Magnet coil cores, demagnetization power and Lenz delay.

[synchro1]

The magnetite core needs a longer interval to totally demagnetize. The de-escalating crescendo is still returning output power past the area you colored in red.

[synchro1]

"I would wire up transistors in order to close down the timing variability and problems with contact bounce"

Transistors are notoriously sluggish and inefficient compared to mechanical contacts. The separation of the mechanical contact points generates a very clean square wave compared to a gummy transistor.

[MarkE]

Says you!

Synchro1 if you want help designing an experiment that will tell you the truth, then I am willing to do that.  If you just want a food fight, then I will defer.

[MarkE]

The magnetite core needs a longer interval to totally demagnetize. The de-escalating crescendo is still returning output power past the area you colored in red.

And you establish that how?  Where are measurements that support your idea?

[MarkE]

"I would wire up transistors in order to close down the timing variability and problems with contact bounce"

Transistors are notoriously sluggish and inefficient compared to mechanical contacts. The separation of the mechanical contact points generates a very clean square wave compared to a gummy transistor.

Mechanical contacts bounce for up to multiple milliseconds and an also arc for up to many milliseconds.  A well designed transistor power switch operating at these levels can easily turn on and off in a tenth of a microsecond.  In other words a well designed transistor switch will make a stable transition thousands of times faster than mechanical contacts.