On Mon, 28 Jul 2014 14:25:54 -0400, Spehro Pefhany
<spef...@interlogDOTyou.knowwhat> wrote:
>If you're blowing soldered-in fuses that often, something else is
>wrong. 
>
>Usually if the pigtail fuse goes on a switching supply it's one of two
>things.. the fuse element was fatigued by years of turn-on surges, or
>one of the main switching transistors has gone short, and probably
>taken a diode or two out of the bridge with it. 
The repair FAQ says:
<
http://www.repairfaq.org/sam/smpsfaq.htm#smpsmcp>
   Supply dead, fuse blown - shorted switchmode power transistor 
   and other semiconductors, open fusable resistors, other bad parts.
   Note: actual cause of failure may be power surge/brownout/lightning
   strikes, random failure, or primary side electrolytic capacitor(s)
   with greatly reduced capacity or entirely open - test them before
   powering up the repaired unit. 
I see a few blown fuses in PC supplies.  It's most often the result of
a bulging low voltage capacitor finally deciding to short.  The
switcher should not oscillate with a shorted output, but with a lesser
load, odd things happen.  Usually the switcher starts for about a
second, blows the fuse, and then quits.  If I replace the fuse with a
much larger value, it follows the same pattern, but without blowing
the fuse.  I guess operating with bulging and leaking capacitors isn't
defined in the specs.
Another common blown fuse problem is on overclocked and heavily
modified machines.  The owner has the audacity to actually run the
power supply at it's rated power output.  For example, a 450 watt
power supply actually running at around 450 watts.  200 watt video
cards are common so this is not science fiction.  Most such supplies
will barely produce the rated power for very long and protect
themselves by either a thermal shutdown, or that's lacking, blowing
the fuse.
I'm also seeing blown fuses caused by a mismatch between the inrush
current limiter, and the two big electrolytics in the switchers input
voltage doubler section.  Normally, these electrolytics hold their
charger for a long time (as I've demonstrated by zapping myself).  If
the power supply is turned on and off suddenly, the caps remain mostly
charged and there's no inrush current.  Nearby is the inrush surge
current limiter (PTC thermistor) the is high resistance when cold, low
resistance when hot, and is suppose to cool down very quickly.  That
usually works as expected until some moron glues the thermistor to the
side of the ATX case, or adds a bleeder to the to big electrolytics.
That causes the capacitors to discharge BEFORE the thermistor cools
down.  When the next power glitch appears, the thermistor is still hot
and low resistance, while the caps are discharged.  The inrush current
then blows the fuse.  Breaking the RTV glue between the thermistor and
the ATX case is usually sufficient to prevent further blown fuses.
Grumble... First day of my vacation and I have to drag myself to 
the office to deal with customer emergency.  I should have left town.