8 Nov
2006
8 Nov
'06
9:30 p.m.
After swapping out 3 HDA's and 2 servo boards, I had one RA81 working -
mostly. For some reason the system wouldn't see it on port A, but it would
on port B. Oddness.
Anyways, since the day I had it working (and got RT11 loaded on it), the
system will no longer see it. I have moved around and changed some of the
cabling between the drive cabinet and the cpu cabinet, but I don't see that
I've done anything wrong. I put the cabling back the way it was (I could be
mistaken) and still no joy. During the basic UDA50 diag it says there are no
drives connected (same thing for >>>B DU0).
I'm still plowing through the docs (again), but can anyone suggest anything
"obvious" that isn't clear in the docs with regards to connecting an RA81 to
a UDA 50 that a newbie like me could be missing?
I'm up for any input :)
Jay West
8 Nov
8 Nov
9:52 p.m.
I think (It's a long time since my '86 DEC Days...) there was one of the
cables between the drive and the cabinet that was kind of null modem'd.
This could be the problem if at one time the drive was direct cabled to the
UDA and is now going through one cabinet-> drive cable without the
corresponding Cabinet--> uda cable hookup.
Been a long time... I could be wrong. If there's an extra cable in the
wiring try to just eliminate it.
(I assume the correct drive number's on the drive select plug etc.)
On 11/8/06, Jay West <jwest at classiccmp.org> wrote:
After swapping out 3 HDA's and 2 servo boards, I had one RA81 working -
mostly. For some reason the system wouldn't see it on port A, but it would
on port B. Oddness.
Anyways, since the day I had it working (and got RT11 loaded on it), the
system will no longer see it. I have moved around and changed some of the
cabling between the drive cabinet and the cpu cabinet, but I don't see
that
I've done anything wrong. I put the cabling back the way it was (I could
be
mistaken) and still no joy. During the basic UDA50 diag it says there are
no
drives connected (same thing for >>>B DU0).
I'm still plowing through the docs (again), but can anyone suggest
anything
"obvious" that isn't clear in the docs with regards to connecting an RA81
to
a UDA 50 that a newbie like me could be missing?
I'm up for any input :)
Jay West
10:20 p.m.
Bill wrote....
I think (It's a long time since my '86 DEC
Days...) there was one of the
cables between the drive and the cabinet that was kind of null modem'd.
Ah, that makes sense. That doesn't seem to be documented (cept perhaps in
the prints which I haven't looked at yet). So, how does one test (or tell)
if a given cable is non-null-modem'd or null-modem'd (with respect to SDI),
and what should it be end to end? That way I can find the right kind of
cable :)
Thanks!
Jay West
9 Nov
9 Nov
1:47 a.m.
Jay West wrote:
Bill wrote....
There is (AFAIK) only one type of SDI cable and you need an odd number
of them
between drive and controller. Typically you have three: controller to
system cab bulkhead (internal cable), system cab to drive cab (external)
and drive cab bulkhead to drive (internal).
Another source of confusion (although probably not here) is that
although
there are two ports (A & B) only one is active at a time and switching
between them takes a second (or two). In this case I presume that you
only have one cabled up anyway. The confusion usually arose because
some older drives could be cabled to two systems simultanbeously
(although you had to mount as read-only on one if you valued your
sanity and your data).
Antonio
I think (It's a long time since my '86
DEC Days...) there was one of
the cables between the drive and the cabinet that was kind of null
modem'd.
Ah, that makes sense. That doesn't seem to be documented (cept
perhaps in the prints which I haven't looked at yet). So, how does
one test (or tell) if a given cable is non-null-modem'd or
null-modem'd (with respect to SDI), and what should it be end to end?
That way I can find the right kind of cable :)
9:32 a.m.
It was written....
There is (AFAIK) only one type of SDI cable and you
need an odd number
of them
between drive and controller. Typically you have three: controller to
system cab bulkhead (internal cable), system cab to drive cab (external)
and drive cab bulkhead to drive (internal).
Understood, but here's the wrinkle
that confuses me... SOME of my ra81
drives have a cable inside the drive that goes from the personality card to
a block on the back of the drive that presents an AB port on the back of the
drive. SOME of my ra81 drives have a cable inside the drive that connects to
the personality card and goes all the way out the back of the drive a few
feet with connectors on the end. So on those drives there is no "port" on
the back of the drive. In the case of the first style drive, does that
internal to back of the drive bulkhead count as one of the 3? Or are you
supposed to have 3 from there out (back of drive to cabinet bulkhead, then
drive cabinet bulkhead to cpu cabinet bulkhead, then cpu cabinetn bulkhead
to uda50)? I'm not sure if the two styles of drive cabling count differently
for an odd number of cables.
