68 pin LVD wide drive on fast SCSI II 50 pin bus?

I have to deal with an old system that cannot be replaced and has a fast
SCSI II controller in it. The last time it came in for repair I had to
change out the drive to a Seagate ST23451W "wide SCSI". That required
use of a 68 to 50 pin adapter with high byte (high 9) termination and
all was well. Time passes...

The system is back again and the Seagate drive is dead. The closest
drives I can find kicking around now with enough capacity are used 68
pin LVD wide drives like a Quantum Viking PX09L011. From what little I
can find out - since Quantum>Maxtor>Seagate - has no posted info on
Quantum drives like this one is: this particular drive is definitely LVD
wide.

Will connecting an LVD wide drive to a Fast SCSI II controller be ok?
The FAQ for comp.periphs.scsi seems to imply that it will be ok as long
as high byte termination is in place on the 68 to 50 pin adapter, and
the drive auto-switches between LVD and SE modes.

Rick wrote:
>
> Will connecting an LVD wide drive to a Fast SCSI II controller be ok?

Yes.

> The FAQ for comp.periphs.scsi seems to imply that it will be ok as long
> as high byte termination is in place on the 68 to 50 pin adapter, and
> the drive auto-switches between LVD and SE modes.

Believe the FAQ, with a correctly wired adapter it should work as
expected. If you want to be sure, verify that the DIFFSENS signal is
correctly wired on your adapter: The wide LVD drive must see Pin 16
connected to GND for SE mode. Either the pin must be grounded on the
adapter itself or connected to Pin 11 (GND) on the narrow SE bus.


Micha

Michael Baeuerle wrote in news:mklqm5-8tc.ln1@micha.freeshell.org
> Rick wrote:
> >
> > Will connecting an LVD wide drive to a Fast SCSI II controller be ok?

> Yes.

As in yes, it can be made to work.

>
> > The FAQ for comp.periphs.scsi seems to imply that it will be ok as long
> > as high byte termination is in place on the 68 to 50 pin adapter, and
> > the drive auto-switches between LVD and SE modes.

That also depends on the SCSI host adapter.
If the host adapter doesn't have difsens wired to ground there
may be a problem if the adapter routes that signal just through.
But that can be solved too.

> Believe the FAQ, with a correctly wired adapter it should work as
> expected.

Problem is that it doesn't say "correctly wired" on the adapters.

> If you want to be sure, verify that the DIFFSENS signal is
> correctly wired on your adapter: The wide LVD drive must see Pin 16
> connected to GND for SE mode.

It's not important, provided that there's a Force SE jumper on the drive.
All you need to do is set the Force SE jumper on the drive.

> Either the pin must be grounded on the adapter itself

Huh? Adapters are not grounded.

> or connected to Pin 11 (GND) on the narrow SE bus.

Or any other uneven pin number (except the middle 3) on the 50-pin connector on the adapter. Just make sure that that pin is
actually
grounded (wired to ground) when connected to the host adapter.
All this doesn't matter if all uneven lines are tied together though.

>
>
> Micha

Folkert Rienstra wrote:
> Michael Baeuerle wrote in news:mklqm5-8tc.ln1@micha.freeshell.org
> >
> > [DIFFSENS]
> > Either the pin must be grounded on the adapter itself
>
> Huh? Adapters are not grounded.

I mean "SE only" hardware from the age before Multimode. Such adapters
may have connected all SE-GND Pins (what includes DIFFSENS) together
onboard.


Micha