New permanent magnet motor on youtube from Roobert33

[mscoffman]

@Omnibus;

Before you modify your experimental setup shown, you should
consider building an electric version of this wheel. The electric version
is solid as it has no mechanical moving pieces except the rotor.

For example you could make magnetic poles by cutting a large iron
washer into quarters and gluing two poles pieces onto the magnet.
Then wrapping some wire coils around the pole pieces. You could
test this by a switch which applies a DC current pulse only during
the sticky spot, to turn the magnet off.

In a perfect world you could charge an electrolytic capacitor during
the wheel rotation and reverse the cap across the coil during the
sticky spot.

The beauty of this is that for a particular voltage, a particular
cap will hold a specific quantity of electricity. So you can tell
by a voltage reading how close the device is to overunity operation.
(plus, it would be neat to see it operating)

:S:MarkSCoffman

[teslaalset]

@teslaalset . so your conclusion seems to be [correct me if I am wrong] that the energy gained by 2 magnets attracting face to face and moving together is greater than the energy needed to separate them by a sideways sliding movement . The difficulty , if this is true of using this to make a motor is as follows . As the 2 magnets attract and move together , The energy produced starts at a very low level , and increases exponentially reaching a maximum at the end of the movement . To separate the magnets , we need a large force initially , decreasing exponentially until the separation is complete . So The two "power curves" do not match .The answer might be to store the energy off attraction in a flywheel , And then ,an instant later , to use this stored energy for the separation along the appropriate path .

@Neptune,
It's difficult to draw conclusions without doing the real measurements myself.
The only thing I tend to believe is that by moving magnets this way there is a surplus of energy.
The patent of Kozeka shows it (confirmed by real measurements), my simulations confirm it and also Omnibus experiments shows this.

As you and Omnibus already indicate (and I agree), the real problem is to convert this principle into a real working mechanical model.

I think there are a few critical parameters that need real accurate tuning as well.
In my initial FEMM simulations, I used  1 cm cube magnets and a displacement of 2 times 50 mm.
(the horizontal axis indicate units of 0.1 mm, representing the steps that my simulations used).
From the third graph in my posted results one can see that there is a maximum of energy gain at a particular distance smaller that 50 mm, rather than take the value of the full stroke of 2 times 50 mm.
You can see that the optimum energy surplus is around 14 mm. 
So, this method can be used to determine the optimized maximum gain and the corresponding optimum stroke if the magnets are given parameters.

There maybe other movement combinations that have larger gain as well, who knows.
If I can find time, I will do some more simulations to see whether I can find these combinations.

[tbird]

omnibus,

Here's where I'm stuck: http://www.youtube.com/watch?v=LSrgTTmHGMQ and that's not at all unusual. That's where everybody gets stuck. Now, when adding a cam you add more parameters, unneeded disbalance and weight let alone friction.

maybe the picture below will give you a new approach.

the idea is to reduce the load of the reset device as much as possible.  by mounting the stator magnet on a bell crank (i think that word works here) with a counter weight, you can reduce the force needed to raise the magnet.  the nice thing about it is it can be applied to a larger unit without any increase in load.  just adjust the weight or length of arm counterweight is mounted on to suit the job.

if adjusted proper, you should be able to blow on the #4 arm and the stator magnet will go up.  cool?

you can determine the height and time by the overlap length of #3 & #4.  once it is in your hand, it will be clear.

the timing in the picture maybe off, but that too you can work out once the parts are in place.

if after you've had a look you don't get it, let me know and i'll try to help.

tom

[Omnibus]

Now, this is getting real close: http://www.youtube.com/watch?v=jftQS2ClBm8 . The gravity effects are basically taken care of and one has to figure out now how to move away that mosquito-like pendulum just a wee bit to have it overcome the sticky spot.

@tbird, hope you'll like this more conservative solution which avoids additional structures.

@mscoffman, your idea is very good but it seems we'll get into the same problem Paul Sprain has -- how is the output energy to be measured. Paul Sprain's device so far is the closest to achieving an OU motor, using brute force -- slight input from the outside to allow for the overcoming of the sticky spot.

[tbird]

Omnibus,

bill wilson used a couple more magnets in repulsion.  he said it worked well.  they don't have to be big, just far enough out not to interfer with the working magnets.

tom