22 Nov
2012
22 Nov
'12
2 p.m.
Hello.
I have a VAX730 with both TU58 drives destroyed (capstan melted, need
replacements).
I also have a bunch of cassettes, but unfortunately all seem to have
problems with the bend and/or bad spots on the tape.
Possibly I would try to replace the broken bands (if I find a source)
and/or replace the magnetic tape when damaged (I was thinking to try
with audio cassette tape, don' t know if metal oxide high density tape
could be good for it).
Anybody has some information about the coercivity of original DEC TU58 tape?
One problem indeed is the need of reformatting the tape, but: if I can
emulate the TU58 drive using a serial, would it be possible to send
raw commands to the drive using the serial and a PC?
Andrea
PS-If possible, some good-condition cassette would be very useful to
me too. I'm located in Italy.
23 Nov
23 Nov
7:06 a.m.
On Thursday (11/22/2012 at 11:00PM +0100), shadoooo wrote:
Hello.
I have a VAX730 with both TU58 drives destroyed (capstan melted, need
replacements).
I also have a bunch of cassettes, but unfortunately all seem to have
problems with the bend and/or bad spots on the tape.
Possibly I would try to replace the broken bands (if I find a source)
and/or replace the magnetic tape when damaged (I was thinking to try
with audio cassette tape, don' t know if metal oxide high density tape
could be good for it).
Anybody has some information about the coercivity of original DEC TU58 tape?
One problem indeed is the need of reformatting the tape, but: if I can
emulate the TU58 drive using a serial, would it be possible to send
raw commands to the drive using the serial and a PC?
It's more than just sending the correct serial command, at least for the
original DEC Tape II, for which the documentation is here,
http://bitsavers.trailing-edge.com/pdf/dec/dectape/EK-0TU58-UG-001_TU58ug.p…
There is no command in the set to accomplish formatting. So, as Eric
pointed out in an earlier message, it would require different firmware
(at least for the DEC Tape II) to accomplish it.
The above document also discusses some parameters of the tape itself.
I do have some DC100A that are not DEC labeled. They are just Scotch
(3M) brand and have a hand written "DEC" paper label stuck to them.
This suggests to me that someone had a method for formatting them. Other
labeling on the tapes suggests VAX 4500 (?) may have been involved so
I wonder if it was possible to format raw DC100A on some flavors of VAX?
I found replacement capstans on eBay and waited until they went "on
sale" and then nabbed a couple. They too are rare. I wonder if there's
something that could be applied to keep them from degrading again?
My TU58 spent its days prior to restoration in a particularly bad
environment. A hot, humid barn. This surely helped the rubber capstan
degrade and also caused rusting on a number of the metal components-- in
particular the motor shaft where that capstan is mounted. Even though
the capstan is an aluminum hub, the rusted motor shaft held onto it
like they were one piece. I really struggled to get the capstan off,
finally heating the whole shaft and capstan with a 150W soldering iron
and then hammering on it with a hammer and center punch to break it loose.
PS-If possible, some good-condition cassette would be
very useful to
me too. I'm located in Italy.
I think the challenge is on :-)
Chris
--
Chris Elmquist
4:39 p.m.
New subject: VAX 4000-300 and a CQD-423 Qbus SCSI adapter questions
I am trying to understand why a CMD CQD-423 qbus scsi adapter in my
VAX4000-300
is functioning in this manner. I think it's configured properly as I can see
the drives once VMS
is booted, sysgen verifies the proper driver is loaded. I can mount the
drives as a DUAn device.
Yet from the console prompt I cannot see the drives and cannot boot from the
attached SCSI
drives. My goal was to retire the ancient DSSI drives on this system before
they all fail, move
OS and data to the SCSI drives and use SCSI as a primary OS and data drives.
I have the manual,
read it several times and seem to have missed something.
Can this controller be used to boot from? I would think so which leads me to
believe, my reading
comprehension may be a stake here.
Any suggestions?
Dan Snyder
Butler, PA
24 Nov
24 Nov
8:03 a.m.
New subject: VAX 4000-300 and a CQD-423 Qbus SCSI adapter questions
What's the output of "SHOW QBUS" and "SHOW DEV" at the console
prompt?
23 Nov
23 Nov
5:40 p.m.
Chris Elmquist wrote:
I do have some DC100A that are not DEC labeled. They
are just Scotch
(3M) brand and have a hand written "DEC" paper label stuck to them.
This suggests to me that someone had a method for formatting them.
Other labeling on the tapes suggests VAX 4500 (?) may have been
involved so I wonder if it was possible to format raw DC100A on some
flavors of VAX?
It is remotely possible that there's an undocumented "format" command in
the protocol. However, I've heard multiple people claim that special
firmware was required.
As far as I can tell, the TU58 drive hardware should be perfectly
capable of formatting a tape, unlike e.g. disk drives that use embedded
servo.
6:31 p.m.
