26 Nov
2000
26 Nov
'00
5:32 a.m.
Serial number appears to be 4469. I got it from John Wilson, moved it 1000
miles to home. It looks too heavy to drag up the stairs so chances are really
good it'll stay in the (heated) garage. We also got a TU45 and a spare formatter
for it. (Question, can I just plug any Pertec Unformatted 1600/800 drive into
that and make it work, or only special Pertec devices?)
Current status of the machine is unknown - It's never been powered on.
We'll be cleaning it up, taking lots of pictures, and checking things out
before it gets powered on. But that should be sometime soon. In any case,
I'll have pictures posted before the week's out.
And now, I'm going to do a jig and generally behave like a child until the
high wears off :) More information to come.
-------
10:37 a.m.
On Sun, 26 Nov 2000, Mark Crispin wrote:
On Sun, 26 Nov 2000 11:32:17 -0800, Daniel A.
Seagraves wrote:
The number is printed on the top rear of the rack, I have a KS10-AA, the
serial number as printed is MRR4469. This is also printed in several
places on stickers affixed to other places in the rack.
Right now, I've finished making sure there are no obvious mechanical
faults (I.E. making sure I didn't destroy it moving it), pulling all the
various boards out, making sure the chips on them are well seated, etc.
Then I might just try turning it on. (The top of the Mighty Mite is rather
scary looking and I'd rather not go poking a voltmeter in there unless I
have to, lest I bridge two terminals or something and destroy it.)
Will it be unhappy with my if I plug all these 120 AC plugs (I found 3 so
far, one for the fan at the base of the rack, one for the KS logic, and
one for the UNIBUS box) into a normal powerstrip instead of DEC's?
I don't have the proper plug for the end of the power controller, and this
is what I do for my PDP-11s anyway.
Does anyone have the jumper settings for the console board? I'l like to
not have to try guessing what speed/stops/bits/etc. it's set to.
Oh, and I did neglect to mention that the machine was earmarked for Gordon
Greene, who graciously let me snag the machine out from underneath him.
You have my eternal thanks, I didn't mean to leave you out.
"Confuse, annoy, and DEE-STROY!" -- Jet Wolf | "Nothing Happens." --
ADVENT
"You'd be surprised what you can live through..." -- Anonymous
"...A man can pass his family and his name down through his sons, but it's
his honour that gets passed through his daughters. He can see the best
and worst of life in his girls. A daughter is something far too precious,
and he'll do anything to protect her."
-- Reichsfuehrer Siegfried Koenig, _Matrose_Mond_, David Oliver
Serial number appears to be 4469.
I'd be interested in you can verify that. My first KS is 4464, which I had
thought was the last one made. 
2:11 p.m.
On Sun, 26 Nov 2000 18:37:11 -0600 (CST), Daniel Seagraves wrote:
You have to rewire jumpers on the RH11. Cut some, add others.
The number is printed on the top rear of the rack, I
have a KS10-AA, the
serial number as printed is MRR4469. This is also printed in several
places on stickers affixed to other places in the rack.
Ah. You do indeed have KS #4469, but I had a brain fade when comparing it. I
have KS #4664.
Will it be unhappy with my if I plug all these 120 AC
plugs (I found 3 so
far, one for the fan at the base of the rack, one for the KS logic, and
one for the UNIBUS box) into a normal powerstrip instead of DEC's?
Yes, and you only need to provide power for the fan and the KS backplane if
you don't need the tape drive or terminals (the CTY and KLINIK are off of the
KS backplane). You do have to fix a bug in TOPS-20 to keep it from crashing
if no DZs respond.
Speaking of
which, what actually told the RH11 that it was an 18 bit
transfer? Piddling a specific register? I've got some UC12's around here
somewhere, it would be simple to test em if I knew which bit to flip.
I have no
idea, myself. You'd have to look it up.
2:40 p.m.
On Sun, 26 Nov 2000, Mark Crispin wrote:
Ah. You do indeed have KS #4469, but I had a brain
fade when comparing it. I
have KS #4664.
