Re: DAT drive headaches

• Previous message: Roger Grosswiler: "Re: IPSEC / VPN - Feature Request for FC4"
• In reply to: J. Epperson: "Re: DAT drive headaches"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

To: fedora-list@redhat.com
Date: Tue, 14 Dec 2004 13:15:58 -0500

On Tuesday 14 December 2004 12:16, J. Epperson wrote:
[...]
>
>The symptoms are consistent with reflectance/termination problems
on
> the bus. Termination for internal drives is usually done at the
> end of the ribbon cable.

Correction here. The scsi cable *is* a transmission line, and
*must* be terminated on both *ends* to control what we in the rf
field call VSWR. A missplaced termination that leaves as little as
4 inches of cable not connected to anything can be enough to make
you start sacrificing virgins, shooting the neighbors cat or
something equally hard on the liguor cabinet.
 
>It's likely that the cable has at least
> two device connectors, so if there's no terminator, it's often
> possible to add one on the last connector. Note that something on
> the bus has to furnish TERMPWR--usually one or more devices are
> jumpered to do so. Check the tape drive doc for jumpers. I'd also
> try running a cleaning tape through two or three times
> consecutively.

By convention, this is generally furnished by the host
controller, although lots of drives do have a jumper so that they
can furnish this TERMPWR. IMO it is a mistake to set this jumper
to on, on any drive. The host controller should furnish this thru
an isolation diode so that no (external to the controller itself)
device connected, is externaly powered, and it has this jumper
set, the resultant nearly 5 volts cannot backfeed into the
computer when the computer is turned off.

Note also that the use of this diode represents a voltage loss,
and for an si (silicon) diode, this is around .7 volts, so TERMPWR
then is not 5 volts, but about 4.3, and that this makes the logic
1 voltage on the buss low enough to have effects on the bus's data
reliability. Any card found with an si diode in that location
really should have it changed to a schotkey(sp) type as the loss
will then be less than .25 volts, helping that noise margin
considerably. The so-called 'active terminations' are much more
capable of handling that in comparison to the usual resistive
terminations consisting of a 220 ohm to TERMPWR, and a 330 ohm to
ground on each data line.

>
>--W.C. Epperson
>***These are my own opinions, but help yourself: I got a plenty.***

Chuckle, I guess you could say I have a few too.

-- 
Cheers, Gene
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
99.30% setiathome rank, not too shabby for a WV hillbilly
Yahoo.com attorneys please note, additions to this message
by Gene Heskett are:
Copyright 2004 by Maurice Eugene Heskett, all rights reserved.
-- 
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: http://www.redhat.com/mailman/listinfo/fedora-list

• Previous message: Roger Grosswiler: "Re: IPSEC / VPN - Feature Request for FC4"
• In reply to: J. Epperson: "Re: DAT drive headaches"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]