On 11/23/2012 08:40 PM, Eric Smith wrote:
Chris Elmquist wrote:
Wishful thinking.
I've disassembled the firmware its not there. It's not that hard to do
the reason I say modded is the 8085
has only has 256bytes of ram and 2k of rom on the board. So bypassing
the 8085 or extending the memory
and firmware by modding the board for more rom space and maybe ram space
could make the needed
formatter. In the end you need the TU58 board for the drive interface
and motor control.
TU58 core:
8085 at 2.4576mhz (4.9152 crystal), 8155 (22 io lines, timer used for
various timing needs), 2K rom,
64s004 ART (serial).
Bob at Spare Time Gizmos did a TU-58 emulator using ram devices and
DS5000 (8051 like) CPU back around 1995.
It's still the way to go save for flash is now available. I'd buy two
in a heartbeat now.
Allison
I do have some DC100A that are not DEC labeled.
They are just Scotch
(3M) brand and have a hand written "DEC" paper label stuck to them.
This suggests to me that someone had a method for formatting them.
Other labeling on the tapes suggests VAX 4500 (?) may have been
involved so I wonder if it was possible to format raw DC100A on some
flavors of VAX?
It is remotely possible that there's an undocumented "format" command
in the protocol. However, I've heard multiple people claim that
special firmware was required.
As far as I can tell, the TU58 drive hardware should be perfectly
capable of formatting a tape, unlike e.g. disk drives that use
embedded servo.
24 Nov
24 Nov
12:01 a.m.
It is remotely possible that there's an
undocumented "format" command in
the protocol. However, I've heard multiple people claim that special
firmware was required.
As far as I can tell, the TU58 drive hardware should be perfectly capable
of formatting a tape, unlike e.g. disk drives that use embedded servo.
If someone wills to create a formatter, I'd like to keep informed. I
always had curiosity of how these things works :)
1:06 a.m.
I wrote:
It is remotely possible that there's an
undocumented "format" command
in the protocol. However, I've heard multiple people claim that
special firmware was required.
I spent some time reverse-engineering the firmware. There is only one
undocumented opcode, decimal 10, and I haven't yet figured out what it
does, but it definitely doesn't format a tape.
I think I understand the code that writes a block, but of course there's
no code to write an address field. Writing formatter firmware shouldn't
be too hard, but it would be best to have a known good tape to examine
to find out the details of the gap lengths.
Eric
11 a.m.
On Saturday (11/24/2012 at 02:06AM -0700), Eric Smith wrote:
I wrote:
I saved an image of the ROM while I had this unit apart but I see you
guys are already way ahead.
If the raw tapes more readily available, I think it would make sense to
pursue the ability to format them. But since it seems even raw tapes
are approaching unobtainium in any quantity, maybe it is time to revisit
the solid-state emulation. Perhaps a job for a BeagleBone or Raspberry
Pi these days...
Chris
--
Chris Elmquist
It is remotely possible that there's an
undocumented "format"
command in the protocol. However, I've heard multiple people
claim that special firmware was required.
I spent some time reverse-engineering the firmware. There is only
one undocumented opcode, decimal 10, and I haven't yet figured out
what it does, but it definitely doesn't format a tape.
I think I understand the code that writes a block, but of course
there's no code to write an address field. Writing formatter
firmware shouldn't be too hard, but it would be best to have a known
good tape to examine to find out the details of the gap lengths. 
25 Nov
25 Nov
12:34 a.m.
On Sat, 24 Nov 2012 13:00:11 -0600
Chris Elmquist <chrise at pobox.com> wrote:
maybe it is time to revisit the solid-state emulation.
Perhaps a job for a BeagleBone or Raspberry Pi these days...
Thats like killing a
fly with a sledge hamer.
A smalish Atmel ATmega and a SD card could do that job perfectly.
See also:
http://www.sparetimegizmos.com/Hardware/TU58_Emulator.htm
http://www.fpns.net/willy/pdp11/tu58-emu.htm
--
\end{Jochen}
\ref{http://www.unixag-kl.fh-kl.de/~jkunz/}
6:38 p.m.
On Sunday (11/25/2012 at 09:34AM +0100), Jochen Kunz wrote:
On Sat, 24 Nov 2012 13:00:11 -0600
Chris Elmquist <chrise at pobox.com> wrote:
Exactly... because then I can transfer images in and out
via SD card,
ethernet, wifi... and I can put it together in an afternoon with parts
on hand.
maybe it is time to revisit the solid-state
emulation.
Perhaps a job for a BeagleBone or Raspberry Pi these days...
Thats like killing a
fly with a sledge hamer. A smalish Atmel ATmega and a SD card could do that job
perfectly.
See also:
http://www.sparetimegizmos.com/Hardware/TU58_Emulator.htm
Yup. No longer
available.
Yup. Lots of solutions existing as code to run on a PC, Unix,
Linux. Should be a no brainer to port to a small ARM platform such as
I mentioned.
