29 Nov
2014
29 Nov
'14
8:40 p.m.
Chuck Guzis wrote:
"It didn't start with the 170, did it Billy? I remember a nice thick mat
of twisted-pair wires on the backplanes of 6000s, as well as in a lot of
attached controllers. Taper pin technology."
You're right - the 170 was the first use of main frame wire wrap at CDC. I've
never seen any other machine that used twisted pair wire wrap. When I left the project,
they were trying to defelop a machine to automate the wiring. It was not going well.
The earlier 6000, 7000 machines used twited pair with taper pins. Their wire mats were
huge, more than 10 inches thick in some places. That was one of the problems they were
trying to solve.
Billy
29 Nov
29 Nov
9:12 p.m.
New subject: EDSAC lives
The National Museum of Computing unveils EDSAC re-creation:
http://www.bbc.com/news/technology-30131447
- John
10:27 p.m.
On 11/29/2014 08:40 PM, Billy Pettit wrote:
You're right - the 170 was the first use of main
frame wire wrap at
CDC. I've never seen any other machine that used twisted pair wire
wrap. When I left the project, they were trying to defelop a machine
to automate the wiring. It was not going well.
I've got a couple of Eurcard 10K ECL boards with twisted pair all over
the place. Seems to be some sort of 12-bit CPU, from the looks of things.
There was also some sort of stapled-wire technology that was automated.
I don't recall the name of it.
The earlier 6000, 7000 machines used twited pair with
taper pins.
Their wire mats were huge, more than 10 inches thick in some places.
That was one of the problems they were trying to solve.
Penlight held in the mouth, go digging into the mat with both hands...
Fun.
--Chuck
30 Nov
30 Nov
8:59 a.m.
On 11/30/2014 12:27 AM, Chuck Guzis wrote:
There was also some sort of stapled-wire technology that
was automated. I don't recall the name of it.
There was something I think was called stitch bond. Wire
that looked very much like
wire-wrap wire, but probably had thinned insulation was
ultrasonically bonded to
flat posts on the board. I have some military-grade gear
with flat pack chips that
used this much like one would make wire-wrap cards. But,
they were very compact
and hardly thicker than a standard PC board.
Then, there was the system that came before wire-wrap. It
had rectangular
posts on card edge connectors, and I think the wire was
stranded. There was
a "gun" that had a roll of little tin-plated clips. You
stuck the wire into the
gun, and squeezed the handle. It drove a clip onto the pin,
pinching the
wire through the insulation. I've forgotten the name of
this system, too.
This was used for connecting up standard logic boards into a
backplane,
not for use at the chip level.
Jon
9:47 a.m.
On 11/30/2014 08:59 AM, Jon Elson wrote:
There was something I think was called stitch bond.
Wire that looked
very much like
wire-wrap wire, but probably had thinned insulation was ultrasonically
bonded to
flat posts on the board. I have some military-grade gear with flat pack
chips that
used this much like one would make wire-wrap cards. But, they were very
compact
and hardly thicker than a standard PC board.
This, I think. Back in the 60s it was advertised somewhat in trade
rags. My impression was that it saw use in military applications, but
that was just an impression. Apparently, very high writing density
could be achieved.
--Chuck
12:19 p.m.
On 2014-Nov-30, at 8:59 AM, Jon Elson wrote:
Then, there was the system that came before wire-wrap. It had rectangular
posts on card edge connectors, and I think the wire was stranded. There was
a "gun" that had a roll of little tin-plated clips. You stuck the wire into
the
gun, and squeezed the handle. It drove a clip onto the pin, pinching the
wire through the insulation. I've forgotten the name of this system, too.
This was used for connecting up standard logic boards into a backplane,
not for use at the chip level.
Jon
Would that be these?:
http://www3.telus.net/~bhilpert/tmp/HP2116C/crimps.jpg
http://www3.telus.net/~bhilpert/tmp/HP2116C/backplane.jpg
It's the backplane of an HP2116C.
I've always wondered what the tradename for the technique was.
Looks like in these machines the wire was stripped before crimping though.
12:28 p.m.
New subject: HP engineering excellence - was Re: Cyber ECL Wiring
On 30/11/14 3:19 PM, Brent Hilpert wrote:
On 2014-Nov-30, at 8:59 AM, Jon Elson wrote:
Lovely. One can see HP's famous devotion to engineering here.
--Toby
Then, there was the system that came before wire-wrap. It had rectangular
posts on card edge connectors, and I think the wire was stranded. There was
a "gun" that had a roll of little tin-plated clips. You stuck the wire into
the
gun, and squeezed the handle. It drove a clip onto the pin, pinching the
wire through the insulation. I've forgotten the name of this system, too.
This was used for connecting up standard logic boards into a backplane,
not for use at the chip level.
