Sjack Abeling Gravity Wheel and the Worlds first Weight Power Plant

[Omnibus]

Would it just be better if the design was built and tested? After all, trying to get peretual motion on wm2d is harder than real life because not only do you have to solve the problem, but work around the faults of the program.

Alex

In the case of @AquariuZ' egg-shaped model the program hasn't shown faults. Knowing this, it's obviously better to try the various forms of the track, weights, materials, dimensions etc. first on a computer model rather than spend money and effort on innumerable real world models.

[Alexioco]

In the case of @AquariuZ' egg-shaped model the program hasn't shown faults. Knowing this, it's obviously better to try the various forms of the track, weights, materials, dimensions etc. first on a computer model rather than spend money and effort on innumerable real world models.

Oh, well yes, if theres no faults then it is better, hows it coming anyway, I havnt seen it because I dont have the program, wont let me have it...


Alex

[Omnibus]

Oh, well yes, if theres no faults then it is better, hows it coming anyway, I havnt seen it because I dont have the program, wont let me have it...


Alex

No joy yet. If there's anything noteworthy we would post it on youtube for people like you who don't have wm2d.

[mondrasek]

Missed a day due to a real life work trip.  But here is the first of two post for you all:

With regards to WM2D.  It is a very valuable tool that can be very helpful if used correctly.  To use it correctly you have to have some engineering knowledge of the parameters.  Please do not dismiss it as causing errors or having bad algorithms.  That simply is not the case.

Springs.  Using an ideal spring in WM2D can cause issues.  But then again, there is no such thing as an ideal spring in real life!  If so, you would be able to deflect such a spring (in a vaccum) and watch it oscillate forever.  But it will not.  That is because in the case of all real springs there is an internal resistance/friction that causes some sort of dampening.  Using dampeners and/or air resistance in WM2D makes their theoretical (yet impossible) springs actually behave more like real world springs.  So using dampeners is not a bug work around.  Using springs without dampeners can create sim world instabilities that can be corrected with proper engineering knowledge.

Collisions.  Elasticity is an actual property of every real material.  And no material has a value or 0 or 1 elasticity.  So again, if you model with elasticity of 0 and/or 1 you should expect results that are also fictional.

Pins vs. Rigid joint.  WM2D must calculated interactions through an iterative process, one by one.  This induces error and cannot be avoided.  But the error can be minimized.  Using two pins spread as far apart as possible will minimize the error.  I won't go into the reasons again, but just understand that it is a more robust way to model the behavior of fixing the three degrees of rotation we deal with in 2D sims.  Again, not a bug.  Just a mathematical fact.  As with every sim, there are various ways to model each interaction.  It may take some engineering knowledge or experience to figure out the best method.  Modeling exactly how you would expect to build a device in the real world is not necessarily the best way to model something in a sim.  This is because the sim has the limitations of working calculations one by one using a relatively large time step.  In order to get the best results, the time step must be made so small that you would need supercomputer processing speeds in order to see a sim in anything close to real time.  Simulations are what supercomputers are used for!  So don't expect your PC to be able to keep up.  But do learn the best way to model each interaction to minimize the errors.  Note:  Why are solid joints so hard to model?  Because in the real world everything (EVERYTHING!) bends a bit when subjected to a force.  So the rigid joints in the sim are also something that does not exist in the real world.

M.

[mondrasek]

Now for the fun post! 

AquariuZ appears to have begun to investigate this idea already:  What if the missing interaction comes from the weights being gyroscopes?  Take the dumbbell shape and let it spin around the axis of the handle.  Why?  Because gyros still have mass and therefore their whole body will react to gravity and fall and cause a wheel to rotate like any other weight.  But when a force is applied to a gyro the equal and opposite reaction force is in a 90 degree direction to the force on a regular object.  So when you push a gyro it does not move in the direction of the force.  It moves to the side.

In the lifting portion of Abeling's (and others) gravity wheel design the wheel, ramps, etc. are pushing on the weights in order to lift them up again.  If those weights were gyros the reaction of the wheel, ramps, etc would not be in the expected direction, but 90 degrees to it.

How to model a gyro in WM2D?  Replace the current round weights with two concentric circles attached at the center with a pin joint.  The smaller inner wheel is the axle/dumbbell handle and should collide with the wheel, ramps, etc.  The bigger, outer circle should be free to spin and not collide with anything (or just some things).

Do gyros react correctly to applied forces in WM2D?  I do not know.  But I can think of some simple tests to try. 

M.