I still like (kind of a love-hate sort of thing) the real TU58 though.
It's love-hate because even my "new" capstan rollers have turned to goo
after just a few cycles of the tape past them. So, I am exploring the
Tygon tubing and some o-ring options next.
Additionally, I found that the DC100A cartridges I have that were
labeled with hand written "4500" stickers are all unreadable in my drive
(before the capstans failed). They all returned "block not found"
error no matter what block I try to seek. I have concluded these were
reformatted into some other non-TU58 format and so they became donors
for belts to fix the other tapes that were labeled "DEC". And so now,
I have recovered five tapes that appeared to work error free in my drive
until the capstans melted.
Chris
--
Chris Elmquist
26 Nov
26 Nov
5:56 a.m.
On Sun, 25 Nov 2012, Chris Elmquist wrote:
On Sunday (11/25/2012 at 09:34AM +0100), Jochen Kunz
wrote:
I've been able to recover tapes that were damaged by a gooey capstan by
reading the tape until a bad block appeared, cleaning the head, and
repeating.
Each pass transfers a little bit of goo from the tape to the head,
eventually removing enough to read the block.
On Sat, 24 Nov 2012 13:00:11 -0600
Chris Elmquist <chrise at pobox.com> wrote:
Exactly... because then I can transfer images in and out
via SD card,
ethernet, wifi... and I can put it together in an afternoon with parts
on hand.
maybe it is time to revisit the solid-state
emulation.
Perhaps a job for a BeagleBone or Raspberry Pi these days...
Thats like killing a
fly with a sledge hamer. A smalish Atmel ATmega and a SD card could do
that job perfectly.
See also:
http://www.sparetimegizmos.com/Hardware/TU58_Emulator.htm
Yup. No longer
available.
Yup. Lots of solutions existing as code to run on a PC, Unix,
Linux. Should be a no brainer to port to a small ARM platform such as
I mentioned.
I still like (kind of a love-hate sort of thing) the real TU58 though.
It's love-hate because even my "new" capstan rollers have turned to goo
after just a few cycles of the tape past them. So, I am exploring the
Tygon tubing and some o-ring options next.
Additionally, I found that the DC100A cartridges I have that were
labeled with hand written "4500" stickers are all unreadable in my drive
(before the capstans failed). They all returned "block not found"
error no matter what block I try to seek. I have concluded these were
reformatted into some other non-TU58 format and so they became donors
for belts to fix the other tapes that were labeled "DEC". And so now,
I have recovered five tapes that appeared to work error free in my drive
until the capstans melted.
Chris
--
Chris Elmquist
9:29 a.m.
On Monday (11/26/2012 at 06:56AM -0700), cctech at vax-11.org wrote:
I've been able to recover tapes that were damaged by a gooey capstan
by reading the tape until a bad block appeared, cleaning the head,
and repeating.
Each pass transfers a little bit of goo from the tape to the head,
eventually removing enough to read the block.
Yes. Unfortunately, what I'm experiencing is the roller in the cartridge
embeds itself into the gooey capstan and then stops turning. It'll start
moving fine, seeking pretty far in and then begins turning slower and
slower until it finally fails with a motor stall error.
Chris
--
Chris Elmquist
10:38 a.m.
On 11/26/2012 12:29 PM, Chris Elmquist wrote:
On Monday (11/26/2012 at 06:56AM -0700), cctech at
vax-11.org wrote:
The trick is if the roller on inspection is soft to the point of goo
is to NOT INSERT a tape and foul it with the goo.
These rollers all by now have failed to goo and the only choice is to
remove the roller and remove the goo and replace with tygon tubing
or other rubber product. I started seeing failed rollers more than
18 years ago so this is not a new problem.
Allison
I've been able to recover tapes that were
damaged by a gooey capstan
by reading the tape until a bad block appeared, cleaning the head,
and repeating.
Each pass transfers a little bit of goo from the tape to the head,
eventually removing enough to read the block.
Yes. Unfortunately, what I'm
experiencing is the roller in the cartridge
embeds itself into the gooey capstan and then stops turning. It'll start
moving fine, seeking pretty far in and then begins turning slower and
slower until it finally fails with a motor stall error.
Chris
27 Nov
27 Nov
3:33 a.m.
On Monday (11/26/2012 at 01:38PM -0500), Allison wrote:
On 11/26/2012 12:29 PM, Chris Elmquist wrote:
Correct. These rollers were not gooey in any way on inspection.
They were NOS replacements I ordered from a DEC parts supplier.
On Monday (11/26/2012 at 06:56AM -0700), cctech at
vax-11.org wrote:
The trick is if the roller on inspection is soft to the point of goo
is to NOT INSERT a tape and foul it with the goo. I've been able to recover tapes that were
damaged by a gooey capstan
by reading the tape until a bad block appeared, cleaning the head,
and repeating.
Each pass transfers a little bit of goo from the tape to the head,
eventually removing enough to read the block.