Jon
Would that be these?:
http://www3.telus.net/~bhilpert/tmp/HP2116C/crimps.jpg
http://www3.telus.net/~bhilpert/tmp/HP2116C/backplane.jpg
It's the backplane of an HP2116C.
I've always wondered what the tradename for the technique was.
Looks like in these machines the wire was stripped before crimping though.
1:07 p.m.
New subject: HP engineering excellence - was Re: Cyber ECL Wiring
On 2014-Nov-30, at 12:28 PM, Toby Thain wrote:
On 30/11/14 3:19 PM, Brent Hilpert wrote:
The 2116 is amazing even just from the mechanical engineering.
Looking at the backplane picture, the major frame pieces are cast aluminum.
Then notice the solid 1.5-by-0.5-inch machined aluminum rails which the backplane slides
out on.
Two of the rails split in half - you can see the stainless-steel alignment pins on the
ends of the rails, and the backplane swings open - the other two rails have integral
hinges. I should make a video showing this in action.
The rail structure is quite hidden, when I first got the machine it was quite perplexing
and I dismantled it from all angles trying to figure out how to get inside, before
realising what was going on.
On 2014-Nov-30, at 8:59 AM, Jon Elson wrote:
Lovely. One can see HP's famous devotion to engineering here.
Then, there was the system that came before wire-wrap. It had rectangular
posts on card edge connectors, and I think the wire was stranded. There was
a "gun" that had a roll of little tin-plated clips. You stuck the wire into
the
gun, and squeezed the handle. It drove a clip onto the pin, pinching the
wire through the insulation. I've forgotten the name of this system, too.
This was used for connecting up standard logic boards into a backplane,
not for use at the chip level.
Jon
Would that be these?:
http://www3.telus.net/~bhilpert/tmp/HP2116C/crimps.jpg
http://www3.telus.net/~bhilpert/tmp/HP2116C/backplane.jpg --Toby
>
> It's the backplane of an HP2116C.
> I've always wondered what the tradename for the technique was.
> Looks like in these machines the wire was stripped before crimping though.
>
2:28 p.m.
On 11/30/2014 02:19 PM, Brent Hilpert wrote:
Would that be these?:
http://www3.telus.net/~bhilpert/tmp/HP2116C/crimps.jpg
http://www3.telus.net/~bhilpert/tmp/HP2116C/backplane.jpg
It's the backplane of an HP2116C. I've always wondered
what the tradename for the technique was. Looks like in
these machines the wire was stripped before crimping though.
Yes, that's it.
No, I am PRETTY sure the tool stripped the
wire as it was
terminated. I actually worked at a place that used this
system, that was
in 1969. I believe the system came from AMP, but i could be
wrong
on that.
Jon
1 Dec
1 Dec
7:11 p.m.
On 2014-Nov-30, at 2:28 PM, Jon Elson wrote:
On 11/30/2014 02:19 PM, Brent Hilpert wrote:
Right, I was just distinguishing between the wire being stripped vs having the crimp bite
through the insulation.
Looking closely at them, there's actually two sections to the crimp, a short bit that
clamps the insulation and a longer section for the metal-to-metal contact.
I haven't experienced any problems with them, thankfully.
Would that be these?:
http://www3.telus.net/~bhilpert/tmp/HP2116C/crimps.jpg
http://www3.telus.net/~bhilpert/tmp/HP2116C/backplane.jpg It's the backplane of an
HP2116C. I've always wondered what the tradename for the technique was. Looks like in
these machines the wire was stripped before crimping though.
Yes, that's it.
No, I am PRETTY sure the tool stripped the wire as it was
terminated. I actually worked at a place that used this system,
that was
in 1969. I believe the system came from AMP, but i could be wrong
on that.
I just looked, the edge connectors at least are indeed from AMP.
30 Nov
30 Nov
8:50 a.m.
On 11/29/2014 10:40 PM, Billy Pettit wrote:
Chuck Guzis wrote:
"It didn't start with the 170, did it Billy? I remember a nice thick mat
of twisted-pair wires on the backplanes of 6000s, as well as in a lot of
attached controllers. Taper pin technology."
You're right - the 170 was the first use of main frame wire wrap at CDC. I've
never seen any other machine that used twisted pair wire wrap.
DEC KL10B had a lot
of twisted pair wire-wrap on the
backplane. VAX 11/780 had
critical clock signals on twisted pair wire-wrap. These
were, in fact, ECL signals.
Jon
3434
days inactive
3436
days old
10 comments
6 participants
participants (6)
-
bpettitx＠comcast.net -
cclist＠sydex.com -
elson＠pico-systems.com -
hilpert＠cs.ubc.ca -
jfoust＠threedee.com -
toby＠telegraphics.com.au