26 Sep
2014
26 Sep
'14
3:18 p.m.
-----Original Message-----
From: cctalk-bounces at classiccmp.org [mailto:cctalk-bounces at classiccmp.org]
On Behalf Of Paul Koning
Sent: 26 September 2014 20:22
To: General Discussion: On-Topic and Off-Topic Posts
Subject: Re: Value of a Tantalum Capacitor
On Sep 26, 2014, at 3:03 PM, Tothwolf <tothwolf at concentric.net> wrote:
> On Thu, 25 Sep 2014, drlegendre . wrote:
>
>> Now this varies based upon construction of the resistors, but it's
almost
impossible for any typical 2W resistor to fail
dead-short - they can drift
up, they
can drift down - but a hard-short is just not what we
see. And wirewound
parts
tend to act like fuses - they simply go red-hot and
open up if they are
forced to
carry high fault currents.
>>
>> If the PSU isn't built for rapid over-current shutdown, and the only
other
alternative is the 6800uF cap, then I'd put the
money on the cap.
>>
>> Have you checked for build errors, like missing insulators between
boards
&
standoffs / mounts? The kind where all is well (often
for many years)
until the
board coating wears-though and creates a short to the
standoff / mount
point?
>
> Those resistors are large ceramic wirewounds mounted vertically. I
highly
doubt any of those could fail short.
>
> This type of PSU has some form of crowbar circuit, so my money would be
on
either overvoltage causing the crowbar to do its job,
or a failure in the
crowbar
circuit itself. I highly doubt that 6800uF
electrolytic capacitor would
fail short,
and although that always possible, I'd think one
of the semiconductors
would be
more likely to fail short.
>
> From a google search, this very problem is apparently common to this
model
of PSU, but I couldn't find anyone who had
troubleshooted it to the point
of
finding the actual cause.
Just wondering: what tool do you use to show that short? A typical
ohmmeter is
a low current source plus an ammeter. If you connect
such a small power
supply to a large discharged capacitor, it will look like a short for a
while.
For
example, if you have a supply of a couple of volts
current limited to 1
mA, it
will take quite a number of seconds for a 6800 uF
capacitor to charge
substantially. So in a test like that, hook up the instrument and leave
it
connected for a minute. Does it still read 0 ohms?
If yes, then I would
indeed
suspect a short.
paul
I use the diode testing function of my digital multimeter. I have tried
leaving it connected for a while to make sure that the capacitor has
charged. I have just tried again for even longer, just to be sure, and after
a good two minutes it is still showing short.
Regards
Rob