"jeff_wisnia" wrote in message
...
Over the years I've had a few thermal fuses in houshold appliances go
open for no apparent reason. When I've replaced them with ones with the
same temperature rating they stayed working fine for years more.

Way back when, my dad used to do forensic engineering - analyzing various
disaster scenes to determine the cause. Thermal switches and circuit
protectors (which were often referred to generically (and incorrectly) as
"Klixons" did indeed fail from time to time. Usually they failed "safe" and
would interrupt the current flow, but on occasion they would "weld"
themselves together and fail to break the circuit. That's why you'll see
two of them in series on devices capable of self-ignition like coffee
makers, electric furnaces, etc. The likelihood of one failing is small, but
not negligible. Regrettably, it's happened often enough that many
manufacturers now use two is series. The probability of both failing
simultaneously *should* be pretty astronomical.

I recall one case dad investigated where the Klixon (and it really *was* a
Klixon branded unit) failed because the thermostat of a large electric
furnace had been wired incorrectly and the Klixon was actually doing the
turning on and off of the unit (very dangerous situation!). The expert sent
by the Klixon Corporation to testify said that the constant cycling of the
protective device as a switch caused a premature failure. He believed the
unit in question had been cycled 1,000's of times before it welded itself in
the closed position. The Klixons had been subjected to a fire of over 1,200
degrees F (determined by the "alligatoring" of the wood beams and the color
of the burned copper wiring). The fire damage to the unit made precise
metallurgical determination of the failure rather difficult.

That's when I learned (30+ years ago) that circuit breakers should not be
used as control devices unless they are specifically rated for that type of
use. I believe most are, nowadays, but that's not always been the case. I
think we've had this discussion before with lots of different opinions.
IIRC, even those breakers labeled SWD (switch duty) are only meant for
low-frequency applications such as turning the lights off and on once a day.

Not sure what the NEC has to say about it - I would think they would require
a switch, other than a circuit breaker, to control a load. However, I do
recall a number of people saying they routinely used the breaker panel to
kill lights, etc. in large stores and shops but it seems odd to me not to
require a switch in the circuit outside the circuit panel. That makes me
assume that the practice is allowed, at least in some jurisdictions. Maybe
there's a low-frequency exception involving lighting loads in large
buildings when using SWD-rated breakers.

A far worse risk (and one that seems to be committed far more often) is
resetting a tripped circuit breaker over and over again. That's definitely
a no-no in terms of asking for the protective device to fail. The fourth
time our kitchen breaker tripped was the signal to me that I needed to
install more 20A lines (eventually 3) to the kitchen. Great SAF. The
kitchen breakers haven't tripped in years. Only the GFCI's have nuisance
tripped in that time.

--
Bobby G.