21 Feb
2011
21 Feb
'11
10:17 p.m.
I'll attempt to keep a long story short here:
I have over the past few years been putting together a MicroVAX III; I
have a KA650-B CPU, 16MB of memory, a CMD SCSI interface and a DEQNA
ethernet interface. A couple of years back I got a VCB02 and finally
this week I got ahold of a BC18Z cable so that I can actually connect
the VCB02 to a monitor/keyboard/mouse :).
I've got everything installed and wired up this evening, the problem is,
I can't seem to get it working. On powerup, the monitor syncs with the
video signal the VCB02 is sending out (and I get a brief screen with
some garbage on it followed by a black screen), and the keyboard emits a
couple of very short beeps during the diagnostic cycle, but the MicroVAX
doesn't seem to acknowledge the existence of the VCB02. All diagnostics
pass.
So, a few questions:
- Is there a specific ROM version I need in order for the KA650 to
recognize the VCB02? I have V1.2/0123 which seems pretty early (given
that I've seen references to 5.3 on the 'net). If so, anyone know where
I can find a dump?
- What keyboards is the VCB02 compatible with? I'm using it with an
LK201 from my VT220 currently, and all four LEDs are lit constantly,
which starts me to wondering...
- Are there any restrictions on slots/card ordering I should be
concerned with? This is in a BA23 backplane (which originally housed a
MicroVAX I)
Thanks!
Josh
22 Feb
22 Feb
1:55 a.m.
On 2/21/2011 10:17 PM, Josh Dersch wrote:
So, a few questions:
- Is there a specific ROM version I need in order for the KA650 to
recognize the VCB02? I have V1.2/0123 which seems pretty early (given
that I've seen references to 5.3 on the 'net). If so, anyone know
where I can find a dump?
- What keyboards is the VCB02 compatible with? I'm using it with an
LK201 from my VT220 currently, and all four LEDs are lit constantly,
which starts me to wondering...
- Are there any restrictions on slots/card ordering I should be
concerned with? This is in a BA23 backplane (which originally housed
a MicroVAX I)
Thanks!
Josh
I think I've answered my own question for the first one -- I found the
V5.3 ROMs at http://www.dunnington.u-net.com/public/DECROMs and burned
them, successfully upgrading my KA650 to 5.3. (I made copies of my V1.2
ROMs just in case)
The machine now definitely sees the VCB02 (it shows up under the SHOW
QBUS command) and appears to be trying to use it as a console at
startup, but is failing:
?62 2 08 FF 00 0000
P1=00000000 P2=00000000 P3=00000028 P4=201406C0 P5=00000000
P6=00000000 P7=00FF2000 P8=2004E1C2 P9=00000000 P10=20051CE0
r0=00000000 r1=00000028 r2=00000062 r3=201407A0 r4=2004E128
r5=00007F90 r6=00FF2000 r7=00000000 r8=00000000 ERF=00000180
KA650-B V5.3, VMB 2.7
Performing normal system tests.
33..32..31..30..29..28..27..26..25..24..23..22..21..20..19..18..
17..16..15..14..13..12..11..10..09..08..07..06..05..04..03..
Normal operation not possible.
Based on my investigations, the ?62 test code is the console QDSS test,
which would point to a failure of the VCB02 in some way. Anyone know
how to decipher the test results?
Thanks!
Josh
5:45 a.m.
Josh Dersch [derschjo at mail.msu.edu] wrote:
I think I've answered my own question for the
first one -- I
found the
V5.3 ROMs at http://www.dunnington.u-net.com/public/DECROMs
and burned
them, successfully upgrading my KA650 to 5.3. (I made copies
of my V1.2
ROMs just in case)
The machine now definitely sees the VCB02 (it shows up under the SHOW
QBUS command) and appears to be trying to use it as a console at
startup, but is failing:
?62 2 08 FF 00 0000
P1=00000000 P2=00000000 P3=00000028 P4=201406C0
P5=00000000 P6=00000000 P7=00FF2000 P8=2004E1C2
P9=00000000 P10=20051CE0 r0=00000000 r1=00000028
r2=00000062 r3=201407A0 r4=2004E128 r5=00007F90
r6=00FF2000 r7=00000000 r8=00000000 ERF=00000180
KA650-B V5.3, VMB 2.7
Performing normal system tests.
33..32..31..30..29..28..27..26..25..24..23..22..21..20..19..18..
17..16..15..14..13..12..11..10..09..08..07..06..05..04..03..
