Subject: Re: scsi question with sparcstation.
To: Andy Ball <andy.ball@earthlink.net>
From: John Gould <johng@powinv.co.uk>
List: port-sparc
Date: 06/21/2006 08:22:57
SCA drives are un-terminated, the termination is provided by 
the SCSI backplane, the ID is provided by the position on the SCSI bus.

See for instance :-

http://www.obsolyte.com/sun_ss5/

There are also links there for the SS5 service manual.

Enjoy...

John.

On Tue, 20 Jun 2006, Andy Ball wrote:

>
> Hello Paul,
>
>  PTS> ok so if it's a more modern standard it won't work yeah?
>
> I think it's a coin toss basically.  I've heard it said that Ultra160
> drives are supposed to be able to fall back to single-ended signalling
> but I don't know how safe it is to assume that they can also work on a
> narrow (8-bit) bus.  I'm interested in finding out though, not least
> because the drive in my own SPARCstation 5 is old, small and screechy.
>
>  PTS> yeah the SCA plug has nothing on it.
>
> But it's still connected to the bus? I think I read somewhere that SCA
> drives ordinarily let the backplane control ID numbers and that the
> drives are unterminated, depending on the backplane to terminate the
> SCSI bus.  If this is true, then even if there's no drive plugged into
> an SCA backplane, it's not safe to assume that there's no termination
> there.
>
> If I get time when I get home tomorrow morning, I'll tear apart my
> SPARCstation5 in an attempt to determine how the SCSI bus is routed.
> One end is bound to be the external SCSI port on the back, perhaps via
> some sort of automatic termination? The bus presumably goes from there
> to the SPARCstation's host adaptor chip and then to a header where it
> jumps off as a ribbon cable.  My guess would be that the ribbon cable
> goes to the CD-ROM drive and then to the SCA backplane where it's
> terminated at the physical end.  It could be the other way around
> though.  Hopefully the CD-ROM isn't on a spur.
>
> - Andy.
>