Another source of confusion (although probably not
here) is that
although
there are two ports (A & B) only one is active at a time and switching
between them takes a second (or two).
Understood already :)
Thanks!
Jay
11:20 a.m.
New subject: RA81 cable issue
I see now why the number of cables is an issue. I rang out the bulkhead
connectors and they aren't straight through in the sense that pin1 on one
side doesn't go to pin1 on the other side. It presents a mirror image of the
lines on the other side of the connector - in other words the pins truly are
"straight through", not in the numbering sense.
As a result, I'm sure it DOES matter how many cables you have (or another
way of looking at it, it matters how many bulkhead connectors you go
through). I don't know if the device is expecting a mirror image end to end
or a "straight through" end to end. But obviously the number of bulkhead
connectors you go through affects which order the pins show up on the far
end.
So with regard to the last cable inside the drive sometimes coming out of
the drive as a long cable and sometimes ending as a connector on the back of
the drive... I hooked it up the way that made sense to me based on the above
and diags didn't see the drive. I inserted an additional bulkhead connector
and cable and voila - diags see the drive now. This problem was exacerbated
by the fact that port A on my drive doesn't work, only port B does. This
made the cabling issue seem more mystifying. Interestingly enough, the way I
originally hooked it up is exactly how it is pictured in the manuals. But I
had to add an extra cable/bulkhead connector for the controller to see the
drive.
Thanks for all the input folks. I'm not out of the woods yet though, the
drive still has various problems but I'll address them one at a time.
Jay West
1:48 p.m.
On 9/11/06 14:32, "Jay West" <jwest at classiccmp.org> wrote:
internal to back of the drive bulkhead count as one of
the 3? Or are you
supposed to have 3 from there out (back of drive to cabinet bulkhead, then
drive cabinet bulkhead to cpu cabinet bulkhead, then cpu cabinetn bulkhead
to uda50)? I'm not sure if the two styles of drive cabling count differently
for an odd number of cables.
I don't know the UDA50 but I'm guessing it's the same as the KDA50, in which
case the cable from the KDA50 to cabkit is 1, the SDI cable to the drive
bulkhead (or cabinet bulkhead) is 2 then from bulkhead to drive gives you 3.
We still have to cable up RA7x/9x drives for testing and we use a disk cab
bulkhead to give us the 3rd cable. The cables are physically different too,
at least for RA7x/9x drives. It's been a long while since I last looked at
the back of an RA81 :)
--
Adrian/Witchy
Binary Dinosaurs creator/curator
Www.binarydinosaurs.co.uk - the UK's biggest private home computer
collection?
10 Nov
10 Nov
5:01 p.m.
"Jay West" wrote:
Bill wrote....
When i'm having sdi problems I try (and this is just me) to run one
cable directly from the uda 50 to the drive. This seems to work great
for ra7x's and ra9x's. It might be a trick with an ra81 due to the bulk
(but you might be able to set them up butt to butt as it were - i just
know I'm going to regret typing that)
The rule is the # of sdi cables needs to be odd. 1 or 3.
going direct with one has always worked for me.
-brad
I think (It's a long time since my '86 DEC
Days...) there was one of the
cables between the drive and the cabinet that was kind of null modem'd.
Ah, that makes sense. That doesn't seem to be documented (cept perhaps in
the prints which I haven't looked at yet). So, how does one test (or tell)
if a given cable is non-null-modem'd or null-modem'd (with respect to SDI),
and what should it be end to end? That way I can find the right kind of
cable :) 
6:13 p.m.
On 11/11/06, Brad Parker <brad at heeltoe.com> wrote:
When i'm having sdi problems I try (and this is
just me) to run one
cable directly from the uda 50 to the drive. This seems to work great
for ra7x's and ra9x's.
Haven't (yet) had the pleasure... I've run systems with RA9Xs, but
never had to swap out drives.
It might be a trick with an ra81 due to the bulk
(but you might be able to set them up butt to butt as it were - i just
know I'm going to regret typing that)
Ow! ;-)
The rule is the # of sdi cables needs to be odd. 1 or
3.
going direct with one has always worked for me.
Sure. Simplify, simplify, simplify.
Speaking of going direct, one of my longer-term DEC cabinet projects
is to fire up an RA70 I have in a BA123. I'd like to find some of the
light-weight "in cabinet" style SDI cables - they are, IIRC, grey and
round, not black and square like the cab-to-cab cables I have enough
of.
