*) 17/25
t(PLH) (B1 to B*) 48/65
t(PHL) (B1 to B*) 17/25
The A*, B* outputs laoded with 15pf, the others loaded with the standard
TTL load network.
That's for a plain 7480, I can't find any other versions.
-tony
Ummm what
about a 7480 :)
A gated full adder? (I had to look it up...). I have a data sheet which
even includes a transistor-level schematic.
8 Sep
8 Sep
3:49 p.m.
On Thu, 08 Sep 2005 16:30:46 -0600
woodelf <bfranchuk at jetnet.ab.ca> wrote:
a.carlini at ntlworld.com wrote:
That's true, but there are decades worth of NOS TTL chips on eBay. I've
gotten a bunch of them recently of the types I think I'll ever use. And
really cheap at that.
I have a brace of CDROMs from TI on the shelf
which
have datasheets for many (possibly all) of their TTL
parts (and a bunch of their analog stuff too). I
expect that, unless you want some really obscure
stuff, it's all on there and, I would expect, on
their website too.
Ummm what about a 7480 :)
Right now, the two CDROMs are easier for me to
reach
than the half-dozen or so TI databooks too :-)
I find most of the time I need a data sheet I have to
surf the web.
Antonio
Too bad I find out about all the
neat parts that
are going bye-bye. Now is the last chance to get
any of the old TTL before they go *POOF*!
6:08 p.m.
That's true, but there are decades worth of NOS
TTL chips on eBay. I've
gotten a bunch of them recently of the types I think I'll ever use. And
really cheap at that.
And after those decades worth are gone, a few more decades worth will turn
up with just a little looking. We will all be dead when the things become
unobtainium.
I am always getting into arguments with folks that are worried that the
1960s transistors used in early minicomputers (PDP-1s, Packard Bell 250,
and the like) are "all gone" or "extremely rare". 'Taint true,
Dooley -
all you have to do is look.
William Donzelli
aw288 at osfn.org
9 Sep
9 Sep
1:37 a.m.
William Donzelli wrote:
Concerning transistors of the 60s (even Germanium-type PNPs), I wouldn't
have a problem to use modern types
as replacements if the go or no-go of an old system would depend on it.
Discrete electronic components are rarely a
problem, even coils and transformers could be replaced. More difficult
are certain kinds of integrated circuits.
The mentioned 7480 is maybe hard to locate even as a datasheet (good
idea to keep old TI TTL databook, IIRC
also at bitsavers), and even more difficult to locate the chip itself,
but this is a function that would easily fit into a GAL.
A real problem are species what can't be simply replaced by some GAL,
namely open collector components or "extendable"
gates like the 7450. Replacing a single chip of such a kind usually
means replacing its environment as well, e.g. that 7450 would
then require to pull the extension gate (was it 7460, my memory is
failing) as well.
In the area of unobtainium, however, you usually don't find SSI or MSI
circuits, but rather nasty special analog circuits and
early LSI MOS circuits. Often a modern uC could emulate the latter, but
this often needs significant tinkering for each problem,
and an many cases also rewiring its environment.
Surely the question original replacement vs. reengineered functionality
has some religious aspect, but if it is the question of whether
an old system should be left inoperable in a museum in contrast to
actually switching its blinkinligths I favor reengineering.
Regards
Holger
That's
true, but there are decades worth of NOS TTL chips on eBay. I've
gotten a bunch of them recently of the types I think I'll ever use. And
really cheap at that.
And after those decades worth are gone, a few more decades worth will turn
up with just a little looking. We will all be dead when the things become
unobtainium.
I am always getting into arguments with folks that are worried that the
1960s transistors used in early minicomputers (PDP-1s, Packard Bell 250,
and the like) are "all gone" or "extremely rare". 'Taint true,
Dooley -
all you have to do is look.
5:37 p.m.
The mentioned 7480 is maybe hard to locate even as a
datasheet (good
idea to keep old TI TTL databook, IIRC
Why do you think I keep (and actively grab) all old databooks :-)
also at bitsavers), and even more difficult to locate
the chip itself,
but this is a function that would easily fit into a GAL.
A real problem are species what can't be simply replaced by some GAL,
namely open collector components or "extendable"
gates like the 7450. Replacing a single chip of such a kind usually
I hate to tell you this, but the 7480 will not fit into a GAL. The A* and
B* pins are effectively open-collector I/O with intenral (6k) pull-ups.
You can extend the input gating via that pin, or you can monitor the
output of the A1/A2 NAND gate there.
means replacing its environment as well, e.g. that
7450 would
then require to pull the extension gate (was it 7460, my memory is
failing) as well.
Great fun if it happens to be on another board (did this ever happen?)
Surely the question original replacement vs.
reengineered functionality
has some religious aspect, but if it is the question of whether
an old system should be left inoperable in a museum in contrast to
actually switching its blinkinligths I favor reengineering.
Me too. Computers that don't work are not interesting. That said, I am in
favour of doing as few mods as possible to the original machine. For
example, I'd fit a turned-pin socket in place of the original chip and
wire up a heeader to the replacement chip (maybe using stripboard or a
PCB to make a plug-in module). That way, if I ever get the real part, I
can go back to the original design.
-tony
8 Sep
8 Sep
4:02 p.m.
woodelf wrote:
Ummm what about a 7480 :)
Nope, although I did have fun trying to get the
app to run :-) The directory with the datasheets
has hundreds of sheets but there seems to be no
obvious way of relating a filename to a device
type :-(
I found the SN7485 and SN7486 but no SN7480 (or
at least no datasheet).
