We tried it before, now I have tried it again. It does not hold, it sticks.It has been manufactured for two years

May be difficult in that way. Two bearings as on my photo above, each made of 4 magnets, may be more stable.

We tried it before, now I have tried it again. It does not hold, it sticks.It has been manufactured for two years
.Jumps off and attracts.Only a gyroscope or use electricity.

It is real finicky and placement/distances of the magnets are not
going to be perfectly symmetrical.
Also, the way you spin it can cause it to jump off if not careful.


I’m not sure what use it has other than a toy. I haven’t found a way
to attach a drive shaft without the 1-ended support.
The force on the shaft or belt pulley destroys it.


Perhaps a two-belt system where the magnetic repulsion gives the belts tension?
(shrug)


Back to this angled magnet thing:


Solid disks makes this a lot easier: (only one field)
To the left of the interaction point magnetic density increases
as you approach the pole. (Peak is at the corner very close to your point)
To the right, magnetic density decreases as the field expands
(a secondary lower peak at ~ 1/4 length of the magnet)
Then it approaches 0 at the meridian.
This will occur in two forms with your stack, one for each magnet, and
a larger (less dense) field of the whole stack.
Density at the poles increases with more magnets in the stack.


The combined field of the stack (at your point of interaction) will behave
similar to a single magnet (solid cylinder or disk)


At the peak: (which is just inside the corners on the flat pole face)
you will have the greatest magnetic density.
The vectored force will try to align the other pole such that the density
is equal on all sides (self centering effect)
At this location, the force will change direction and oppose any change in location
this is easiest observed with magnetic attraction, but can also be observed in repulsion
when you apply a pressure to force the magnet to this point. (inverse)


If you map the field lines with filings or a magnetic viewer you get a visible representation
of the field density gradient. In short, there are more lines to the left of your point
and less lines to the right (per distance/volume increment)
Turning the larger magnet slightly more vertical will cause the centering effect to be more
prominent.
The actual vector of the forces will come in one or two forms.
Depending on the particular method of creating the magnet.
Magnets made by the first method will center in the middle of the pole.
Magnets made by the second method (inductive) will center in a ring
around the pole center. (the location depends on the active state of the magnet/variant)

Assuming a consistent force for experiment:
(ie: don’t move the lower stack between tests)
Maintaining the angle of approach - moving towards or away
from the stack will give you different measurements at different
points. Record the distances and measurements.


This way you can determine a “force gradient” over distance.
Not just change in force between point A and point B.
But changes in force at points A1, A2, A3, etc. leading to B.
Take a change in force from say A1 to A2:
The difference between them is your change in Energy.
(caused by the motion)
The + or - sign of this change will be the direction of motion.
+ energy towards the magnet, - energy away from it.
(from our perspective)


The magnitude of the changes can be added up to get a
total change in Energy between A and B.
This change is a proportional conversion of PE to KE or
vice versa.
The energy input or output is the cause or result of this change.
(in your test this is you physically moving the magnet within the field)

As long as the measurements were taken at the same distance
You can safely assume the proportionality to be consistent at
any distance. (margin of error being less than that of our equipment)


The actual magnitudes of the field, while not known, in most cases, will
maintain the same proportions.
How true this is in theory and practice is material dependent for permanent magnets.
But for the most part your numbers are valid enough to set up a gradient scale.
Then compare to your springs for verification.

moving towards or away
from the stack will give you different measurements at different
points. Record the distances and measurements.

Yes in my video i did exactly that. That is, all that is necessary is to move the magnet by small increments, then every time stop the scales. The distance moved and the scales reading can then be seen from the video. Then i calculated energy for all such small movements, by multiplying the force by the distance, and summed up all the energies, getting the energy at one side.

Another thing is, it is equally important to measure forces when the scales stand still, and when the scales move. Because in both cases, the friction force is to the opposite directions. I did that too, but could only properly measure the force when moving, when moving though all the measurable distance. I did the calculations too, considering the friction, and calculated friction, though with the accuracy the friction could been up to two times more.

I don’t want to make you read this whole lesson
but the first section pretty much gives a summary
of the problem.
Which is why I originally asked
 “what your were trying to measure”.

http://www.brynmawr.edu/physics/courses/109/lectures/SpringScales2.pdf



Measuring acceleration is done while they are moving.
Which requires a time-based perspective.


Measuring (magnetic weight?) happens when they are still.

Measuring (magnetic weight?) happens when they are still.

Yes by that, i measured "magnetic weights" and distances. But i also measured the force during movement, by moving through all the field. In order to measure friction.

What concerns measurement, this experiment is too difficult to do, by separately measuring the force and distance, at all a number of points. Thus the solution is to just move the scales by small increments, away from the neutral point, and stop the scales for a moment every time. Which i did. The forces and distances can then be measured from the video.

It would be desirable to measure forces when moving, from more points too. But the problem, it takes time until the spring of the scales would stretch enough to get force equal to the measured force. Thus i could only properly measure during the movement, when moving through all the field.

I'm not so great experimenter, i'm more like a theoretical thinker like you. Doing things by my own hands is not my desire. I'm sure that some others can make it much more beautiful. But i had to do the experiment, because no one had done that before.

I named the experiment asymmsurf, meaning asymmetry on the surface. Because the small magnet was on the surface of a lid of the box. But it also sounds like surfing the asymmetry of the magnetic field. Which sounds good.

That’s why I was suggesting using a constant force
Such as a counterweight of known weight (measured
on the scale in N)

Then how do you measure, and what do you measure? And how you then calculate the energy?