15 Oct
2001
15 Oct
'01
8:13 a.m.
I'm looking for documentation for a 1963 minuteman missile navigation
computer. Any leads would be appreciated.
Thanks,
John Tinker
9:41 a.m.
That would be the Minuteman I if memory serves me. I know a few people that
were operators and repair techs on that system but I'm not sure if I can
still contact them. I will email you direct if I do contact them.
I don't think it's too weird to have a nav-puter for the old gear, just as
long as you don't have the reentry vehicle to go with it.
-> -----Original Message-----
-> From: owner-classiccmp(a)classiccmp.org
-> [mailto:owner-classiccmp@classiccmp.org]On Behalf Of Gene Buckle
-> Sent: Monday, October 15, 2001 12:59 PM
-> To: classiccmp(a)classiccmp.org
-> Subject: Re: looking for documentation for 1963 minuteman missile
-> computer
->
->
-> > I'm looking for documentation for a 1963 minuteman missile navigation
-> > computer. Any leads would be appreciated.
-> >
-> *spits coffee on monitor*
->
-> You're kidding, right?
->
-> g.
->
->
->
11:02 a.m.
On Mon, 15 Oct 2001 jtinker(a)coin.org wrote:
I'm looking for documentation for a 1963 minuteman
missile navigation
computer. Any leads would be appreciated.
Thanks,
John Tinker
I have a little info if it's the D17B (1962)
(from "Proceedings of the minuteman computer users group" Tulane Univ
1973)
Type: Serial synchronous
Number system: Fractional binary fixed point 2's complemant
Logic levels: 0V = false, -10V = true
Data word length: 11 or 24 (double precision)
Instruction length: 24 bits
Intructions: 39
Maximum I/O words/Second: 25600
Execution times: add ~78 uSec, multiply ~546 uSec, Double prec multiply
~1015 uSec
Memory: 24 bits + 3 timing bits
Type: Oxide coated disk
cycle time ~78 uSec
Capacity 5454 words 2727 double prec (24 bit).
I/O: 43 digital lines, 31 pulse type, 12 analog (from 3 8 bit A-D)
program I/O 800 5 bit c/s
Size: 20" high x 29" diameter
Power 28VDC +- 1V @ 17 A
Circuits: DTL+DRL double copperclad gold plated glass fiber laminate
with polyurethane conformal coat
The newer D37C (1964) is mostly the same except has a larger disk and more
instructions
Peter Wallace
Mesa Electronics
5:14 p.m.
New subject: looking for documentation for 1963 minuteman missile computer
At 01:02 PM 10/15/01 -0700, you wrote:
I have a little info if it's the D17B (1962)
(from "Proceedings of the minuteman computer users group" Tulane Univ
1973)
Number system: Fractional binary fixed point 2's complemant
Makes absolute sense in order to maximize the throuput.
Logic levels: 0V = false, -10V = true
Data word length: 11 or 24 (double precision)
Execution times: add ~78 uSec, multiply ~546 uSec, Double prec multiply
~1015 uSec
It probably used 24 bit accumulation... I'm impressed. However,
that's way too slow to implement any modern scheme (several orders
of magnitude too slow).
It probably used a mixture of analog/digital control; analog
for the single-variable, high bandwidth servoing; digital for
the higher hierarchy control.
Carlos.
--------------------------------------------------------------
Carlos E. Murillo-Sanchez carlos_murillo(a)nospammers.ieee.org
6:55 p.m.
New subject: looking for documentation for 1963 minuteman missile computer
Most of the older AF missile equipment had an A/D and D/A convertor, usually
sent "discretes" out to the servos, servopositioners, accelerometers, etc.
-> -----Original Message-----
-> From: owner-classiccmp(a)classiccmp.org
-> [mailto:owner-classiccmp@classiccmp.org]On Behalf Of Carlos Murillo
-> Sent: Monday, October 15, 2001 9:14 PM
-> To: classiccmp(a)classiccmp.org
-> Subject: Re: looking for documentation for 1963 minuteman missile
-> computer
->
->
-> At 01:02 PM 10/15/01 -0700, you wrote:
-> >I have a little info if it's the D17B (1962)
-> >
-> >(from "Proceedings of the minuteman computer users group" Tulane Univ
-> >1973)
-> >Number system: Fractional binary fixed point 2's complemant
->
-> Makes absolute sense in order to maximize the throuput.
->
-> >Logic levels: 0V = false, -10V = true
-> >
-> >Data word length: 11 or 24 (double precision)
-> >Execution times: add ~78 uSec, multiply ~546 uSec, Double prec multiply
-> >~1015 uSec
->
-> It probably used 24 bit accumulation... I'm impressed. However,
-> that's way too slow to implement any modern scheme (several orders
-> of magnitude too slow).
->
-> It probably used a mixture of analog/digital control; analog
-> for the single-variable, high bandwidth servoing; digital for
-> the higher hierarchy control.
->
-> Carlos.
->
-> --------------------------------------------------------------
-> Carlos E. Murillo-Sanchez carlos_murillo(a)nospammers.ieee.org
->
->
8295
days inactive
8295
days old
6 comments
6 participants
participants (6)
-
cmurillo＠emtelsa.multi.net.co -
fernande＠internet1.net -
geneb＠deltasoft.com -
jtinker＠coin.org -
pcw＠mesanet.com -
rhblakeman＠kih.net