Yes. Unfortunately, what I'm
experiencing is the roller in the cartridge
embeds itself into the gooey capstan and then stops turning. It'll start
moving fine, seeking pretty far in and then begins turning slower and
slower until it finally fails with a motor stall error.
Chris
These rollers all by now have failed to goo and the
only choice is to
remove the roller and remove the goo and replace with tygon tubing
or other rubber product. I started seeing failed rollers more than
18 years ago so this is not a new problem.
Yes. Unfortunately, they became soft after some use. I did not have
a reference for how hard they should be when new and so they seemed OK
(a lot better than the disintegrated ones that were on the transports
when I started).
I have correct diameter (5/8"OD, 7/16"ID) tygon tubing as well as
various types of o-rings on order that I will try as replacements.
I am guessing about the durometer of the rubber so have a selection of
different hardnesses to try and see what works.
Chris
--
Chris Elmquist
7:50 a.m.
On 11/27/2012 06:33 AM, Chris Elmquist wrote:
On Monday (11/26/2012 at 01:38PM -0500), Allison
wrote:
NOS rollers are sill OLD and shot. The only thing is NEW build.
When new the roller had a hardness close to rubber tires and
when they started to fail they would finger nail dent easily.
On 11/26/2012 12:29 PM, Chris Elmquist wrote:
Correct. These rollers were not
gooey in any way on inspection.
They were NOS replacements I ordered from a DEC parts supplier.
On Monday (11/26/2012 at 06:56AM -0700), cctech
at vax-11.org wrote:
The trick is if the roller on inspection is soft to the point of goo
is to NOT INSERT a tape and foul it with the goo. I've been able to recover tapes that were
damaged by a gooey capstan
by reading the tape until a bad block appeared, cleaning the head,
and repeating.
Each pass transfers a little bit of goo from the tape to the head,
eventually removing enough to read the block.
Yes. Unfortunately, what I'm
experiencing is the roller in the cartridge
embeds itself into the gooey capstan and then stops turning. It'll start
moving fine, seeking pretty far in and then begins turning slower and
slower until it finally fails with a motor stall error.
Chris
These rollers all by now have failed to goo and
the only choice is to
remove the roller and remove the goo and replace with tygon tubing
or other rubber product. I started seeing failed rollers more than
18 years ago so this is not a new problem.
Yes. Unfortunately, they became soft
after some use. I did not have
a reference for how hard they should be when new and so they seemed OK
(a lot better than the disintegrated ones that were on the transports
when I started). I have correct diameter (5/8"OD, 7/16"ID)
tygon tubing as well as
various types of o-rings on order that I will try as replacements.
I am guessing about the durometer of the rubber so have a selection of
different hardnesses to try and see what works.
Since there was no new build back 18 years ago I went to a repair
solution and have use it since. Mins was specifically a .150-.160
wide ring of tygon of the right ID to fit tight on the cleaned hub
and anchored with a drop of superglue. I tried Orings and they
were hard to find with the right ID/OD and it required machining
the hub to have a grove to keep it.
Since the tape has servo data for speed (part of the encoding)
the OD of the roller can vary with no ill effects so long as mechanical
fit and traction are good.
Allison
28 Nov
28 Nov
11:38 a.m.
Correct. These rollers were not gooey in any way on
inspection.
They were NOS replacements I ordered from a DEC parts supplier.
Unfortuantely, they deacy with time, not really with use. So a NOS one is
likely to be unusable.
I have correct diameter (5/8"OD, 7/16"ID) tygon tubing as well as
various types of o-rings on order that I will try as replacements.
I am guessing about the durometer of the rubber so have a selection of
different hardnesses to try and see what works.
Rik Bos reported that a Citroen seelve seal (the largest size) could be
stretched ovr the hub of the capstan in an HP tape drive that uses
simialr-sized cartridges. That mey be suitabkle for the TU58 too.
-tony
29 Nov
29 Nov
6:33 a.m.
On Wednesday (11/28/2012 at 07:38PM +0000), Tony Duell wrote:
Yes. I've learned this now. I mistakenly assumed my original ones had
decayed due to how the unit was stored (basically outside)-- and it was
extreme heat, cold and ozone that did it in. Now, it's pretty clear
it's an issue of the material and time.
OK. Thanks for that. I've got the Tygon tubing in hand and will try
that first. Having a little trouble coming up with a good way to cut
it squarely though as it likes to deform while cutting. This is causing
the first attempts to come out non-uniform in height. I think I need
to put a 7/16" dowel inside it and then cut it on the dowel (or through
the dowel in a small mitre box)-- unless others have a good technique
for getting a nice square cut?
Chris
--
Chris Elmquist
Correct.
These rollers were not gooey in any way on inspection.
They were NOS replacements I ordered from a DEC parts supplier.
Unfortuantely, they deacy with time, not really with use. So a NOS one is
likely to be unusable. I have correct
diameter (5/8"OD, 7/16"ID) tygon tubing as well as
various types of o-rings on order that I will try as replacements.
