From: Phil Hobbs
Subject: Re: cheap IR camera?
Date: Thu, 19 Sep 2002 13:07:00 -0400
Organization: IBM T. J. Watson Research Center
NNTP-Posting-Date: 19 Sep 2002 17:07:01 GMT
X-Mailer: Mozilla 4.61 [en] (OS/2; U)
Ian Stirling wrote:
> In sci.electronics.design George Gonzalez wrote:
> > Lately I've been resuscitating a pallet-load of Tek 465 and 475 scopes.
> > Most of the rest have shorted tantalum capacitors from one of the power
> > supplies (-8, -15, +5, +15) to ground. These are a bit difficult to find,
> > as there are many caps sprinkled around the boards. The method I'm using
> > now is to push a few amps into the shorted power bus with an external power
> > supply, then feel around for warm or hot capacitors.
> As others have said, IR cameras are expensive.
> If you have a shorted decoupling capacitor, then if you poke around looking
> at all the cap negatives, then the one you want will have the highest
> positive voltage.
Seconded. You can easily make an audible microvoltmeter from a
low-drift op amp driving a VCO and small speaker. I think Pease has a
wide-range picoammeter circuit like this, based on a couple of 2N3904
base-collector junctions in inverse parallel as the FB elements for the
op amp. Shove a few hundred milliamps through the power supply bus, and
look for tens-of-microvolt to millivolt drops across the grounds. Just
slide the probe around until you get the highest frequency tone.
This frequently works better than looking for drops in the supply
traces, because in the ground plane the voltage peak is localized to the
short circuit itself (although of course it's much smaller). On a
trace, the location of the short and all points further from the supply
will have the same drop, so narrowing it down is a little harder.
Sometimes people use pulsating DC (e.g from a 555 driving a 2N3055 in
series with one supply lead) and look for the short with a signal
tracer. It's nearly the same effect that gives rise to ground loops,
only used for a beneficial purpose.