Does anyone have a) part numbers, or b) a spare cable or two? I
already have the drives and the KDA50s, etc. I suppose one way might
be to rig up something right from the KDA50 directly to the drive,
bypassing the I/O bulkhead, but I don't know enough about the nature
of the cabling to make an end-to-end scratch cable.
Thanks,
-ethan
11 Nov
11 Nov
7:10 a.m.
Ethan wrote:
Speaking of going direct, one of my longer-term DEC
cabinet projects
is to fire up an RA70 I have in a BA123. I'd like to find some of the
light-weight "in cabinet" style SDI cables - they are, IIRC, grey and
round, not black and square like the cab-to-cab cables I have enough
of.
Does anyone have a) part numbers, or b) a spare cable or two? I
already have the drives and the KDA50s, etc. I suppose one way might
be to rig up something right from the KDA50 directly to the drive,
bypassing the I/O bulkhead, but I don't know enough about the nature
of the cabling to make an end-to-end scratch cable.
Thanks,
-ethan
I have two of those cables. They came from the innards of a microVAX 3900,
and they handle nicer than the black stiff rectangular ones. Sorry, they
are used to connect two RA81's to my VAX-11/750. The cables have a male
and a female plug, and they are is +/- 1.5 meters. One of them is labeled:
P/N 17-01482-01
MFG. MC 46461-000
05-15-89
Hope that helps (a little). .
- Henk
This message and attachment(s) are intended solely for the use of the addressee and may
contain information that is privileged, confidential or otherwise exempt from disclosure
under applicable law.
If you are not the intended recipient or agent thereof responsible for delivering this
message to the intended recipient, you are hereby notified that any dissemination,
distribution, or copying of this communication is strictly prohibited.
If you have received this communication in error, please notify the sender immediately by
telephone and with a "reply" message.
Thank you for your cooperation.
9 Nov
9 Nov
2:59 a.m.
Bill wrote:
-----Original Message-----
From: cctalk-bounces at classiccmp.org
[mailto:cctalk-bounces at classiccmp.org] On Behalf Of Bill Pechter
Sent: donderdag 9 november 2006 3:53
To: General Discussion: On-Topic and Off-Topic Posts
Subject: Re: ra81's are irking me
I think (It's a long time since my '86 DEC Days...) there was
one of the cables between the drive and the cabinet that was
kind of null modem'd.
This could be the problem if at one time the drive was direct
cabled to the UDA and is now going through one cabinet->
drive cable without the corresponding Cabinet--> uda cable hookup.
Been a long time... I could be wrong. If there's an extra
cable in the wiring try to just eliminate it.
(I assume the correct drive number's on the drive select plug etc.)
Never heard of an SDI "null-modem" cable! It makes sense however,
as you normally need an *odd* number of cables between the UDA and
the RA8x drive. Since the (red) cable has a different connector at
the UDA side, you often need 2 cables to get to the drive. But that
does not work as you have to get a *third* cable. That "null-modem"
SDI cable would come in handy!
I have not had problems with my RA81 (till now, but it's been a year
that it ran for my 11/44). Like Jay, I have some six RA81's.
One recenty booted VAX-VMS on the VAX-11/750, so that one is fine.
The other 4 ... never had a look! Same goes for the RA60 ...
I always wanted to measure an SDI cable to understand why you need
three of them to get it working... Perhaps Bill has the answer,
I never got round to check them, yet.
Anyway Jay, if it worked before you moved things around, you either
had hardware (HDA, boards) that were about to fail, one of the
cables was "old", perhaps bent a lot and using/moving it developed an
open wire in it, or perhaps a loose connection?
Try some other cables if you have them.
Depending on which cable you connect (the RA81 drive has two, correct?)
you connect to port A or port B. If port A is not responsive, connect
the UDA to the other drive cable. Now port A is connected and port B
is "off". Also possible, if port A *is* connected, check the bulb in
the button!
- Henk.
This message and attachment(s) are intended solely for the use of the addressee and may
contain information that is privileged, confidential or otherwise exempt from disclosure
under applicable law.
If you are not the intended recipient or agent thereof responsible for delivering this
message to the intended recipient, you are hereby notified that any dissemination,
distribution, or copying of this communication is strictly prohibited.
If you have received this communication in error, please notify the sender immediately by
telephone and with a "reply" message.
Thank you for your cooperation.
6469
days inactive
6471
days old
10 comments
7 participants
participants (7)
-
arcarlini＠iee.org -
brad＠heeltoe.com -
ethan.dicks＠gmail.com -
henk.gooijen＠oce.com -
jwest＠classiccmp.org -
pechter＠gmail.com -
witchy＠binarydinosaurs.co.uk