I am guessing about the durometer of the rubber so have a selection of
different hardnesses to try and see what works.
Rik Bos reported that a Citroen seelve seal (the largest size) could be
stretched ovr the hub of the capstan in an HP tape drive that uses
simialr-sized cartridges. That mey be suitabkle for the TU58 too.
8:26 a.m.
On 11/29/2012 09:33 AM, Chris Elmquist wrote:
On Wednesday (11/28/2012 at 07:38PM +0000), Tony Duell
wrote:
Yes. I've learned this now. I mistakenly assumed my
original ones had
decayed due to how the unit was stored (basically outside)-- and it was
extreme heat, cold and ozone that did it in. Now, it's pretty clear
it's an issue of the material and time.
OK. Thanks for
that. I've got the Tygon tubing in hand and will try
that first. Having a little trouble coming up with a good way to cut
it squarely though as it likes to deform while cutting. This is causing
the first attempts to come out non-uniform in height. I think I need
to put a 7/16" dowel inside it and then cut it on the dowel (or through
the dowel in a small mitre box)-- unless others have a good technique
for getting a nice square cut?
Chris
I just put an over size slice on the hub after cleaning and used
a bit of superglue to lock it. With the motor powered and tuning
I use an exacto knife to slice to top level with the top of the hub
and then make a cut down to metal about .160 down from there
for a clean edge. Any excess is peeled off. Then I used an emery
board while spinning to smooth and take the glazed surface off.
The whole process takes me a few minutes and I'm done.
I don't try to over complicate it. It needs to be the right width
for proper contact and no interference with the case and you
can eyeball that. That and reasonably round is good too.
One thing, even with tygon do not leave the tape in contact
when not in use, as everything I've tried (other materials)
get a flat spot in the roller that makes the drive noisy even
though it works fine.
Allison
Correct.
These rollers were not gooey in any way on inspection.
They were NOS replacements I ordered from a DEC parts supplier.
Unfortuantely,
they deacy with time, not really with use. So a NOS one is
likely to be unusable. I have
correct diameter (5/8"OD, 7/16"ID) tygon tubing as well as
various types of o-rings on order that I will try as replacements.
I am guessing about the durometer of the rubber so have a selection of
different hardnesses to try and see what works.
Rik Bos reported that a Citroen
seelve seal (the largest size) could be
stretched ovr the hub of the capstan in an HP tape drive that uses
simialr-sized cartridges. That mey be suitabkle for the TU58 too.
12:04 p.m.
OK. Thanks for that. I've got the Tygon tubing in
hand and will try
that first. Having a little trouble coming up with a good way to cut
it squarely though as it likes to deform while cutting. This is causing
the first attempts to come out non-uniform in height. I think I need
to put a 7/16" dowel inside it and then cut it on the dowel (or through
the dowel in a small mitre box)-- unless others have a good technique
for getting a nice square cut?
What I do is this :
Cut an over-long piece roughly, making sure I haev enough to cut the
ragged ends off and still have the bit I want.
Put the tubing on suitable rod as a mandrel Any material will do, wood,
steel, etc.
Put it i nthe 3 jaw chucj of the lathe. Hold a knife against the side of
the toolpost and move the altter to get the blade where I want if over
the tubing.
Tunr the lathe _by hand_. Don't even think of using the motor unless you
wanta knife blade stuck in you. Prest the blade agaisn the tubing to cut it.
Move the toolpost to ge the lrnght of tubing I wnt. cut again
Slide the bits off the mandrel and keep the one I want.
Since you don't need the accuracy of the lathe, you might be able to rig
soemthign up usin a hand drill, a block of fwood to guide the knife
against, etc.
-tony
2:17 p.m.
On Thursday (11/29/2012 at 08:04PM +0000), Tony Duell wrote:
What I do is this :
Cut an over-long piece roughly, making sure I haev enough to cut the
ragged ends off and still have the bit I want.
Put the tubing on suitable rod as a mandrel Any material will do, wood,
steel, etc.
Put it i nthe 3 jaw chucj of the lathe. Hold a knife against the side of
the toolpost and move the altter to get the blade where I want if over
the tubing.
Tunr the lathe _by hand_. Don't even think of using the motor unless you
wanta knife blade stuck in you. Prest the blade agaisn the tubing to cut it.
Move the toolpost to ge the lrnght of tubing I wnt. cut again
Slide the bits off the mandrel and keep the one I want.
Since you don't need the accuracy of the lathe, you might be able to rig
soemthign up usin a hand drill, a block of fwood to guide the knife
against, etc.
Yes. Thanks Tony. That seems very workable and along with Allison's
suggestion of just cutting it on the hub of the capstan being refurbished,
I think I can get there. I could mount the salvaged hub to a drill bit,
put that in a drill motor, and then turn this assy by hand while cutting.
Thanks guys.
Chris
--
Chris Elmquist
25 Nov
25 Nov
1:10 p.m.
