Sjack Abeling Gravity Wheel and the Worlds first Weight Power Plant

[mondrasek]

@eisenficker2000 and @mondrasek,

These are the results from the wheel shifted at 20 degrees CW compared to first @eisenficker2000’s wheel (unlike @eisenficker I’m counting the spheres CW starting from the top one; notation -- same as @eisenficker2000’s):

1. 0.070m    -0.094N    -0.0066Nm
2. 0.070m    -0.250N    -0.0175Nm
3. 0.070m    -0.253N    -0.0177Nm
4. 0.067m    -0.084N    -0.0057Nm
5. 0.044m   +0.162N    +0.0071Nm
6. 0.032m   +0.247N    +0.0079Nm
7. 0.034m   +0.253N    +0.0086Nm
8. 0.054m   +0.203N    +0.0110Nm

Net Torque = -0.0129Nm

Recall that @eisenficker2000 obtained for the net torque the value -0.0088522Nm. It is seen, that the calculated torque stays practically the same (it’s even higher this time but I’m not rubbing it because @mondrasek expected it to be lower). As expected, the above result is very well in concert with the mass-axle observation of this kind of wheel.

P.S. Please see attached also the jpg of the drawing with the arms and the normal vectors drawn in Cyan.

Well, there are a couple errors.  The big one is on ball #4 where you set up your vectors as if the ball was on the wheel (circle).  Actually it is on the ellipse.  The other errors are minor and are likely due to the spherical weights not having been placed exactly at the intersection of the slot centerline and the circle or ellipse path.  I corrected those CAD imperfections before drawing my vectors.

One other thing to point out is how tricky it is to draw lines perpendicular to the ellipse path of the lower ramp.  It appears to have been exploded and is now many small straight segments.  So I drew another ellipse using the major and minor axis and used it to find perfectly perpendicular vectors.  I left it (offset 10 units) in the attached JPG.

The corrected vectors now sum as:

1. 0.0700000m    -0.092345N    -0.0064642Nm
2. 0.0700000m    -0.244703N    -0.0171292Nm
3. 0.0700000m    -0.253717N    -0.0177602Nm
4. 0.0677295m    -0.027306N    -0.0018494Nm
5. 0.0446039m   +0.164653N    +0.0073442Nm
6. 0.0325249m   +0.245266N    +0.0079773Nm
7. 0.0341605m   +0.256491N    +0.0087619Nm
8. 0.0534965m   +0.201651N    +0.0107876Nm

Net Torque = -0.008332Nm

Let me know if you want the DWG or DXF file of this.

[Omnibus]

@mondrasek,

I'm using a small Lisp application called plperp.lsp to draw the perpendicular lines. Unfortunately, not all lines are aligned exactly to begin with so that's another source of error. I think, however, that the order of magnitude of the effect is correct. In order to be sure there are no determinate errors of this kind a program in Lisp has to be written for all those calculations. Wonder if it won't be easier to handle that by a script in SolidWorks? Will post more to discuss this further when I'm ready.

[i_ron]

That's exactly what I am trying to point out with my drawings â€" again and again ... but nobody seems to care about doing the calc 'of enlightment'...

ruggero

ruggero,

Most often what happens is the idea is not fully understood
and then the tendency is to, "let somebody else answer"

I thought that it added an un-necessary level of complication,
replacing an ellipsoid ramp with a circular ramp with its attendant friction and complication so didn't follow through.

But don't be discouraged, we are all in learning mode at this time.

Ron

[mondrasek]

I did a new analysis of the wheel at 0 degrees rotation.  Interesting thing here is that the weight at 12 o'clock is in transition between the circular and elliptical path.  So it either had zero net torque value or a small CCW one.  Results are:

1. 0.0700000m   +0.071373N    +0.0049961Nm
2. 0.0700000m    -0.190919N    -0.0133643Nm
3. 0.0700000m    -0.270000N    -0.0189000Nm
4. 0.0700000m    -0.190919N    -0.0133643Nm
5. 0.0552271m   +0.106415N    +0.0058770Nm
6. 0.0359726m   +0.215966N    +0.0077689Nm
7. 0.0317787m   +0.261549N    +0.0083117Nm
8. 0.0422302m   +0.230257N    +0.0097238Nm

Net Torque with #1 on circle = -0.0139472Nm --> Wheel starts Clockwise
Net Torque with #1 on ellipse = -0.0089511Nm --> Wheel starts Clockwise

[eisenficker2000]

@Mondrasek: Nice results, so what bring  all 9 positions? 

The other errors are minor and are likely due to the spherical weights not having been placed exactly at the intersection of the slot centerline and the circle or ellipse path.  I corrected those CAD imperfections before drawing my vectors.

Those "Errors" were due to the fact that the axles of the weights are 3 mm in diameter and the slots are 4 mm wide. So yes it is easier to move them to  the centerlines to simplify the model.

As for falling weights, I do hope that size does matter