@Jetijs,

Pardon me for stepping in here a little bit late. I have just joined this great forum after reading it for months.

I have not built a Perendev motor, but I have build a rotor and stator setup to try to verify the concepts before I poured lots of $$ into a full motor. I have tried to watch every video of the motor and read all the documentation supporting it. As a matter of fact, this was the idea that got me started researching OU phenomenon.

I found lots of videos, but very little scientific study on the concept, with the exception of Tom Bearden who actually tried to explain the physics. In his paper "Permanent Magnetic Motors", he showed some hand drawn charts which explained the concept. So I set out to duplicate and verify it. He showed these diagrams to explain what happens when you angle and shield a cylinder magnet on a rotor:



However this is not the exact result I found experimentally.
I had cut (with great toil) some cylindrical neo magnets at an angle, and cut crescent shields from steel, and mounted them at opposing angles on the rotor and stator. The whole principal is to minimize the magnetic opposition as the magnets approach, and then let the strong force kick in after they hit top dead center.

What I found when I did this, was that there was much less opposition as the magnets approached each other, but just before center, there was a HUGE opposing force, much greater than without the shields. I believe there should be a large peak on that graph before center, and the right side of the graph will also be deformed. I think science will back me up when I say that shields do not eliminate magnetic force, but they REDIRECT it. I wish a had a sexy dynomometer setup where I could plot the forces over distance, but I don't.

And since I tried this I have been searching forums and experimenters' sites to find anyone who has shown experimentally that shielding gives you a NET force in one direction over the entire path of travel. I WANT to find this info so I can continue working on this, but so far, I have not. This and the fact that there has been a 3 year silence from the manufacturers leads me to believe that my instincts were right. You can rearrange the flux lines as much as you want, but when you integrate over the entire path (in whatever dimension you would like), your net is zero.

But keep in mind I'm usually wrong.