On 11/24/2012 01:00 PM, Chris Elmquist wrote:
But since it seems even raw tapes are approaching
unobtainium in any
quantity
Modify tapes to contain an embedded USB stick, modify drive* to accept
them, retain user experience of inserting/removing cartridge... ;-)
* including delays, occasional errors etc., of course.
7:49 p.m.
On Sun, 25 Nov 2012, Jules Richardson wrote:
On 11/24/2012 01:00 PM, Chris Elmquist wrote:
I sure do miss the 15 minute boot times of my 11/750, which only failed
20% of the time due to a tape error near the end of the console tape.
But since it seems even raw tapes are approaching
unobtainium in any
quantity
Modify tapes to contain an embedded USB stick, modify drive* to accept them,
retain user experience of inserting/removing cartridge... ;-)
* including delays, occasional errors etc., of course.
27 Nov
27 Nov
12:07 p.m.
> Modify tapes to contain an embedded USB stick,
modify drive* to accept them,
> retain user experience of inserting/removing cartridge... ;-)
>
> * including delays, occasional errors etc., of course.
But this does not retain the user (OK,'hacker') experience of repairing
decayed captans, cleaning heads, retrheding tapes that have come off the
spool i nthe cartridge, and so on.
-tony
25 Nov
25 Nov
3:23 p.m.
On 11/24/2012 02:00 PM, Chris Elmquist wrote:
On Saturday (11/24/2012 at 02:06AM -0700), Eric Smith
wrote:
This is clearly using a tactical nuke to clear an ant farm. The original
TU-58 uses a very bare 8085 with only 256 bytes or ram and 2K of Rom
running at a clock speed that is 5/6ths the nominal speed of that CPU.
This should give a clear indication that very little processing power is
needed.
As to what a TU58 is, it is only two drives of 256K bytes each, with
serial interface.
The communications is serial to 38Kbaud and the protocol is very specific
but fairly simple.
For people with PDP11s this thing without tapes (solid state memory flavor)
is adequate to boot an 11 and even do useful thing s and without the slow
down (seek and rewind) of mechanical tapes it's fast enough to be very
useful.
For VAX11/750 and 730 users thsi is the microcode boot device and also
what is used to run diags. So it's important to have a tape or its analog.
Allison
I wrote:
I saved an image
of the ROM while I had this unit apart but I see you
guys are already way ahead.
If the raw tapes more readily available, I think it would make sense to
pursue the ability to format them. But since it seems even raw tapes
are approaching unobtainium in any quantity, maybe it is time to revisit
the solid-state emulation. Perhaps a job for a BeagleBone or Raspberry
Pi these days...
Chris It is remotely possible that there's an
undocumented "format"
command in the protocol. However, I've heard multiple people
claim that special firmware was required.
I spent some time reverse-engineering
the firmware. There is only
one undocumented opcode, decimal 10, and I haven't yet figured out
what it does, but it definitely doesn't format a tape.
I think I understand the code that writes a block, but of course
there's no code to write an address field. Writing formatter
firmware shouldn't be too hard, but it would be best to have a known
good tape to examine to find out the details of the gap lengths.
6:55 p.m.
allison wrote about the TU58:
For VAX11/750 and 730 users thsi is the microcode boot
device and also
what is used to run diags. So it's important to have a tape or its
analog.
And the HSC50, which is why I'm interested. Though I haven't been able
to obtain an actual HSC50. I've got two HSC70s, but they aren't useful
to me, so I'd be happy to trade one for an HSC50.
Eric
13 Nov
13 Nov
12:46 p.m.
On Sat, Nov 24, 2012 at 4:08 AM Eric Smith <eric at brouhaha.com> wrote:
Is Opcode #10 the "Enter MRSP Mode" or is that something else?
-ethan
It is remotely
possible that there's an undocumented "format" command
in the protocol. However, I've heard multiple people claim that
special firmware was required.
I spent some time reverse-engineering the firmware. There is only one
undocumented opcode, decimal 10, and I haven't yet figured out what it
does, but it definitely doesn't format a tape. 
1:30 p.m.
On Tue, Nov 13, 2018 at 1:47 PM Ethan Dicks via cctalk <
cctalk at classiccmp.org> wrote:
On Sat, Nov 24, 2012 at 4:08 AM Eric Smith <eric at
brouhaha.com> wrote:
I'm still not sure what it does, but it was present in the early 23-089e2
firmware, and gone (acts as no-op) in the later 23-294e2 and 23-389e2
firmware versions.
MRSP mode requires a command 12 (hexadecimal?) according to this page:
http://www.willsworks.net/pdp-11/tu58-emu
but neither the 23-089e2 nor 23-389e2 firmware seem to include that. I
haven't checked 23-294e2.
Maybe there are other firmware versions that include the Enter MRSP
command, or maybe I'm not correctly understanding how the command packet
parsing works in the firmware.
