This one.

The rod magnet polarity must be the same as the center point of the block magnet mounted onto the 45 degree frame.. the rod magnet polarity must pint directly at this center.

here's a side view;

That is, in this model, magnets move in three planes. Have you noticed this?
And not like a two-dimensional classic track V-gate.

Magnetman, great work. It will have immense importance if replicated. Until then i assume that you did succeed and got some results.

I remember talking with you many years ago about magnet motors, maybe what we talked then helped you to find your solution.

The rest i can say is just occam's razor, i don't exclude any unknown effects. That magnet under 45 degrees can provide overunity, yes, my experiment shows that. Either the magnetic field of the ring magnet is not even, then the overunity has the same reason as in my experiment (see the thread below this one, it's also replicated). But also, if the field is even, when the magnet is at 45 degrees, the force in one direction is greater than the force in the opposite direction, this is also measured in my experiment.

Why the rod magnets have to be in the plastic tubes, i don't know. In principle there could be overunity even with fastened rod magnets and one ring magnet. But it is difficult to have a gain of energy that can overcome friction, so everything there may depend on small things. Like these moving rod magnets go more easily over the possible "bumps" in the magnetic field. But again i don't know, i said it's just occam's razor.

But one thing is certain, doing research together helps. What one may find may help others sometimes even in unknown ways.

Hello
in the video the second ring magnet is not visible. The tubes with bar magnets should also have a large distance.
Magnets have arrived. 156 mm diameter knife. 3D printer is already printing.
Here's a description.

Greeting
Lota

First off I just want to say thank you so much to magnetman12003 for sharing this wonderful design! I thought I would have a go at this with what I already had available and so I have not used the magnet sizes that you have used.

So far I have not achieved any continuous rotation so perhaps the sizing is very important to your design but there also seems to be a sticky point to my build which possibly might mean weak or uneven magnet rings?
All pieces printed for the magnet holders were at 45 degrees.

https://www.youtube.com/watch?v=DF1ltIug60w




I have a lot of time on my hands atm so will continue to experiment with the available magnets I already have.

Try placing the rod magnets inside the tubes.  The ends of the rod magnets should face the same polarity as the block magnets. To be certain the rod magnets are close to the block magnets a weight on the opposite end of the rod magnet should be added.  The rod magnets and weight themself’s should float free and not be fixed in any way to the supporting tubes!!!

In that few seconds movie, I can clearly see that the rotor is slowing down.

Please.

Yes i see it too. I don't think it showed continuous rotation, and it could have been given initial speed. All i think it shows is what he considers that he got a gain of energy, not necessarily greater than friction. He never claimed anything more either. And i assume that's true, i know him and know that he is a honest researcher.

Who replicates, btw, this thing looks like delicate. The strength and the weight of the rod magnets, etc, may all matter in minor detail. So no wonder when you cannot replicate. This is the problem of wanting to get all that at once. Why don't you replicate my experiment, much simpler, where you did succeed for certain. But yes, it's already replicated.