🚀 go-pugleaf

Active termination is required if you're using any UltraSCSI or UltraWide
devices on the SCSI chain.  Otherwise the passive terminator is fine.

Both physical ends, and only the ends, of the cable must be terminated,
and any devices between the two ends must have termination disabled.

You can read about termination and other SCSI issues at
http://www.paralan.com/glos.html

As far as the scanner locking up your system, that's caused by using a
non-busmastering SCSI card.  You say you're not using the card that
came with the scanner, so what card are you using?  An Adaptec 2906
or 2902E (or similar busmastering card) should fix the problem.

-- 
Frank

Shannon Jacobs <shanen@my-dejanews.com> wrote in message news:7gfcn0$202$1@tipgw2.t-kougei.ac.jp...
> I have a SCSI PC card with a relatively short cable.  The scanner came with
> a PCI SCSI card that I'm not using and a 25-pin passive terminator, and the
> SCSI PC card came with a 50-pin active terminator.  Right now, the active
> terminator is unused and has no involvement with the scanner, but I'm
> mentioning it in case that's somehow an important piece of data--only the
> passive terminator will fit on the scanner.
> 
> There are various ways to hook up the cables, but I've only discovered three
> configurations that actually work.  I'm pretty sure I've tried all the other
> possibilities, but if you think I might have missed an important one, let me
> know and I'll try that, too, or confirm that I already checked it and it
> didn't work.
> 
> 1)  If the SCSI card is connected directly to the scanner with no terminator
> it works.
> 
> 2)  If the SCSI card is connected to another device and then daisy chained
> to the scanner with no terminator on the scanner, and the other device is
> turned off, then the scanner works.  Actually, this seems to work best and
> is the configuration I am using now.
> 
> 3)  Finally, if the scanner is connected via the daisy chain, but with its
> terminator, then it will work, but ONLY if the other device is turned on.
> 
> However, in general I have to say that so far the SCSI is not very
> impressive.  When using the USB scanner the computer remains basically
> functional, and I can continue doing other work.  When using the SCSI
> scanner, there are long periods when the machine is basically locked up and
> unable to do anything but scan.  As already mentioned, these periods seem to
> be shortest with configuration 2, and the scans also seem to be faster.  I
> don't know if it's an important clue, but the times when the computer is
> locked up are the same times as when the scanner is making a particular
> sound.  During the SCSI scans the CPU remains at 100% continuously, though
> the machine goes back and forth between the active and locked up state.

Shannon Jacobs <shanen@my-dejanews.com> wrote in message news:7gkj59$a0e$1@tipgw2.t-kougei.ac.jp...
> Well, it's been an interesting thread already, but I'm feeling a lot of heat
> without seeing much light so far....  But I'll try to read through some of
> the new FAQs which have been referenced.
> 
> 
> <snip>
> >Both physical ends, and only the ends, of the cable must be terminated,
> >and any devices between the two ends must have termination disabled.
> 
> That's what I thought, but...  Right now it continues to work best with NO
> terminator on the scanner.  I vaguely think this proves the scanner must
> have internal termination of some kind, though I see no way to disable it.
> So why did they provide the passive terminator in the first place?  Beats
> me.
> 
> I'm also concerned about the fact that the passive terminator is the 25-pin
> type.  However, all the cables are 50-wire and the scanner is at the end,
> connected to a normal cable.  Someone in another post suggested it was a Mac
> scanner, but even though some of the new Macs have PCI bus slots, I'm pretty
> sure there is no Mac software here, and that this is intended for use with
> Windows 9x.

Most low-end scanners for both PC and Mac now come with 25 pin SCSI
connectors.  It's called "Apple type" SCSI because Apple was the first company
to use it as a "standard", even though it never existed in the SCSI specs.

Some scanners do provide internal termination.  Ask the manufacturer.


> <snip>
> >As far as the scanner locking up your system, that's caused by using a
> >non-busmastering SCSI card.  You say you're not using the card that
> >came with the scanner, so what card are you using?  An Adaptec 2906
> >or 2902E (or similar busmastering card) should fix the problem.
> 
> The SCSI card I'm actually using is a PC card type.  It's a Japanese brand,
> so this probably isn't going to do much good, but on the card it says:
> Logitec LPM-SCSI2E SCSI-2 Card UltraSMIT.  This is the card that came with
> the active terminator (which I'm not using).  But I think that this implies
> it is supposed to be a faster and newer SCSI interface than the one that
> came with the scanner, though I don't know how to find out about the bus
> mastering.  I'll try digging in the manual, but this is going to be pushing
> my Japanese to new limits...