I spent some time reverse-engineering the
firmware. There is only one
undocumented opcode, decimal 10, and I haven't yet figured out what it
does, but it definitely doesn't format a tape.
Is Opcode #10 the "Enter MRSP Mode" or is that something else?
1:31 p.m.
Mark J. Blair (NF6X) has a git repo with various TU58 firmware
disassemblies here:
https://gitlab.com/NF6X_Retrocomputing/tu58firmware
2:02 p.m.
On 11/13/2018 03:46 PM, Ethan Dicks via cctalk wrote:
On Sat, Nov 24, 2012 at 4:08 AM Eric Smith <eric at
brouhaha.com> wrote:
Is Opcode #10 the "Enter
MRSP Mode" or is that something else?
-ethan
Same here,
I asked while I was at DEC and the designer said need a special rig to
do that and the important
things was EOT and BOToptical sensors and the code with the basic
patterns.? The TU58 does not
have an EOT or BOT sensor as it detects end marks on the tape.
I think using disassembled code and adding the sensors its possible to
put down the marks needed.
Never explored it fully to prove that.? It was easier to make a
RAM-based TU58 work alike and at
least one company did a floppy based TU58 work alike.
Allison
It is
remotely possible that there's an undocumented "format" command
in the protocol. However, I've heard multiple people claim that
special firmware was required.
I spent some time reverse-engineering the firmware.
There is only one
undocumented opcode, decimal 10, and I haven't yet figured out what it
does, but it definitely doesn't format a tape.
14 Nov
14 Nov
7:53 a.m.
On Sat, Nov 24, 2012 at 4:08 AM Eric Smith <eric at brouhaha.com> wrote:
I spent some time reverse-engineering the firmware.
There is only one
undocumented opcode, decimal 10, and I haven't yet figured out what it
does, but it definitely doesn't format a tape.
I looked at the firmware disassemblies. Opcode 10 appears to only
point to code (vs NOP) on one version of the firmware
https://gitlab.com/NF6X_Retrocomputing/tu58firmware/blob/master/disassembly…
op_undoc_10:
ld hl,l07e4
call s06a
ld a,018h
ld (r2003),a
jp op_nop
(I think the jump to op_nop is just there to handle regular opcode
processing cleanup)
I have TU58 drives in VAXen (11/750, 11/730, 11/725), a VT103, and one
standalone unit that I pulled from an 11/725 that we bought to strip
for parts to keep our 11/730 running. I really haven't explored the
differences in the firmware since mostly, I've used them as intended,
for loading SA Backup, Microcode patches, and, of course, as Console
Media to boot the 730 and 725.
-ethan
9:33 a.m.
On Wed, Nov 14, 2018 at 8:53 AM Ethan Dicks via cctalk <
cctalk at classiccmp.org> wrote:
On Sat, Nov 24, 2012 at 4:08 AM Eric Smith <eric at
brouhaha.com> wrote:
Well, yes, that's my disassembly. So now you know about as much about
opcode 10 as I do. :-)
I spent some time reverse-engineering the
firmware. There is only one
undocumented opcode, decimal 10, and I haven't yet figured out what it
does, but it definitely doesn't format a tape.
I looked at the firmware disassemblies. Opcode 10 appears to only
point to code (vs NOP) on one version of the firmware
https://gitlab.com/NF6X_Retrocomputing/tu58firmware/blob/master/disassembly…
10:59 a.m.
On Wed, Nov 14, 2018 at 12:35 PM Eric Smith <spacewar at gmail.com> wrote:
Right. But I guess what I trying to was was more about me being
surprised that it's only in what may be _older_ firmware (based on the
assumption, which I _think_ is valid, that Firmware part numbers are
ordered by date).
I'm not surprised there's an undocumented opcode. I'm surprised to
think it's older and was removed. That's not the usual trend.
-ethan
I looked at
the firmware disassemblies. Opcode 10 appears to only
point to code (vs NOP) on one version of the firmware
https://gitlab.com/NF6X_Retrocomputing/tu58firmware/blob/master/disassembly…
Well, yes, that's my disassembly. So now you know about as much about opcode 10 as I
do. :-)
9:59 a.m.
On 11/14/2018 10:53 AM, Ethan Dicks via cctalk wrote:
On Sat, Nov 24, 2012 at 4:08 AM Eric Smith <eric at
brouhaha.com> wrote:
Mine I use to boot a 11/23.? The code, I have both is functionally the
same for PDP11 use.
There are also two version of the TU58 board, one is serial IO (used for
most everything)
and a parallel version used in the PDT110 (I have one).? The parallel
one uses the same
bits and all to match the UART so the same code.? The parallel IO
hardware fakes the
uart framing error error (attention signal).? Very clever as the 8085
code is the same.
Both samples of disassembled code do not relate bits and ports in the
actual drive if they
did then it would be clearer what the code was doing.? The biggest thing
is it reads or
writes a 128byte section of a 512 byte block to tape without doing IO to
the host from
local ram as the CPU is fully consumed doing that.?
