From: John De Armond
Newsgroups: rec.outdoors.rv-travel
Subject: Re: Sinemaster Inverter Generator:  Anyone have one?
Date: Thu, 17 Mar 2005 14:27:25 -0500
Message-ID: <kflj311t7kbklpbg4jl1g9migb0nnk5hkn@4ax.com>

On Thu, 17 Mar 2005 06:46:06 -0800, Dapper Dave <expurgated@gmail.com>
wrote:


>I know you are talking about a fused circuit here, but I have a related
>circuit breaker question.
>
>I thought circuit breakers would carry only 80% of their rated capacity
>indefinitely. E.g., a 30 amp breaker will trip if it carries more than
>24 amps for an extended period.
>
>It that not true? Do fuses work the same way?

That's one of those simple questions with a really, really complicated
answer.  There are probably thousands of let-thru curves for circuit
breakers and fuses.  For example, a semiconductor fuse will blow
instantly when the rated current is exceeded by a certain percentage
while a motor starting time delay fuse will pass many times its rating
for a considerable period.

The response curve is specified according to what is being protected.
A semiconductor fuse is protecting the semiconductors in the
controller and not the load.  An individual motor fuse is protecting
the windings from thermal overload and not the branch.

An ordinary household breaker is a dual trip device designed to
protect the branch wiring but not anything connected to it.  There is
a magnetic trip that opens the breaker instantly on gross overload
like a short.  I think the magnetic trip is usually set at 6x the
rated current.  There is also a thermal trip that responds to the time
and magnitude of the overload.  The higher the overload the quicker
the trip.

Most all breakers and fuses must carry the rated current indefinitely.
A fuse is somewhat dependent on its mounting to conduct away heat so
it isn't all that precise.  Most breakers will actually carry a bit
more than their rating more or less indefinitely, depending on the
ambient temperature.

The normal way of sizing breakers is according to the wire size in the
branch.  If one can get the response curves from the MFR, however, one
can do better than just protecting the wiring.  Example:

When a compressor in a refrigeration unit locks up the thermal
overload will cycle until power is interrupted.  If it isn't
interrupted fairly quickly the thermal overload overheats and the
contacts weld shut.  At that point the system is entering a "burn out"
condition where the windings burn up, the heat breaks down the
refrigerant into chlorine and fluorine compounds and the whole mess
diffuses throughout the system.

A burn out is frightfully expensive to repair, as the system has to be
flushed with solvent and a burnout cleaner/dryer installed and later
removed.  Usually it's cheaper to just replace the system.  Unless
it's something like a walk-in freezer.

If one chooses a breaker with the right thermal response curve then
the breaker will trip after a few seconds of locked rotor current
before the windings have time enough to start decomposing.  The
breaker has to be slow enough to ride through even the worst starting
inrush and yet trip on locked rotor current.  Fairly easy to engineer
with the breaker curves in hand.

I took that approach with the restaurant.  It has saved me a
bucketfull of money.  I've had two AC locked rotor incidents and one
with the walk-in freezer.  In all three cases the breaker tripped
before the compressors could burn up.  All I had to do was recover the
freon, acid test it for breakdown products, braze in a new compressor,
recharge the system and go.

John