Ah, OK.
Yes, and you only need to provide power for the fan
and the KS backplane if
you don't need the tape drive or terminals (the CTY and KLINIK are off of the
KS backplane). You do have to fix a bug in TOPS-20 to keep it from crashing
if no DZs respond.
You were right - It didn't like me. I just went downstairs, plugged my
VT100 into the CTY port (I set it for 9600 N81 as just a guess), plugged
the KS backpanel and it's fan into a 20 amp powerstrip, bridged pins 1 and
3 on the UNIBUS box's DEC power controller plug (to let you turn the box
on without the controller at the bottom of the rack - I had to do this
trick for awhile to turn on the second BA11 in my 11/44), plugged the
UNIBUS in, and let 'er rip.
The Mighty-Mite whines at me (but it's not as loud as my 34A whines, it's
just there), some LED's above the KS boards light, the POWER light comes
on, a light comes on in the bottom of the UNIBUS box, the fans all come
on, and the machine just sits there. I checked the serial cable, it's
good, and I noticed that if I unplug the Vt100 from the CTY port and push
BOOT, the FAULT light comes on. I can clear the light by pushing RESET.
The power switch on the frontpanel doesn't do anything, and, if the BOOT
or RESET switches are pressed while the CTY is connected, they don't do
anything. So obviously it wants either the missing RH11 board (I think I
have a spare, I'll check tomorrow) or something inside the power
controller.
Since I can leave the BA11 off, I'm going to unplug it and pull the board
from it's RH11 and shove it in the KS backpanel,
and we'll see what
happens.
"Confuse, annoy, and DEE-STROY!" -- Jet Wolf | "Nothing Happens." --
ADVENT
"You'd be surprised what you can live through..." -- Anonymous
"...A man can pass his family and his name down through his sons, but it's
his honour that gets passed through his daughters. He can see the best
and worst of life in his girls. A daughter is something far too precious,
and he'll do anything to protect her."
-- Reichsfuehrer Siegfried Koenig, _Matrose_Mond_, David Oliver
3:18 p.m.
One last question, and them I'm done for tonight.
The Mighty-Mite's whining apparently is an alarm or something - because
it will shut itself off after more or less 5 minutes. There are
temperature sensors above the cardcage - I've heard you get this sort of
misbehaviour if these go bad and continuously tell the PSU that it is over
temperature. How can I test for and or eliminate this? The CPU boards
aren't getting hot at all. (Not even warm - in fact, the blower air is
kinda cold.)
"Confuse, annoy, and DEE-STROY!" -- Jet Wolf | "Nothing Happens." --
ADVENT
"You'd be surprised what you can live through..." -- Anonymous
"...A man can pass his family and his name down through his sons, but it's
his honour that gets passed through his daughters. He can see the best
and worst of life in his girls. A daughter is something far too precious,
and he'll do anything to protect her."
-- Reichsfuehrer Siegfried Koenig, _Matrose_Mond_, David Oliver
2:58 p.m.
Took the missing RH11 board from the UBA, added it in the KS backplane,
and unplugged the UBA. Booted the KS, and I have a KS10> prompt.
It wanted 1200 baud N81, if that helps anyone.
Now, to see if I have the missing board, get the normal DEC power
controller working, and put the skins on the machine so it looks nicer. :P
I took pictures of things, as soon as I get to work tomorrow I'll post
'em.
"Confuse, annoy, and DEE-STROY!" -- Jet Wolf | "Nothing Happens." --
ADVENT
"You'd be surprised what you can live through..." -- Anonymous
"...A man can pass his family and his name down through his sons, but it's
his honour that gets passed through his daughters. He can see the best
and worst of life in his girls. A daughter is something far too precious,
and he'll do anything to protect her."
-- Reichsfuehrer Siegfried Koenig, _Matrose_Mond_, David Oliver
3:33 p.m.
