On Wed, 30 Mar 2005 19:47:47 -0500, Jeff Wisnia
wrote:

Ken Finney wrote:

snipped


I knew I was too close to the problem, and that there had to be a simple
solution. Your solution is likely what we will use.

As for the why, we are testing a switch to ensure it won't set off an
explosive atmosphere. The test method isn't too clear other than we need to
test the switch at maximum current into an inductive load. Rather than try
to justify why this really means "a partially inductive load", we'd rather
worst-case the test conditions.



snip


Your original post didn't turn up on my server so I
may have missed some of the earlier comments.

For specifying switching conditions an inductive
load can be accurately defined as a pure inductance in
series with a resistance.

On switching "break" or contact bounce the inductive
stored energy release is 1/2 L x I squared. This means that
energy storage is directly proportional to inductance so
that worst case testing requires an infinite inductance !!

Testing is meaningless unless the value of the
inductance of the 12A load is specified or switching is
tested with a worst case sample of the device(s) that are to
be switched.

If you need a larger design margin it's better to
test at increased current (energy is I squared) rather than
mess about with non representative inductive components.

Jim