[3412] in linux-scsi channel archive
Re: CD-ROM resets SCSI bus on Buslogic BT-958
daemon@ATHENA.MIT.EDU (Jeff Collins)
Thu Feb 26 19:08:28 1998
Date: Thu, 26 Feb 1998 13:19:19 -0800 (PST)
From: Jeff Collins <collinsj@ece.ucdavis.edu>
To: Michael Weller <eowmob@exp-math.uni-essen.de>
cc: Brendan Miller <brmiller@wco.com>, linux-scsi@vger.rutgers.edu
In-Reply-To: <Pine.A32.3.95.980225105317.43378A-100000@werner.exp-math.uni-essen.de>
On Wed, 25 Feb 1998, Michael Weller wrote:
> On Tue, 24 Feb 1998, Jeff Collins wrote:
> >
> > That's active termination. Termination that's enabled with software,
> > jumpers, or switches is active. If it were passive, you'd have 2-3 inline
> ====== Nope!
> > resistor packs that you'd have to mess with. You don't have the problem I
> > described. I don't have any other brainy ideas. Good luck. --jw
>
> While you describe passive termination correctly, active termination does
> not use (passive) resistors but active OpAmps (Operation Amplifiers using
> transistors and such) to 'actively' pull the (floating) bus lines to a
> defined level. If termination is enabled by a jumper/software or not is
> not directly related to active or passive termination at all. External
> terminator plugs using active termination are available since ages (though
> the resistor pack type is cheaper) (usually they are those with control
> LEDs for working termination power)
I did not say that active termination uses resistors. You are correct; it
does not. It uses an active circuit (ie. op-amps). I was merely
describing what a device employing active termination typically looks like
from a black-box point of view. Namely, that active termination (ie power
to the op-amp circuit) is enabled using a jumper, DIP switch, or software
setup as opposed to a socket for plugging in resistor packs.
> One might argue though that those devices using a jumper or software
> setup to enable termination are so decent that they'll use active
> termination anyway, but I don't know if this holds in reality.
This holds for any device that I've ever seen. Note here that up until a
few months ago, I was also (in addition to being a grad student in EE) a
sysadmin here. Believe me, we have *plenty* of SCSI devices kicking
around. I suppose you could design a device with a jumper that would
switch in a (passive) resistor pack and do termination that way. You
would also have to be some kind of moron to propose and implement such a
mess. The active circuit would be much simpler and cleaner. --jw
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Jeff Collins, collinsj@ece.ucdavis.edu, Computer Engineering Research Lab
Electrical and Computer Engineering, The University of California, Davis
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