On Sun, 26 Nov 2000 22:58:42 -0600 (CST), Daniel Seagraves wrote:
It wanted 1200 baud N81, if that helps anyone.
You can set dip switches on the console board to change that.
The Mighty-Mite's whining apparently is an alarm
or something - because
it will shut itself off after more or less 5 minutes.
I've never experienced this, and it doesn't use any external overtemp sensors.
It costs several hundred bucks to get one of those Mighty-Mites repaired; it's
a lot cheaper to get a replacement from salvage. If you open one of them up,
you'll see why; unlike DEC power supplies a Mighty-Mite is a nightmare to
repair.
27 Nov
27 Nov
12:27 a.m.
Mark:
Hm. Regarding that bug, does T20 blow up immediately when it doesn't find
the DZ11's?
This might explain why I can't boot the system with the second UBA shut
down. I'll check it out tonight. Running the T20 diags is a bit boring :-)
Chris
1:27 p.m.
New subject: Fire up the iron...
Ok, it's fall, it's cool, and I can finally start working on AI again.
With the cold weather, I can pop open a window instead of fighting a
losing battle with an 8,000 BTU AC unit.
Quick bit of status: Sitting in my computer room here at the house is:
AI (20/20 system, fully loaded with memory, passes all diags)
RM03 (Fully operational, need to dust off and purge)
RM80 (Operational, however currently unformatted)
RM03 (Logic problem in getting the heads to load)
TU77 (Doesn't work, why?)
TU77 (Spare for parts)
I've got a pair of RM02's, a TU45, and an RP07 sitting in storage. The
TU45 is a bit of a shambles, maybe with Dan's system as a plate I can
diagnose it someday. The RM02's will work on a 20, bit slower than an
RM03, but same format. The RP07 is in God knows what kind of shape (the
lid for it disintegrated on Interstate 40...). A dream would be to get
that drive operational and running as the system disk.
Once running, the systems will move to a place where power and AC is not
an issue. Till then, they run at my house.
The big problem with AI is that the tape MASSBUS is dead. I don't have a
clue as to why. Diagnostics are very vague, closest thing I get on the
controller is a control bus parity error/8915 bit fiddler error.
According to the limited TOPS20 diags, this could be anything.
So in order to figure this out, I've spent the evening pulling out the
DZ11's, the chaos controller and the Micon Ethernet card. In it's place
I have installed a pdp11/23+ for running pdp11 diagnostics on the RH11.
Yes, I know: 11/23's are Q-Bus and the RH11 is a Unibus tank. However
I've also got a Qunivertor, and it seems to be working (albeit in 18 bit
mode). At least I can fire up RSX11/M and begin walkin through tests on
the TU77/TM03/RH11. And I don't have to deal with the main CPU, the 18
bit fiddler, etc...
Thus my question: Did the tape drives work on AI before it went offline.
Somehow I seem to recall that they didn't, and this might be due to a
bad RH11 MASSBUS controller (as I have swapped in 2 other TM03
controllers with the same lack o results).
CZ
28 Nov
28 Nov
4:09 a.m.
New subject: Fire up the iron...
The big problem with AI is that the tape MASSBUS is
dead. I don't have a
clue as to why. Diagnostics are very vague, closest thing I get on the
controller is a control bus parity error/8915 bit fiddler error.
According to the limited TOPS20 diags, this could be anything.
CZ
First -- this is a couple of points on quick and dirty massbuss
troubleshooting.
There are a number of problems that can cause MCPE (Massbuss Control
Parity Errors)...
The first thing is -- bad round bus cable with a bent pin and bad
contact.
The second can be bad flat cables from the three massbus transceivers
which also can have backplane seeting problems -- the double height
cards wiggle a lot. (common in TM03's and RPxx DCL's).
It also could be the 8915 bit fiddler -- but the failure sounds
like the equivalent of the TM03 wrap the world data wrap through the
controller.
