3D printing a structure for an experiment with magnets

[lumen]

Forget that last picture and post. My mistake was to se the sum of all measured torque readings. Each and individual was + and - 50Nm, but the sum was around 2-3Nm. Deviding this on all 72 samples, and virtually nothing is left.


So I have to stick with the alternative on the previous picture (With all the black long rods)
The individual torque of each outer magnets would normally be in addition to the torque around center of the outer rotor, but with the control-rods that keeps these outer magnets in same direction all the time, will reduce the problem to a minimum.
I will be very cautious about what I say now, because I am most probably wrong.
Just by going ahead and try something that my mind says doesn't work feels wrong, and it feels just stupid to actually go building a prototype of something that physics says isn't possible to do...but that is the reason why I will build it


I will come back and report my findings later.


Vidar

With the small rotor moving 1.5 rotations to the larger outer ring rotating 1 the result is zero gain.
Maybe it could be changed so the outer ring with three magnet sets would have the smaller radius and the center ring with the two magnet sets would have the larger radius.
This would have the greater leverage from the increased radius plus the gearing leverage both on the same side providing a gain of 3 to 1 ?

[Low-Q]

With the small rotor moving 1.5 rotations to the larger outer ring rotating 1 the result is zero gain.
Maybe it could be changed so the outer ring with three magnet sets would have the smaller radius and the center ring with the two magnet sets would have the larger radius.
This would have the greater leverage from the increased radius plus the gearing leverage both on the same side providing a gain of 3 to 1 ?

Thanks for the input lumen. I will try that as long the inner and outer magnets doesn't bump into eachother - they have to be synced in 2 to 3 ratio and out of the way of each other :-)


What I have simulated is that there is the same torgue on two gears that has different size. These gears are connected like in the picture below, and rotate in the same direction but at different speed.
The torque is in opposite directions on those gears. However, a small gear attached to a large gear, and apply the same torque on both in opposite directions, I have a feeling that the small gear will win. That is what I have been struggeling with....


This scenario only applies when the outer magnets are externally forced to stay in the same magnetic orientation.
If I use a gear system that is attached to the outer gear, so the outer magnet can rotate in the opposite direction relative to the outer rotor, each magnet will apply approx 50% more countertorque, so the final torque readings on each rotor as a difference of 1.5 times. In that case, the rotors will not go anywhere because the energy is conserved.


Therfor I had the idea of using relatively long rods that is attached to each of the outer magnets, and let them keep the magnetic orientation instead. Then I release the individual countertorques from the actual rotation. That what I hav been thinking. I am probably very wrong, but the thought bugs me.


I also attach a zip-file containng the .FEMM-file that I am working on. If you want to work with this file, just rememer that the inner rotor is 10mm vertically offset from the outer rotor.
Also remember to simulate torque around 0,0, and therfor move the whole structure 10mm up or down depending on which rotor you want to measure.
Also remember to change magnetic orientation for the outer magnets to 0 each time you rotate.

PS! The arrows on the gears on the picture is direction of TORQUE - not rotation!

Maybe someone will try to build it... I am going to anyways

[Low-Q]

Made an animation of the concept. hmmm, animated gif doesn't seem to work that well. The local file works on my computer...

[Low-Q]

Here is a video from my screen...

https://www.youtube.com/embed/IaTF-ofT4AM

[Low-Q]

Lets print out the parts and see what happens next...