Sjack Abeling Gravity Wheel and the Worlds first Weight Power Plant

[Omnibus]

@Jubjub,

The diameter of your wheels appears to be 2m which is realistic. However, it isn’t clear what the units of mass used are. The masses are some huge numbers and the interesting thing is that the program doesn’t calculate smaller numbers to be proper. Also, some realistic value of elasticity has to be used which is now set to zero for all parts. Wonder if changing to real materials would change that too. Friction is set to 0.1 for the dumbbells but is 0 for the wheels. Not clear what the meaning of that is. Could it be that the program makes calculations separately for each part and when calculations are for the dumbbells friction is considered non-zero but is zero when calculating the wheels. Not clear. I don’t mind simulations for ideal conditions as long as they are meaningful. Of course, one may argue that even if the ideal conditions are properly set the continuous running of the wheel is due to the fact that at these ideal conditions the device is only a very efficient re-distributor of an initially imparted energy. Therefore, seeking some realism won’t hurt, to say the least.

As for the dxf file, it is of the type exported by SolidEdge, for instance. When drawing it in SolidEdge, however, one has to have in mind that millimeters appear as meters in WM2D. Also, in order for the drawing to be usable in wm2d all the grooves have to be part of a closed curve (polygon) â€" see instructions how to do that in SolidEdge by @X00013 in this video: http://www.youtube.com/watch?v=Ad3RZIwxQAM.

[Omnibus]

@Jubjub,

Tried to assign materials to the elements and see the result. Assignimg clear glass to the rotor and the dumbbells (following what Abeling appears to be doing) ostensibly renders well. However, when trying to run the animation the whole thing expodes. Tried also Al disk w/ glass dumbbells as well as other materials. These seem to render and run well as long as one keeps friction of the wheel to 0.00. Also, again, the program doesn't like the dumbells to be of a mass less than around 2000 mass units (whatever these mass units are). You've set it to 200 mass units and, although it renders and runs well (at friction 0.00), the Analyze World considers it un-physical. Aslo, the program doesn't like the wheel to be set to anything less than around 100,000 mass units. What the hell is this I can't fathom. Anyway, I put up with it for now. Now, as soon as you set friction to the wheel, even just a little bit, say, 0.01 units, the whole picture changes. Rendering becomes very slow and of what I've seen wheel exhibits somewhat erratic behavior. Hope you'll get a better picture of that influence with your faster computer. That question of friction is bothersome and has to be understood well. It well may be that setting to much of it is an overkill. Also, ideally non-elastic condition may be reasonable in view of the fact that there are practically no collisions (if one ignores the bumping of the dumbbell into the wall after sliding). But even so, we can imagine some plastilin (putty)-like material that acts ideally non-elastic. Anyway, I think that elasticity thing we'll take care of later after the friction is understood. This post was somewhat repetitious in terms of the friction question but the materials issue is new. Seems choosing materials doesn't change the friction value which has to be specifically indicated through setting it by hand.

Forgot to mention, when materials are assigned at friction 0.00 almost in all cases a slight CCW motion is observed after which the wheel picks up CW motion. This differs from the original case, first uploaded here, whereby no materials were assigned and the wheel goes straight into CW run.

EDIT: I'll add here some more observations. So, after assigning Al to the wheel and stainless steel to the dumbbells, setting friction of the wheel to 0.01 I rendered it for somewhat longer period. What I observed was an initial hesitation to and fro of the wheel after which it picked up the CW rotation. At that the dumbbells that are stuck between other two dumbbells (the size of the dumbbells is such that every other one gets stuck between the previous and the follow-up) slide with a delay . Thus, as I said, even this much of friction imposed on the wheel (0.01) may turn out to be an overkill. I wonder if it's worth making vids of these small steps of the analysis or just stating the above just in words is enough.

Now, tried the above w/ even higher wheel friction value (0.1 inits), matching the dumbbell friction. Similar thing, if not better -- after initial hesitation picks up CW rotation quite decisively. That looks pretty promising.

[LarryC]

Allright! Dusty,

"Then this last video is showing the new upper track layout, with super fast acceleration."

Great job, you are very close to showing what Sjack states: In the topleft of the system the weight is accelerated (like with shot put). The weight is moving faster than the system, and as the system catches the weight it is propelled forward.

Also, in the patent, FIG 4, it shows the reason for the super fast acceleration in the top left.  As all of the weight ball slots are equidistant but the the weight balls are not. The forms/guide force the action.

Isn't it amazing how one builder following the inventors patent and information can out perform a pack of simulators? Some simulations are still rotating backwards

Regards, Larry

[Omnibus]

Allright! Dusty,

"Then this last video is showing the new upper track layout, with super fast acceleration."

Great job, you are very close to showing what Sjack states: In the topleft of the system the weight is accelerated (like with shot put). The weight is moving faster than the system, and as the system catches the weight it is propelled forward.

Also, in the patent, FIG 4, it shows the reason for the super fast acceleration in the top left.  As all of the weight ball slots are equidistant but the the weight balls are not. The forms/guide force the action.

Isn't it amazing how one builder following the inventors patent and information can out perform a pack of simulators? Some simulations are rotating backwards

Regards, Larry

No doubt, @Dusty is doing a great job and, of course, should there be a working model the simulations can only help in clarifying the theory behind it. As for the backward rotation, it may be the natural way of rotation. Don't forget that that rotation is about the center of mass and not only it isn't seen in the sim but also it constantly changes it's position. Sense of rotation is the least of our worries. And, by the way, in @Jubjub's contraption the rotation is in the intuitively expected sense (probably because of the 30 degree offset).

[Omnibus]

As you may expect, I got ambitious and said to myself what if I set friction to 0.3 and elasticity to 0.5as much as the default values are in wm2d. What happened was a very very slow rendering which then showed hesitation but the turning was CW with hick-ups. This is to be expected because I imposed friction on all, including on the axis of rotation. Now I'm doing it again but the friction on the wheel is now 0.01 (retaining the 0.5 elasticity). Will see what happens.

Wish I could show you this (don't know how to make an avi yet). Rotor is hesitating constantly but inching forward. Decelerating and then accelerating.