--Bill
--
bpechter(a)monmouth.com | FreeBSD since 1.0.2, Linux since 0.99.10
| Unix Sys Admin since Sys V/BSD 4.2
| Windows System Administration: "Magical Misery Tour"
30 Nov
30 Nov
12:33 a.m.
New subject: Fire up the iron...
On Mon, 27 Nov 2000, Christopher Zach wrote:
Quick bit of status: Sitting in my computer room here
at the house is:
AI (20/20 system, fully loaded with memory, passes all diags)
RM03 (Fully operational, need to dust off and purge)
RM80 (Operational, however currently unformatted)
RM03 (Logic problem in getting the heads to load)
TU77 (Doesn't work, why?)
TU77 (Spare for parts)
Question, what did AI originally have for system disks? (I.E. do you have
the old AI disk packs or are you restoring from a tape or something?)
I've got a pair of RM02's, a TU45, and an RP07
sitting in storage. The
TU45 is a bit of a shambles, maybe with Dan's system as a plate I can
diagnose it someday.
I replaced the line cord ending on my TU45 last night so I could turn it
on. It doesn't do a whole lot. It makes an impressive amount of noise,
one of the fans needs a can of WD-40 really badly. :P I think it's messed
up though, I tried getting it to mount a tape and all it does is make a
mess unspooling the tape all over the place. When you push LOAD, it runs
the take-up reel backwards slowly for a few seconds, then kicks both reels
forwards (with the tape now loose on the takeup reel) and stops,
unspooling the first 20 feet or so of tape around inside the drive. I
think it wants adjusting, and I was told you need 4 hands, a lot of
alcohol, and the ability to swear in 42 languages to successfully do this.
If anyone with manuals has any idea what to do about this and can tell me,
that would be good.
Of course, I might also be threading it the wrong way, but I tried both
ways with similar results.
The big problem with AI is that the tape MASSBUS is
dead. I don't have a
clue as to why. Diagnostics are very vague, closest thing I get on the
controller is a control bus parity error/8915 bit fiddler error.
According to the limited TOPS20 diags, this could be anything.
I think I remember this discussion... Was it ever determined if the KS
itself or the drive was at fault?
Yes, I know: 11/23's are Q-Bus and the RH11 is a
Unibus tank. However
I've also got a Qunivertor, and it seems to be working (albeit in 18 bit
mode). At least I can fire up RSX11/M and begin walkin through tests on
the TU77/TM03/RH11. And I don't have to deal with the main CPU, the 18
bit fiddler, etc...
My question is, as the 2020's RH11 is modified to be special to the
2020, will it work right on a PDP-11?
"Confuse, annoy, and DEE-STROY!" -- Jet Wolf | "Nothing Happens." --
ADVENT
"You'd be surprised what you can live through..." -- Anonymous
"...A man can pass his family and his name down through his sons, but it's
his honour that gets passed through his daughters. He can see the best
and worst of life in his girls. A daughter is something far too precious,
and he'll do anything to protect her."
-- Reichsfuehrer Siegfried Koenig, _Matrose_Mond_, David Oliver
3:36 a.m.
New subject: Fire up the iron...
The RH11 on the tape drive side is just a normal RH11. The one for the
discs is specially hacked to run in what's called "Hog mode DMA". This
is where it grabs the bus for transfers and never lets go.
On a pdp11, this would starve any other DMA device. On the 2020, the
RH11 is the only thing on that Unibus, and the 20's UBA is designed to
cache stuff coming in from the Unibus (in 18 bit chunks) and forward it
to the CPU/memory in 36bit chunks.
Chris
3:47 a.m.
New subject: Fire up the iron...
On Thu, 30 Nov 2000, Christopher Zach wrote:
The RH11 on the tape drive side is just a normal RH11.
The one for the
discs is specially hacked to run in what's called "Hog mode DMA". This
is where it grabs the bus for transfers and never lets go.
Oh, OK. I thought the tape one was odd too. I'm sorry.
