Thanks Geoff, I knew I could count on you for helpful hints. I
will measure the 300V box tomorrow. First I'll see what comes in
and out of it. I suspect that it will not have normal output.
In that case I might have to open it and measure things there.
Of course it's hard to tell what potentials I should be measuring
but will get some clarity on that on site.

Now to the voltage and null-wire question. You see, a U.S. 
houshold is fed with a 220 V single phase line. A German 
houshold is fed with a 380 V 3-phase line. The U.S. 220
V line is then split into 2 110 V lines with a shared null
in the middle. At least that's what I learned on this list.
And so, we have 220 as a pseudo-two phase thing. But one or
two phase is all relative, in the end its just a sinus wave
undulating around some point of reference, whichever you take
changes the amplitude, but if you only take the two phases
you get 220V AC. I chose to leave the null-wire open because 
it is much harder for me to tell what would happen if I closed 
it. Having it open will give the VAX exactly the same feeling 
of power inside as your machine has. The shouldn't notice the 
missing null wire. Please correct me if I'm wrong.

Of course I will close protective GND in my final setup. Right
now I'm in the garage to test things out. Later I move it down
into the basement with its very own wire on its very own 
breaker with the right receptacle and everything as good as it
gets.

Geoff Roberts wrote:

> The fault light will only go out after switching to ENABLE and
> completion
> of POST without any hard faults.

how long does this take? Like 10 seconds, 1 minute 3 minutes, 
10 minutes? Might the self test have sensed some abnormality with
the power or does it only check for the busses and boards? Now
I remember having put two RAM boards into the cage that I found
lying loose in the cabinet. These might have been faulty and 
might have caused the fault signal to stay on ... but a RAM
board should not cause a complete power shutdown or should it?


> > Time to hook up a terminal to see whether I would get any
> > messages, after all that's what I was doing this for. I had
> > a box of adapters so I took a D-25 to RJ something adapter
> > with one of those Cisco rollover flat wires hooked into
> > a RJ something to D-9 adapter on my laptop. (I have used
> > the RJ to D-9 adapter before and I know it works as a
> > NULL modem just fine.)
> 
> ISTR having used one of those on a VAX and it didn't work at all.
> 
> > My terminal at 9600 bps
> 
> Could be as low as 2400, but you should see garbage during post if you
> have the connections right.  What sort of term?  Some VT's don't like
> the
> console port levels, which seem to be 5V on at least one of my 6K's,
> this
> meant a VT320 didn't talk to it.

I just used my laptop with tera-term. I use this all the time for
my dayjob work with networking equipment and SBCs lately. I always
run it in 9600 bps first. But of course I don't know what the VAX
console needs. If it had given me more time I would have tested 
lower speeds and swapped cables. I guess an oscilloscope would have
been helpful here to see which is the TX line and at what level.
Do you have an oscilloscope? If so, could you tell from there?
The console port has just 3 pins coming out, 2, 3, and 7. 7 is 
GND and 2 and 3 are RX and TX or vice versa which is to be found out. 
The oscilloscope trick should even allow one to estimate the speed.


> You get nothing from console in standby.  It must be in enable to power
> up
> the console port and start the POST. 

"start the post"? Does that mean I have to manually start the 
test from the console? I was in enable / autoboot, so I would
expect the machine to go as far as it gets without terminal
intervention?

> Ok, sounds like the 300v DC supply has shut down.
> This is either because it has an internal fault which may or may not
> have been caused by the power feed, or one of the other supplies has
> failed, putting it in a fault condition.  If there are no lights on the
> 300v supply, I'd say it was the likely fault area, as it will normally 
> show a FAULT led if it's trying to work.

Where should the light have been? Front or back?

> Start at the front end of the 300v supply (the long skinny one)
> Where the mains feeds into that is the first port of call, I'm assuming
> the
> switchgear box is ok, since it ran that long.  My bet is that leaving
> the neutral
> off may be the cause of the problem, and it's likely to be confined to
> the input
> side of the 300vDC supply box.

> I do, unfortunately, it's on a 240vac SINGLE phase circuit.  The 110v
> 220v 2 phase issue.

So, what did you do with neutral? With a single phase you probably
hooked neutral to L1 (2 or 3) and the one phase to L2 (3 or 1). Then
you did not connect the 4th neutral wire, I mean the real neutral
in the sense of the 3-phase world, the X-axis around which the three
phases undulate? So effectively you have you null open too, or don't
you? I would appreciate to know exactly what you did. Was that detail
on your web site?

> What is the ACTIVE to NEUTRAL voltage on your dryer circuit?
> This could be an undervoltage condition.  The 300v supply needs around
> 220v to operate, much under this might cause higher than usual currents
> to be drawn.

Ah, higher current doesn't sound good. So you say an undervoltage
could cause a component to blow up? That would be consistent with
what happened. But isn't power (P = U x I) what fries components 
rather than higher current at lower voltage? I'm not necessarily 
confident in my judgement here, but I wonder.

> > I still hope the machine isn't lost
> 
> Unlikely to be fatal, any problems are almost certainly confined to the
> power supply, and specifically,
> almost certainly to the 300v DC supply.

I have a spare one which isn't necessarily good either (that 
came out of a machine that was *known* to have similar power 
problems (i.e. shutdown after 15 minutes). I am hesitant to 
throw in the other box, but will do so if I can confirm that 
I made a mistake with the hookup and once I understand how to 
avoid a similar mistake. Here is again what I did:

Black -----------------+------ P1
                      /         |
Black ---------------+        ~230V
                                |
Brown ------------------------ P2

Blue ----------(open)

Green/Yellow---(open)

> > but everything is now
> > much harder.
> 
> Not that much, but you need to sort the mains feed out as a priority.
> 
> Cheers

Indeed, Geoff, you cheered me up, giving me hope!

thanks,
-Gunther

-- 
Gunther Schadow, M.D., Ph.D.                    gschadow@regenstrief.org
Medical Information Scientist      Regenstrief Institute for Health Care
Adjunct Assistent Professor        Indiana University School of Medicine
tel:1(317)630-7960                         http://aurora.regenstrief.org