Subject: Re: VAX 6420 IS RUNNING!!! (Yeah!)
To: Gunther Schadow <port-vax@netbsd.org>
From: Geoff Roberts <geoffrob@stmarks.pp.catholic.edu.au>
List: port-vax
Date: 05/28/2001 11:50:53
----- Original Message -----
From: "Gunther Schadow" <gunther@aurora.regenstrief.org>
To: <port-vax@netbsd.org>
Sent: Saturday, May 26, 2001 3:17 PM
Subject: VAX 6420 IS RUNNING!!! (Yeah!)


> Hi,
>
> we did it! we did it! Watch in awe:
>
> >>> BOOT
> Initializing system.
>
> #123456789 0123456789 0123456789 01234567#
>
> F   E   D   C   B   A   9   8   7   6   5   4   3   2   1   0   NODE #
>     A   A   .   .   .   M   M   M   M   .   P   P   P   P       TYP
>     o   o   .   .   .   +   +   +   +   .   +   +   +   +       STF
>     .   .   .   .   .   .   .   .   .   .   E   E   E   B       BPD
>     .   .   .   .   .   .   .   .   .   .   +   +   +   +       ETF
>     .   .   .   .   .   .   .   .   .   .   E   E   E   B       BPD
>
> .   .   .   .   .   .   .   .   .   +   .   .   .   .   +   .   XBI D
+
> .   .   .   .   .   .   .   .   .   .   .   -   .   .   +   .   XBI E
+
>
>     .   .   .   .   .  A4  A3  A2  A1   .   .   .   .   .       ILV
>     .   .   .   .   . 128 128 128 128   .   .   .   .   .      512 Mb
>
> ROM0 = V4.00 ROM1 = V4.00 EEPROM = 2.03/4.00  SN = SG231LLR72
>
>
> Loading system software.
>
> ?06 Halt instruction executed in kernel mode.
>     PC     = 200D5882
>     SAVPSL = 041F0601
>     ISP    = 00000300
>
> ?3D Error initializing I/O device
>
> Bootstrap failed due to previous error.
> >>>


Ok, looks like it tried to boot from a HSC that isn't there now.
Looks good to me.  512Mb!  Whoa!  Lucky boy!
And it looks to me like it's a 6440 not a 6420 judging by the 4 'P'
marks.
Major box.

> Thanks to all your handholding and mental support :-) I am thus
> far, finally! I swapped the power & logic box and rewired things
> again on my circuit breaker box. I wanted to be really smart giving
> 110V power to the auxillary power autlets at the back, those that are
> accessible even with closed back door. In order to do that, I would
> have to connect the brown wire and one certain black wire. However,
> when I did that, not even the relay clicked when turning the key!
> STRANGE! I simply swapped the two black wires. Now one of those
> two aux. power outlets at the back is dead, but hey, it started!!!

they are on different phases, and one of the phases is used to supply
110v for
the outlets and the switching and control logic.  Makes sense.

> Before I turned the key, however, I debugged the terminal link.
> I started all over making myself a null-modem cable, D25 female
> to D9 female. Turns out that the D25 adapters. It's simple, just
> make sure you got the following three pins connected:
>
>   D25 female        D9 female
>   ----------        ---------
>   TX   2 ------------ 2 RX
>   RX   3 ------------ 3 TX
>   GND  7 ------------ 5 GND
>
> Then I took my laptop in the garage, terminal program, and configured
> 1200 bps (the default and recommended setting as per VAX 6400 manual.)
> I did see garbage appear on the terminal and responding to input
> from my side. With a bit of luck I found her at 9600 bps 8 bit 1 stop
> no parity. The "break" trick didn't quite work. Anyhow, BINGO I had
> connection.

9600 is pretty much the commonest console setting you will find, very
few leave it
at 1200 unless it's talking to a decwriter (which mine originally did)

> Next was self test and looking at all the colorful LEDs :-). I had
> yellow lights everywhere, except the CI boards and the TKB70,
> physically on the XBI E. Somehow my tape controller seems not right,
> so I  swapped it with a spare one. As soon as I sent this of, I'll
> try it.

Might just need reseating.  And check all the plugs and cables to the
backplane and TK70.

> About the power, I now believe two things, without necessarily
> understanding them:
>
> (1) It is better to conect one phase to one power input line and
> not to bridge over two phases. This is the difference I had to
> my first attempt.

Uh, I think this will depend on which phase you use.  Any one of the
thyristor/rectifier gizmos is capable of running the whole show, but if
you pick the wrong phase, the switching stuff won't work, since it's
powered by
an active/neutral 110v phase.
This was a lot easier/different for me, since I had true single phase
240 to
work with.

> (2) It does matter which phase you pick. Use a brown and one black
> phase. But use the correct black cable or you won't even power the
> relay to switch main power.

Yes, see above.

> However, it may also be that the second power & logic box has a
> problem by itself (this came from the machine that was reported
> to have power problems.)

Seems likely, leaky caps most probably.

> So, it may be that one line is foul. Or,
> may be, they had the p&l boxes already switched, so that first
> time I had the known faulty one. BTW: are those p&l boxes not
> screwed to the cabinet at all? Mine weren't, as if someone had
> messed with those already!

They ARE, so it appears someone had been dinking around.

> I can't wait actually booting the machine. But until then I need
> a boot device. I have one DSSI drive RF73 loose in the cabinet.
> I am hoping to get a KFMSA board with cable kit soon (a honorable
> donor promised them to me, but he is very busy.) I will get
> KDB50 boards (SDI) on Tuesday, but those have neither cables
> nor do I have an SDI drive.

this presents a problem.  I had a look in the back of my 6430 and the
the
KDB50 cables and connectors are unlikely to be something you could fudge
up yourself in a reasonable time, even assuming you had/can get the
connectors,
which are quite high density and unusual.  You are going to need the
cabinet
kit (the cable harness and the outlet plate for the back of the machine)
or you
will get nowhere.  I suggest contacting the source of the KDB50 cards,
they may
have the cab kit somewhere.

> But even if I had drives, I do not
> have a bootable image just yet. Isildur, if your TK50 drives
> are all unuseable and you just can't write a tape, I will
> try to get a hold of someone who can show me how to do it on
> VMS on the one remaining 6400 in our IS department.

Hmm, never tried that, will look into it, bear in mind that TK50 carts
can't be written to
in a TK70 drive unless they are bulk erased first.  They can be read
from however.

> The adventure continues ...

The worst is over I think.

Cheers

Geoff in Oz