"Confuse, annoy, and DEE-STROY!" -- Jet Wolf | "Nothing Happens." --
ADVENT
"You'd be surprised what you can live through..." -- Anonymous
"...A man can pass his family and his name down through his sons, but it's
his honour that gets passed through his daughters. He can see the best
and worst of life in his girls. A daughter is something far too precious,
and he'll do anything to protect her."
-- Reichsfuehrer Siegfried Koenig, _Matrose_Mond_, David Oliver
2:31 a.m.
New subject: Fire up the iron...
Greetings folks,
This weekend I will see if I can find some old friends ( Former DEC20
Service Tech's) who used to work on those Beasts.. I will pass on
this Info and see I can get some contact Info from them, and if they
are willing to Help...
enjoy,
--TY
At 10:27 PM -0500 11/27/00, Christopher Zach wrote:
Ok, it's fall, it's cool, and I can finally
start working on AI again.
With the cold weather, I can pop open a window instead of fighting a
losing battle with an 8,000 BTU AC unit.
Quick bit of status: Sitting in my computer room here at the house is:
AI (20/20 system, fully loaded with memory, passes all diags)
RM03 (Fully operational, need to dust off and purge)
RM80 (Operational, however currently unformatted)
RM03 (Logic problem in getting the heads to load)
TU77 (Doesn't work, why?)
TU77 (Spare for parts)
I've got a pair of RM02's, a TU45, and an RP07 sitting in storage. The
TU45 is a bit of a shambles, maybe with Dan's system as a plate I can
diagnose it someday. The RM02's will work on a 20, bit slower than an
RM03, but same format. The RP07 is in God knows what kind of shape (the
lid for it disintegrated on Interstate 40...). A dream would be to get
that drive operational and running as the system disk.
Once running, the systems will move to a place where power and AC is not
an issue. Till then, they run at my house.
The big problem with AI is that the tape MASSBUS is dead. I don't have a
clue as to why. Diagnostics are very vague, closest thing I get on the
controller is a control bus parity error/8915 bit fiddler error.
According to the limited TOPS20 diags, this could be anything.
So in order to figure this out, I've spent the evening pulling out the
DZ11's, the chaos controller and the Micon Ethernet card. In it's place
I have installed a pdp11/23+ for running pdp11 diagnostics on the RH11.
Yes, I know: 11/23's are Q-Bus and the RH11 is a Unibus tank. However
I've also got a Qunivertor, and it seems to be working (albeit in 18 bit
mode). At least I can fire up RSX11/M and begin walkin through tests on
the TU77/TM03/RH11. And I don't have to deal with the main CPU, the 18
bit fiddler, etc...
Thus my question: Did the tape drives work on AI before it went offline.
Somehow I seem to recall that they didn't, and this might be due to a
bad RH11 MASSBUS controller (as I have swapped in 2 other TM03
controllers with the same lack o results).
CZ
11:52 p.m.
New subject: Fire up the iron...
In some classic computer stuff I read that heating your basement with
your computer was a "Myth".
I don't know what the author was smoking, but I want some. Even my poor
collection vents enough heat to keep the place livable. Of course I'm not
a masochist. I still run the furnace.
regards,
Jeff
At 10:27 PM -0500 11/27/00, Christopher Zach wrote:
>Ok, it's fall, it's cool, and I can finally start working on AI again.
>With the cold weather, I can pop open a window instead of fighting a
>losing battle with an 8,000 BTU AC unit.
--
-----------------------------------------------------------
Jeffrey S. Worley
President
Complete Computer Services, Inc.
30 Greenwood Rd.
Asheville, NC 28803
828-277-5959
Visit our website at HTTP://www.Real-Techs.com
THETechnoid(a)home.com
-----------------------------------------------------------
8554
days inactive
8559
days old
16 comments
7 participants
participants (7)
-
cz＠alembic.crystel.com -
DSEAGRAV＠toad.xkl.com -
MRC＠Panda.COM -
pechter＠pechter.dyndns.org -
root＠bony.umtec.com -
THETechnoid＠home.com -
ty＠mit.edu