>> Personally I suspect something relating to the terminator-power SCSI
>> pin, [...]

> Yes, Sun3/60's short termpwr to ground, unless you modify them with
> some wire snippers.  I remember reading that somewhere--maybe you
> did, too.  I hope not, or the rest of this message will be boring.

I've read something of the sort, certainly - but your prediction about
the rest of your message was overly pessimistic, I'm happy to say. :)

I thought it was only some revs of the -3/60...basically, when I don't
forget about the issue :(, I just consider that pin "dangerous to go
near" if the host is a Sun-3.

> Anyway Seagate drives have that messy ``Power from drive, Power To
> Bus,'' three-way two-jumper sideways-backwards-chicken nonsense, so
> they kind of bypass this problem by powering their resistors off the
> power supply rather than the SCSI bus, sometimes.

The drives in question have four pins that up to two jumpers can be
placed on in various combinations, yes.  Looking at the doco for the
various combinations, it's pretty obvious that one pin is power to the
terminator, one pin is connected to the drive power supply, and two
pins are connected to the SCSI termpwr line.

> However, if your drives worked when terminating themselves, then you
> very likely blew their termpwr ``to-bus'' fuses, both of 'em :) :(.

I've been very careful to never use any of the combinations that
connect power from the drive to the SCSI termpwr pin, specifically
because of this danger.  I either power the terminator off the SCSI bus
or off the drive's power supply...or put the terminator-disable jumper
on (yes, the drives in question have terminator disable jumpers, not
terminator enable jumpers - they're ST51080Ns).

> If so, even after you correct the Sun3 short-of-death, you still
> won't have termpwr.  Maybe you have one of those handy active
> terminators with the LED on it that tells you if there's termpwr?

I have five terminators I've been able to find.  One of them is
explicitly labeled as being an active terminator.  Unfortunately none
of them have power LEDs.  I suppose I could kludge one up if I open up
one of my terminators and drop in a LED and resistor....

> I think there are docs for snipping that stupid Sun-3/60 pin-of-death
> on the picarefy hardware reference faq.

Since the SCSI connector is a right-angle PCB-mount connector, it's as
simple as cutting the appropriate wire between the connector and the
PCB.  Unless it's one of the outer tier of connections, though, this
could mean cutting some others as well and reconnecting them
afterwards.

Though I should check it out; it could be there's a better place to do
it - perhaps some accessible etch run can be cut instead....

Or, I suppose, if you don't mind committing to using a particular
cable, breaking off that pin on the mating connector.  I did this once
with the transmit pins on an AUI Ethernet connector, to set up a
machine that I was *certain* would never send *anything*.

Fortunately it seems to work fine to have the drive do its own
termination and ignore the termpwr line.

Except when I forget and try for separate termination.... :(

If I'm going to be messing with the SCSI stuff on the board, I'd be
tempted to try to solder in a 50-pin header at the spot obviously
designed for one, just behind the SCSI connector, and then add a drive
at the big blank spot beside the memory, next to the VME connector at
the back.  All I'd need to do is find ground, +5, and +12...and I'm
sure I saw a document saying where that could be done, somewhere.

					der Mouse

			       mouse@rodents.montreal.qc.ca
		     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B