Sjack Abeling Gravity Wheel and the Worlds first Weight Power Plant

[Cloxxki]

Hello,

What did you think of this one ?

Hi Obelix,
I for one, like it as an idea
However, which speed ratio would you suggested for the tower vs. the outer wheel?

For instance, Vwheel = 1, then Vtower=1 for smooth transitions might be troublesome, as vertical lift is 57% faster than the weights average downward speed. Only exactly at 3 o'clock can the wheel match that.

I would rather amend this setup to let the weight on top roll to the side first (losing zero height), and doing work on a wheel with much larger radius, its axle positioned well to the left of the tower.
Still, I doubt that torgue would be able to over come the slight height loss each revolution. Could the tower be geared up to have fewer weights to lift? I doubt it.

[exnihiloest]

...Instabilities in the numerical solving of differential equations (difference equations, rather) are a well known problem among others and it appears wm2d suffer from such problems to an unacceptable extent.

I agree it is quite easy to modelize a perpetual motion with WM2D due to inaccuracies in numerical solving.
But in this thread we have a reverse problem.
We try to modelize a said perpetual motion in WM2D and it doesn't work.

[Omnibus]

@exnihiloest,

We try to modelize a said perpetual motion in WM2D and it doesn't work.

That isn't as clear as you put it.

[spinner]

OMG...
This thread is still alive?
And... there are still people who believe this Abeling stuff could really work?

Ahhh...

[mondrasek]

As for the calculations, I think it'd be interesting to see your methodology first so I'd be curious to see calculations done the way you can do them for the moment.

Well here is an analysis of the sim with the hockey stick design.  I let it run in WM2D with extremely low time steps and integration errors with object frictions turned off but air resistance on low until it settled.  Once it found it's balance point I saved the file out as a DXF and imported into AutoCAD.  As I stated before, the DXF output of WM2D does not allow for the ellipses of the guides to come in to AutoCAD properly.  Instead it imports diamonds that have their corners on the major and minor axis end points for the ellipses.  I drew new ellipses using these corner points, but keep in mind that the ellipse shape in AutoCAD is slightly different than in WM2D.  I then moved the weights into the proper positions in these new guides and used AutoCAD to perform some vector calculations.

I first drew a line in each weight from the center to the bottom of the spheres.  This line is exactly 10 units long and represents the weight vector.  I then drew lines from the center to where the sphere contacts the wheel slot and the guides.  These are used to find the length of the force vector being applied to the wheel slot.  A line of this exact length was then transferred to where the ball is in contact with the wheel slot.  A line is drawn from the center of the wheel to this same point and is dimensioned to show the exact distance that this vector is acting on the wheel from the axle (torque distance).  Drawing a line from the end of the ball force vector perpendicular to this line gives a line (vector) that is the amount of force acting to create torque on the wheel due to this ball.  So now we just need to multiply the distance times the length of this vector and add them up.  This length was also dimensioned by AutoCAD so both the force and the distance are to the precision of a modern CAD system.  If you start with the weight at approx. 1 o'clock and go clockwise calling each torque that results in CW motion + and each that results in CCW motion - you will get:

+1334.4
+3658.8
+3934.5
+540.7
-2335.1
-1779.4
-2165.7
-3117.8
Total = 70.4

This is as balanced as I can get with this analysis.  I believe the reason the total is not exactly zero is easily explained by the fact that I had to substitute guides of slightly different shape (ellipses) and the obviously poor resolution of the DXF exports from WM2D.

The DXF file with the vectors and dimensions is attached.  Also, here is a screen shot of balls 5 and 6 for those who cannot view the DXF and are interested.

Thanks,

M.