Normal operation not possible.
Based on my investigations, the ?62 test code is the console
QDSS test,
which would point to a failure of the VCB02 in some way. Anyone know
how to decipher the test results?
The VCB02 tech manual (EK-104AA-TM-001) should be available online. If
you cannot
find it, I'll upload it somewhere. It covers configuration and so on.
Doesn't
seem to include diagnostic output, but I did skim rather quickly.
KA650-B suggests, to me, a VAXserver board. That would (iirc) have an
effect
on OpenVMS, but I've no idea what (if any) effect it would have on the
console
ROM working (or not) with the QDSS. You might want to see if anyone has
KA650-A ROMs available.
I do remember that V1.2 was the "minimum port from a KA630", early rev
of
the KA650 console code. I had a few in the lab at DEC and they were
certainly less capable: no probing the bus or looking for devices etc.
You definitely want the later version that you now have. (For historical
purposes, it would be nice to feed your original ones back, just in case
there is a use for V1.2!)
Antonio
arcarlini at iee.org
9:59 a.m.
On 2/22/2011 5:45 AM, arcarlini at iee.org wrote:
Josh Dersch [derschjo at mail.msu.edu] wrote:
Thanks -- I don't know why I didn't think to look at that in the first
place. Turns out there are diagnostic LEDs on the board (helpful!) and
they indicate "keyboard failure". This is the only compatible keyboard
I have, and I had previously been using it with my VT220, so I was
fairly sure it was working (but it's been at least a year since I last
powered it on). I dug out my VT220 and sure enough, the keyboard fails
there too...
So... problem solved, I just need to find a working keyboard :).
I think I've answered my own question for the
first one -- I
found the
V5.3 ROMs at http://www.dunnington.u-net.com/public/DECROMs
and burned
them, successfully upgrading my KA650 to 5.3. (I made copies
of my V1.2
ROMs just in case)
The machine now definitely sees the VCB02 (it shows up under the SHOW
QBUS command) and appears to be trying to use it as a console at
startup, but is failing:
?62 2 08 FF 00 0000
P1=00000000 P2=00000000 P3=00000028 P4=201406C0
P5=00000000 P6=00000000 P7=00FF2000 P8=2004E1C2
P9=00000000 P10=20051CE0 r0=00000000 r1=00000028
r2=00000062 r3=201407A0 r4=2004E128 r5=00007F90
r6=00FF2000 r7=00000000 r8=00000000 ERF=00000180
KA650-B V5.3, VMB 2.7
Performing normal system tests.
33..32..31..30..29..28..27..26..25..24..23..22..21..20..19..18..
17..16..15..14..13..12..11..10..09..08..07..06..05..04..03..
Normal operation not possible.
Based on my investigations, the ?62 test code is the console
QDSS test,
which would point to a failure of the VCB02 in some way. Anyone know
how to decipher the test results?
The VCB02 tech manual (EK-104AA-TM-001) should
be available online. If
you cannot
find it, I'll upload it somewhere. It covers configuration and so on.
Doesn't
seem to include diagnostic output, but I did skim rather quickly. KA650-B suggests, to me, a VAXserver board. That would
(iirc) have an
effect
on OpenVMS, but I've no idea what (if any) effect it would have on the
console
ROM working (or not) with the QDSS. You might want to see if anyone has
KA650-A ROMs available.
I do remember that V1.2 was the "minimum port from a KA630", early rev
of
the KA650 console code. I had a few in the lab at DEC and they were
certainly less capable: no probing the bus or looking for devices etc.
You definitely want the later version that you now have. (For historical
purposes, it would be nice to feed your original ones back, just in case
there is a use for V1.2!)
I have made images, if anyone's looking :).
Thanks!
Josh
Antonio
arcarlini at iee.org
10:11 a.m.
Josh Dersch [derschjo at mail.msu.edu] wrote:
Thanks -- I don't know why I didn't think to
look at that in
the first
place. Turns out there are diagnostic LEDs on the board (helpful!)
and they indicate "keyboard failure". This is the only
I should have read the manual more carefully but I had something to
get done on the car before it got dark. I'll pretend it was much better
in the long run you get you to read it :-)
I was surprised that in my quick skim I didn't see any mention of the
sorts of errors you might see on the console for various errors.
I think there's a way of booting the uVAX as though the graphics
board were not there: some setting(s) of the various switches.
Hopefully someone with a better memory will remember how that's done.
