@Floor

If you will, ....please give us all, a fantastic replication, well described, ... and answer any an all [/size]inquires as to it's mechanical operation. Straight up empirical only, IE. videos, measurements, descriptions, calcs and so on. No abstractions here please.I hope my reasoning for this understandable.

A couple of years ago I saw a video of a working machine which was identical to this concept using four or five rotor sections translating a reciprocal magnet motion to a rotating magnet disk. I believe the device had multiple patents over many years thus this is by no means a new technology.


The inventor also mentioned that out of the many working machines he had built only a few demonstrated efficiency high enough to warrant a practical machine. As such I think that rather than trying to reinvent the wheel 10 years too late some thought should go towards a more practical design. I will try to find the name of the inventor and the patents covering this effect if i can.


AC

Hi floor


yes you're right!... I've addressed it in test 2 video and also yesterday made modification to correct it (see below pic) by adding a counter weight on the opposite side of the arm which was intended for that purpose.


Thanks


Luc

The below is test 3 Input Torque curve chart.
Keep in mind these are Foot Grams measured on the 12 inch torque arms.


Luc

  Good work, Luc!
 It is indeed curious.  I've been thinking of ways to convert the action here to a continuous motion, using gravity to drop an upper magnet (at right angles) in close proximity to the lower magnet prior to the "twisting phase" - then using energy from the twisting phase to raise the upper magnet back up to the starting height then turn it 90deg to the starting position. That is the more difficult action to resolve (for me). 


This would be a small model involving oscillating (engage-disengage) as well as rotary (torque) motions ... 
A small test device that would keep moving would be an awesome Christmas present to the world!

  Good work, Luc!
 It is indeed curious.  I've been thinking of ways to convert the action here to a continuous motion, using gravity to drop an upper magnet (at right angles) in close proximity to the lower magnet prior to the "twisting phase" - then using energy from the twisting phase to raise the upper magnet back up to the starting height then turn it 90deg to the starting position. That is the more difficult action to resolve (for me). 


This would be a small model involving oscillating (engage-disengage) as well as rotary (torque) motions ... 
A small test device that would keep moving would be an awesome Christmas present to the world!

Thanks Dr Jones for your post and thinking of ways to convert this action into continuous motion.

Luc

@ Gotoluc
QUOTE from Gotoluc
"Are you recommending the output torque arm (RO) should
stop at the center of the torque cycle? (see chart below)" END QUOTE
    yes

QUOTE from Gotoluc
"If so, why?... other then it would be next to free for the
input torque arm to return to starting point
END QUOTE

No its just because it would be next to free for the input torque
arm to return to starting point.  Thanks
.....................................................................
QUOTE from Gotoluc
"Not sure I understand your suggestion"
END QUOTE

In general, the maximum rotation motivating force is available
during the first part of the rotation....  this is not true during the
VERY FIRST FEW degrees of  rotation away from exact parallel.
                because
The rotation motivating forces toward clock wise rotation are in
balance with the rotation motivating forces toward counter clock
wise, when RO is exactly parallel to SL

This cw to ccw balance shifts greatly.... within a few degrees of
rotation from exact parallel.

While

The linear motivating force (which pushing SL away) is at its absolute
maximum potential when RO and SL are exactly parallel.  This linear
force decreases substantially with a few degrees of rotation.

There is no need to do input against this absolute maximum, linear repelling
force, when the RO out doesn't give back its maximum until after
a few degrees of rotation from exact parallel.

                  thanks luc
                       best wishes
                                  floor

  Mechanical work - which is one form of energy - is defined in basic terms as
Work = Force X Distance moved.


  For a rotating object, we apply a torque to get it to move/rotate.  Also, the distance moved = distance along a portion of a circle called an "arc", which = R x Theta (where the angle Theta is measured in radians= actually, unitless).
Then
Work = Force X Distance moved  = Force x R x Theta  = Torque x Theta, [/size]


So the work = mechanical energy = Torque x Theta, not just Torque alone.


I'm concerned that Theta has been left out in the analysis so far in this thread - and hope that Theta will be included in the future.