minniescar is partly correct in the context of what he calls decline and return.Actually what he is talking about is sinusoidal decadence.This is part
of the problem of resistence I mentioned earlier.Every reference I have been able to dig up on these pmms have the same problem.
I have an idea that I may work on later to overcome this to a partial extent but I doubt there is any real cure for it.
So,let me put it this way

I am not a scientist here so if anyone reads this and I am wrong or off base,please let me know.)
Magnetism is merely a field of energy created by concentrating matter in like patterns so that all particles are 'facing' the same way.This causes
energy to be transmitted in a specific pattern.This energy flows in a constant flux.When that flux is interupted the energy produced in the flux can
be carried away by whatever is interupting that flow.
For instance,say you were to make a coil of wire around a toilet paper roll and attach the two ends to a voltmeter.If you introduced a cylindrical
magnet into the coil you would be introducing a magnetic field as well.In a stationary mode there would be no activity on the voltmeter.However,if you
moved that magnet back and fourth the voltmeter would register a fluctuating voltage.
That voltage would go from tV- to tV+.Otherwise known as a sine wave.The resistance from many factors would cause the movement of that cylinder to
become progressivly harder until it would take more work to produce voltage than the effort would be worth,thus causing the device to slow down and
stop.
A pmm will work.There have been many built.One of them I read about ran for three years before stopping.So,building a pmm is not a real problem.The
problem we need to solve is how do we build one that ctually provides a meaningful return,and is not just another toy.
This is a rather simplistic view of the project,but hey,I'm tired.I've been working all day.Gotta pay for all those magnets somehow.