If you can do that, then the OS might still initialise the graphics
and you'll have at least half a workstation.
That manual also covers the LK250 in some detail. Possibly not
enoguh detail to fix it, but I *think* the schematics are up on
bitsavers (perhaps part of another manual, I'm sure I've seen
them somewhere).
How hard can it be to fix a keyboard ?
I have made images, if anyone's looking :).
Hopefully Pete will take them and add them to the archive.
I'd take them but I'd probably lose them before needing them :-)
Antonio
arcarlini at iee.org
11:47 a.m.
That manual also covers the LK250 in some detail.
Possibly not
Is it an LK250 or an LK201? The latter goes with the VT220, etc. The
former from what I refall is the PC/AT compatible keyboard with much the
smae layout.
enoguh detail to fix it, but I *think* the schematics
are up on
bitsavers (perhaps part of another manual, I'm sure I've seen
them somewhere).
I reverse-engieered it years ago. It's very simple. An 8051
microcontrolelr, a couple of TTL decoders, RS232 level shifter and a 7805
regualtor basically. And the membrane keybaord assembly.
How hard can it be to fix a keyboard ?
Well, the mcirocontroller is custom-programmed... The membrane assembly
(whcih is what normally fails) is heat-staked toether. You can unclip
all the keycaps but that doens't do you a lot of good. Yes, you can cut
away the heat stakes and take the whole lot apart, but how on earth do
you get it to stary reassembled. There isn't enough plastic left to form
the mins over for a second time
-tony
9:03 p.m.
On 2/22/2011 10:11 AM, arcarlini at iee.org wrote:
Just for future reference, I've thrown them up here:
http://yahozna.dyndns.org/computers/software/vax/ka650/v1.2/
Feel free to grab them :).
- Josh
Josh Dersch [derschjo at mail.msu.edu] wrote:
I didn't see anything about test codes/errors either, but I found the
section on the diagnostic LEDs and that's fortunately all I need :).
Thanks -- I don't know why I didn't think
to look at that in
the first
place. Turns out there are diagnostic LEDs on the board (helpful!)
and they indicate "keyboard failure". This is the only
I should have
read the manual more carefully but I had something to
get done on the car before it got dark. I'll pretend it was much better
in the long run you get you to read it :-)
I was surprised that in my quick skim I didn't see any mention of the
sorts of errors you might see on the console for various errors. I think there's a way of booting the uVAX as
though the graphics
board were not there: some setting(s) of the various switches.
Hopefully someone with a better memory will remember how that's done.
I can still get it to boot via a serial console without any issues,
fortunately.
If you can do that, then the OS might still initialise
the graphics
and you'll have at least half a workstation.
All I have running at the moment is OpenBSD, I'm hoping to get VMS
running on it shortly once I get a working keyboard.
I have made
images, if anyone's looking :).
Hopefully Pete will take them and add them to
the archive.
I'd take them but I'd probably lose them before needing them :-) 
10:37 a.m.
On Tue, Feb 22, 2011 at 9:59 AM, Josh Dersch <derschjo at mail.msu.edu> wrote:
So... problem solved, I just need to find a working keyboard :).
Last time I was at RE-PC Tukwila they had a few LK201s in their random
keyboard box for something like $5 each and I picked up one or two at
the time. That was a few months back and no idea if they have any
there now.
-Glen
11:42 a.m.
Thanks -- I don't know why I didn't think to
look at that in the first
place. Turns out there are diagnostic LEDs on the board (helpful!) and
they indicate "keyboard failure". This is the only compatible keyboard
I have, and I had previously been using it with my VT220, so I was
fairly sure it was working (but it's been at least a year since I last
powered it on). I dug out my VT220 and sure enough, the keyboard fails
there too...
Most of the time LK201s fail becuase of a problem with the membrane
switches, not the cotnroller board.
You might try taking it apart (the screws are under the push-in plastic
feet), remvoig nthe unit from the bottom case (frob the clip and slidt it
to release it), then removing the control PCB fixings (clip on the
regulator, and 1 screw). Then bend up the tabs on the metal clamps that
hold the mebrrane tails to the PCB and take them off.
Try connectin the PCB alone to the VT220. If it passes, try it on the
microvax too. If will at least let yoy know if the latter is likely to
work with a good keybaord.
Alas I have never managed to dismantle the heat-staked membane in a way
taht it can be reassembled :-(
-tony
9:25 p.m.
