On Wed, 12 Jan 2011 05:16:56 -0500, JW wrote:

On Tue, 11 Jan 2011 15:01:48 -0500 PeterD wrote in
Message id: :

...

Regarless the authors have seen similar occurances where turn on
failures were related to a necessary time constant that was not
required on turn off phases. Such a time constant was not present as
the switch contacts seemed to fail, such as might happen after many
years of continious use.

I just ran the ClueMeter over your last post, Peter, and I'm sad
to say the reading was...


There is likely a time constant to prevent undesired turn-ons which is
why the symptoms appear on only with on and not off. That's why the
last paragraph was important! g