[LAD] [Somewhat OT] Strange failure mode of a PC

[Folderol]

On Sat, 17 Jul 2010 23:27:46 +0200
Ralf Mardorf <ralf.mardorf at alice-dsl.net> wrote:

> On Sat, 2010-07-17 at 23:23 +0200, fons at kokkinizita.net wrote:
> > The strange thing about that PC power switch is that it is
> > almost never used. The rendering machines wake up by a LAN
> > message from the master, and shutdown by remote control as
> > well. They are never touched.
> 
> That makes it more easy for particulate matter to hurt the switch. Using
> a switch will shorten it's live time, but never using it will do it too,
> because when you use the switch, also dirt gets hurt, not only the
> switch.

Hmmm.
I would say the black deposit was almost certainly silver oxide, and
silver contact switches are not suitable for zero wiping action - indeed
the only safe ones I know of are gold ones. See, it's all your fault.
No use means no wipe :)

However I'm slightly puzzled. I confess to not knowing *precisely* the
start-up operation of these PSUs, but wonder of the contacts are
normally closed, and start signal is from their break action.

Dirty switches and relays don't usually inadvertently close, but they
very frequently randomly open or partially open, and that would
fully account for the symptoms you saw.

There was a time when us poor trainees had to learn all about contact
materials, ratings, actions, yada yada :(

Kids today don't know how well off they are :P

-- 
Will J Godfrey
http://www.musically.me.uk
Say you have a poem and I have a tune.
Exchange them and we can both have a poem, a tune, and a song.