On 2/22/2011 11:42 AM, Tony Duell wrote:
I just got done trying this. No dice, with the membrane disconnected 3
of the LEDs light up constantly and that's it (no beeps or anything
else). With the membrane connected I get 4 LEDs and a few very short
(shorter than I recall under normal circumstances) beeps. So perhaps
the membrane isn't to blame in this case.
Thanks for the suggestion, though. Maybe later this week I'll have some
time to do some debugging on it...
Josh
Thanks -- I
don't know why I didn't think to look at that in the first
place. Turns out there are diagnostic LEDs on the board (helpful!) and
they indicate "keyboard failure". This is the only compatible keyboard
I have, and I had previously been using it with my VT220, so I was
fairly sure it was working (but it's been at least a year since I last
powered it on). I dug out my VT220 and sure enough, the keyboard fails
there too...
Most of the time LK201s fail becuase of a problem with the membrane
switches, not the cotnroller board.
You might try taking it apart (the screws are under the push-in plastic
feet), remvoig nthe unit from the bottom case (frob the clip and slidt it
to release it), then removing the control PCB fixings (clip on the
regulator, and 1 screw). Then bend up the tabs on the metal clamps that
hold the mebrrane tails to the PCB and take them off.
Try connectin the PCB alone to the VT220. If it passes, try it on the
microvax too. If will at least let yoy know if the latter is likely to
work with a good keybaord. Alas I have never managed to dismantle the heat-staked
membane in a way
taht it can be reassembled :-(
-tony
23 Feb
23 Feb
11:02 a.m.
Try connectin
the PCB alone to the VT220. If it passes, try it on the
microvax too. If will at least let yoy know if the latter is likely to
work with a good keybaord.
I just got done trying this. No dice, with the membrane disconnected 3
of the LEDs light up constantly and that's it (no beeps or anything else). With the membrane connected I get 4 LEDs and a
few very short
(shorter than I recall under normal circumstances) beeps. So perhaps
the membrane isn't to blame in this case.
Thsi, AFAIK, makes no sense. A correctly working membrane assembly is
open-circuit between all conenctions. So the cotnroller PCB can't tell
whether it's connected or not.
-tony
25 Feb
25 Feb
8:17 p.m.
On 2/23/2011 11:02 AM, Tony Duell wrote:
Affirmative. Clips removed, etc.
Hey, just reporting on what I'm seeing :). Perhaps both the controller
and the membrane have got something wrong with them.
At any rate, I dropped by a local recycler (RE-PC in Tukwila, WA) and
they had a box of LK201s for $5 each. I picked up a few just to be on
the safe side :).
Everything seems to be working fine now, just need to get VMS up and
running...
Thanks all for the help!
Josh
I just got
done trying this. No dice, with the membrane disconnected 3
of the LEDs light up constantly and that's it (no beeps or anything
You did
remove the metal clips from the PCB and not just pull out the
membran tails I trust. If you do the latter, said clips short all the
pads togther and cause all sorts of problems. else). With
the membrane connected I get 4 LEDs and a few very short
(shorter than I recall under normal circumstances) beeps. So perhaps
the membrane isn't to blame in this case.
Thsi, AFAIK, makes no sense. A
correctly working membrane assembly is
open-circuit between all conenctions. So the cotnroller PCB can't tell
whether it's connected or not.
26 Feb
26 Feb
8:27 a.m.
On 2/23/2011 11:02 AM, Tony Duell wrote:
Affirmative. Clips removed, etc.
OK.. One thing I have learnt from years of 'remote fault finding' is the
qurstion that Iforget to ask is the one that gives the real clue to what
is going on :-)
I just
got done trying this. No dice, with the membrane disconnected 3
of the LEDs light up constantly and that's it (no beeps or anything
You did
remove the metal clips from the PCB and not just pull out the
membran tails I trust. If you do the latter, said clips short all the
pads togther and cause all sorts of problems. else).
With the membrane connected I get 4 LEDs and a few very short
(shorter than I recall under normal circumstances) beeps. So perhaps
the membrane isn't to blame in this case.
Thsi, AFAIK, makes no sense. A
correctly working membrane assembly is
open-circuit between all conenctions. So the cotnroller PCB can't tell
whether it's connected or not.
4831
days inactive
4835
days old
13 comments
4 participants
participants (4)
-
arcarlini＠iee.org -
ard＠p850ug1.demon.co.uk -
derschjo＠mail.msu.edu -
glen.slick＠gmail.com