What reminds me of the experience I had is the dead short you reported
between -12 and ground. A tantalum cap will make it look like that, and,
once some current runs through them you find out it's not really a short,
since they get hot and catch fire. Once they do that, of course, it opens
the short and you can readily see what's up. I don't recommend putting on a
higher current supply to do this, however, since traces may go before the
cap.
Dick
----- Original Message -----
From: Neil Cherry <ncherry_at_home.net>
To: <classiccmp_at_classiccmp.org>
Sent: Sunday, August 13, 2000 12:38 PM
Subject: Re: [OT] Power and NIC questions
> Richard Erlacher wrote:
> >
> > I'll certainly second the statement about the Tantalum cap's. I once
had a
> > fire in my AMPRO system because of a shorted Tantalum, and, having seen
the
> > smoke emanating from the rear of the PSU (I hd used a PC-style case)
> > immediately assumed the fault was in the often considered flaky power
supply
> > (this was in the mid-80's).
> >
> > In the process of prying the grommet with all the various power leads
out of
> > the supply case, I stuck a 1/2"-wide screwdriver I was using as a
prybar,
> > about 1" into my hand. (think about where you can stick a screwdriver
1"
> > into your hand and not go out the other side ... ) Half a day and over
$1k
> > in medical expenses later, I looked at the Little Board and saw the
burned
> > spot where the cap had been. If Only I'd looked before I lept . . .
<sigh>
>
> Ouch, that's all I have to say about that ....
>
> > About a year ago I once again was powering up an old Ampro Little Board,
> > and, sure enough! ... the -12 was pulled down by a shorted tantalum
bead.
> > Apparently they do not age well.
> >
> > BTW, AUI uses -12 and +12 from the supply on most cards that support it.
> > The coax gets an isolated negative voltage generated locally on the NIC,
> > generally with a little switcher module. These are handy to hang onto
when
> > you're faced with a broken NIC. They generally make a nice -9-volt
source
> > for boxes that have no other negative supply, and that's satisfactory
for
> > RS-232.
>
> Ah. the cap can still be a problem (many tant's on the MB). Well I'll see
> what's going on a little later as I'll check the ub to see if it can see
> a MAC address. This will fill me in on further details of the problem.
>
> --
> Linux Home Automation Neil Cherry ncherry_at_home.net
> http://members.home.net/ncherry (Text only)
> http://meltingpot.fortunecity.com/lightsey/52 (Graphics)
> http://linuxha.sourceforge.net/ (SourceForge)
>
>
Received on Sun Aug 13 2000 - 20:46:05 BST