The actual TU58 system is 2K of Eprom,? 8155 (256 bytes of ram and 8255
style of port
IO plus a timer), serial port and 8085.? So the core hardware outside of
the read amps
is pretty trivial.? It takes all of the available IO and most of the
interrupt pins plus SID/SOD
pins to implement it.? Its very IO intense for controlling the tape and
reading/writing it.
Best use for the code is extract the parts that do IO to the host (MRSP)
and make a dual
256Kbyte ram (or eprom if IPL for VAX) to replace the physical tape.
For 11/23 use I run a 512K ram board an it boots RT-11 XM copies it to
XD(ram disk)
and then boots it.? Takes about 7 minutes to boot but once there its
fairly fast save for
file IO to tape as seeks are very slow.? The version of a solid state
TU58 I have is not
much faster as the serial rate is the same but seeks are much faster.
Allison
I spent some time reverse-engineering the
firmware. There is only one
undocumented opcode, decimal 10, and I haven't yet figured out what it
does, but it definitely doesn't format a tape.
I looked at the firmware
disassemblies. Opcode 10 appears to only
point to code (vs NOP) on one version of the firmware
https://gitlab.com/NF6X_Retrocomputing/tu58firmware/blob/master/disassembly…
op_undoc_10:
ld hl,l07e4
call s06a
ld a,018h
ld (r2003),a
jp op_nop
(I think the jump to op_nop is just there to handle regular opcode
processing cleanup)
I have TU58 drives in VAXen (11/750, 11/730, 11/725), a VT103, and one
standalone unit that I pulled from an 11/725 that we bought to strip
for parts to keep our 11/730 running. I really haven't explored the
differences in the firmware since mostly, I've used them as intended,
for loading SA Backup, Microcode patches, and, of course, as Console
Media to boot the 730 and 725.
-ethan
24 Nov
24 Nov
2:10 p.m.
I wrote:
As far as I can tell, the TU58 drive hardware should
be perfectly
capable of formatting a tape, unlike e.g. disk drives that use
embedded servo.
After a bit more study, I'm not completely certain that the TU58
hardware can write the "mark" patterns used at BOT, EOT, and in
inter-record gap, without some minor hardware modification. Even more
study required.
25 Nov
25 Nov
5:02 a.m.
New subject: VAX 4000-300 and a CQD-423 Qbus SCSI adapter questions - reply
From powerup:
KA670-A V3.4, VMB 2.12
1) Dansk
2) Deutsch (Deutschland/?sterreich)
3) Deutsch (Schweiz)
4) English (United Kingdom)
5) English (United States/Canada)
6) Espa?ol
7) Fran?ais (Canada)
8) Fran?ais (France/Belgique)
9) Fran?ais (Suisse)
10) Italiano
11) Nederlands
12) Norsk
13) Portugu?s
14) Suomi
15) Svenska
(1..15): 5
Performing normal system tests.
66..65..64..63..62..61..60..59..58..57..56..55..54..53..52..51..
50..49..48..47..46..45..44..43..42..41..40..39..38..37..36..35..
34..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..
Tests completed.
>>show device
DSSI Bus 0 Node 5 (DISK45)
-DIA5 (RF72)
DSSI Bus 0 Node 7 (*)
DSSI Bus 1 Node 7 (*)
UQSSP Tape Controller 0 (774500)
-MUA0 (TK70)
UQSSP Tape Controller 1 (760444)
Ethernet Adapter
-EZA0 (08-00-2B-0B-B3-6E)
>>show qbus
Scan of Qbus I/O Space
-200000DC (760334) = 0000 RQDX3/KDA50/RRD50/RQC25/KFQSA-DISK
-200000DE (760336) = 0B00
-20000124 (760444) = 0000 TQK50/TQK70/TU81E/RV20/KFQSA-TAPE
-20000126 (760446) = 0B40
-20001940 (774500) = 0000 TQK50/TQK70/TU81E/RV20/KFQSA-TAPE
-20001942 (774502) = 0BC0
-20001F40 (777500) = 0020 IPCR
Scan of Qbus Memory Space
>>
Maybe a KA670 firmware limitation, I have not tried this in a 705a.
The card is configured with factory default settings and the only
other Qbus card is a TK70 controller.
Dan Snyder
Butler, PA
1983
days inactive
4166
days old
35 comments
14 participants
participants (14)
-
ajp166＠verizon.net -
allisonportable＠gmail.com -
ard＠p850ug1.demon.co.uk -
cctech＠vax-11.org -
chrise＠pobox.com -
ddsnyder＠zoominternet.net -
eric＠brouhaha.com -
ethan.dicks＠gmail.com -
glen.slick＠gmail.com -
jkunz＠unixag-kl.fh-kl.de -
jules.richardson99＠gmail.com -
pu1bzz.listas＠gmail.com -
shadoooo＠gmail.com -
spacewar＠gmail.com