Is it used in anything I'm likely to care about?
Is there a datasheet for it on the web?
Antonio
--
Antonio carlini
arcarlini at iee.org
3:46 p.m.
woodelf wrote:
> Ummm what about a 7480 :)
[...]
I found the SN7485 and SN7486 but no SN7480 (or
at least no datasheet).
Is it used in anything I'm likely to care about?
Well, it's obscrue enough that I had to look it up, and I've worked on
many TTL-based devices. I don't think I've ever seen one.
Is there a datasheet for it on the web?
Dunno. If anyone needs it, I do have the datasheet, and would find some
way to get the information to you if you were trying to repair a classic
computer or something like that.
-tony
6:04 p.m.
I found the SN7485 and SN7486 but no SN7480 (or
at least no datasheet).
7480s were made by several manufacturers, but it was never a hugely
popular chip - sort of in the "second-string" functions of early 74xx TTL.
There are some real oddballs - 7459 is one (there was a thread on this a
few yars ago). I forget what it is - and I am not going to looking it up
again.
William Donzelli
aw288 at osfn.org
6:54 p.m.
On Thu, 8 Sep 2005 21:04:34 -0400 (EDT)
William Donzelli <aw288 at osfn.org> wrote:
Has anybody ever tried to compile a complete, numerical list of the
functions or even copies of the datasheets, for the entire 74xx series?
Who was the issuer of the numbers who determined the sequence?
I used to (it's not as easy to find anymore) love going into surplus
stores and picking up random boards full of TTL, and figure out what the
board was just by the TTL gate part numbers.
I found the
SN7485 and SN7486 but no SN7480 (or
at least no datasheet).
7480s were made by several manufacturers, but it was never a hugely
popular chip - sort of in the "second-string" functions of early 74xx
TTL.
There are some real oddballs - 7459 is one (there was a thread on this
a few yars ago). I forget what it is - and I am not going to looking
it up again.
9 Sep
9 Sep
5:06 p.m.
There are some real oddballs - 7459 is one (there was
a thread on this a
7456 and 7457, which are /50 and /60 counters (to get 1Hz from european
and US mains). AFAIK they're the only TTL chips in 8 pin packages.
And the 74S262, which is a teletext-compatible character generator ROM.
Very few databooks list that one.
-tony
14 Sep
14 Sep
12:35 p.m.
I wrote:
I found the SN7485 and SN7486 but no SN7480 (or
at least no datasheet).
I still have no reason to care about the SN7480
(nor the '57 and friends that Tony mentioned later
in this thread) but I gave up on the TI CDROM and
looked at some paper instead. All of the above
(and many more) are in the 1985 TTL databook(s).
I guess TI rip out the stuff they no longer
sell from their web site (and the CDs that they
presumably produce from it).
Antonio
--
Antonio carlini
arcarlini at iee.org
8 Sep
8 Sep
3:34 p.m.
Tony Duell wrote:
My experience with simulators suggests that's a massive _underestimate_
of the time it would take.
Well, I did manage to make a microcode simulator and it boots the lisp
machine.
I think there are a few very subtle bugs in the pipe, however, and I
think the only way I'll find them is to watch the real hardware work.
I think the timing on this machine, and the 74S/74LS parts it uses are
slow enough that verilog will do fine. In reality what I am saying is
not about verilog, but the accuracy of the models I have created. So,
restated, I think I can make models which are accurate enough for this
type of design.
heh. we'll see.
-brad
8 Dec
8 Dec
12:15 p.m.
Just catching up with these posts :>)
My old Quarndon cat. gives it as a Write while Read RAM with o/c O/Ps. 32x2.
It has a logic level schematic ( quite small ) if that's any use- I can get
it scanned and enlarged.
Geoff.
----- Original Message -----
From: "Brad Parker" <brad at heeltoe.com>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Sent: Wednesday, September 07, 2005 10:53 PM
Subject: Re: datasheets for 82S21 - Signetics 32x2 SRAM?
Tony Duell wrote:
--
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.371 / Virus Database: 267.13.13/195 - Release Date: 08/12/05
Here's the pinout. I have no idea what some of the signals are for (does
it, for example, have separate write-enable inputs for the 2 bits?), but
it may be better than nothing.
Thanks - I know the pinout. I think this ram would allow some sort of
simultaneous read/write (not dual port, just overlapped) but I don't
know how it works.
The NatSemi DM86S21 might be an equivalent, I can
find no
cross-references to it in other memory databooks, though.
Huh. I could not find anything on the 86S21, but I don't have national
books. Anyone else?
-brad
--
No virus found in this incoming message.
Checked by AVG Anti-Virus.
Version: 7.0.344 / Virus Database: 267.10.18/91 - Release Date: 06/09/05
6721
days inactive
6813
days old
26 comments
11 participants
participants (11)
-
arcarlini＠iee.org -
ard＠p850ug1.demon.co.uk -
aw288＠osfn.org -
bfranchuk＠jetnet.ab.ca -
brad＠heeltoe.com -
chenmel＠earthlink.net -
geoffreythomas＠onetel.com -
gkicomputers＠yahoo.com -
holger.veit＠ais.fraunhofer.de -
jim＠jkearney.com -
rigdonj＠cfl.rr.com