Low-end scanners don't send enough information to max out even a SCSI-1,
5MB/sec card.  What's important for performance is that the card supports
bus mastering.  You could try an Adaptec 2906 to see if makes any difference,
then return it if it doesn't.

Frank

Shannon Jacobs wrote in message <7gkj59$a0e$1@tipgw2.t-kougei.ac.jp>...
>Well, it's been an interesting thread already, but I'm feeling a lot of
heat
>without seeing much light so far....  But I'll try to read through some of
>the new FAQs which have been referenced.
>
>
><snip>
>>Both physical ends, and only the ends, of the cable must be terminated,
>>and any devices between the two ends must have termination disabled.
>
>That's what I thought, but...  Right now it continues to work best with NO
>terminator on the scanner.

Are you sure that your scanner does not have auto termination?

> I vaguely think this proves the scanner must
>have internal termination of some kind,


That's the implication but you should be able to find info on that in the
scanners specs/documentation.

> though I see no way to disable it.
>So why did they provide the passive terminator in the first place?


That implies even more stringly that the scanner does NOT have internal
termination,  automatic or otherwise.

  Beats
>me.
>
>I'm also concerned about the fact that the passive terminator is the 25-pin
>type.  However, all the cables are 50-wire and the scanner is at the end,
>connected to a normal cable.  Someone in another post suggested it was a
Mac
>scanner, but even though some of the new Macs have PCI bus slots, I'm
pretty
>sure there is no Mac software here, and that this is intended for use with
>Windows 9x.
>
>
><snip>
>>As far as the scanner locking up your system, that's caused by using a
>>non-busmastering SCSI card.  You say you're not using the card that
>>came with the scanner, so what card are you using?  An Adaptec 2906
>>or 2902E (or similar busmastering card) should fix the problem.
>
>The SCSI card I'm actually using is a PC card type.  It's a Japanese brand,
>so this probably isn't going to do much good, but on the card it says:
>Logitec LPM-SCSI2E SCSI-2 Card UltraSMIT.  This is the card that came with
>the active terminator (which I'm not using).  But I think that this implies
>it is supposed to be a faster and newer SCSI interface than the one that
>came with the scanner,


That doesn't follow.

> though I don't know how to find out about the bus
>mastering.


That very likely has nothing to do with the problems.

What exactly does your SCSI chain look like?

Well, it's been an interesting thread already, but I'm feeling a lot of heat
without seeing much light so far....  But I'll try to read through some of
the new FAQs which have been referenced.


<snip>
>Both physical ends, and only the ends, of the cable must be terminated,
>and any devices between the two ends must have termination disabled.

That's what I thought, but...  Right now it continues to work best with NO
terminator on the scanner.  I vaguely think this proves the scanner must
have internal termination of some kind, though I see no way to disable it.
So why did they provide the passive terminator in the first place?  Beats
me.

I'm also concerned about the fact that the passive terminator is the 25-pin
type.  However, all the cables are 50-wire and the scanner is at the end,
connected to a normal cable.  Someone in another post suggested it was a Mac
scanner, but even though some of the new Macs have PCI bus slots, I'm pretty
sure there is no Mac software here, and that this is intended for use with
Windows 9x.


<snip>
>As far as the scanner locking up your system, that's caused by using a
>non-busmastering SCSI card.  You say you're not using the card that
>came with the scanner, so what card are you using?  An Adaptec 2906
>or 2902E (or similar busmastering card) should fix the problem.

The SCSI card I'm actually using is a PC card type.  It's a Japanese brand,
so this probably isn't going to do much good, but on the card it says:
Logitec LPM-SCSI2E SCSI-2 Card UltraSMIT.  This is the card that came with
the active terminator (which I'm not using).  But I think that this implies
it is supposed to be a faster and newer SCSI interface than the one that
came with the scanner, though I don't know how to find out about the bus
mastering.  I'll try digging in the manual, but this is going to be pushing
my Japanese to new limits...