Sjack Abeling Gravity Wheel and the Worlds first Weight Power Plant

[Omnibus]

If you create a reactor object with no trajectory (via. moving it from one position to another after moving the frame time) it will simply fall, endlessly if there's no ground plane...

Correct. Question is what has really been done here.

[Omnibus]

@Jubjub,

What is the accuracy set to in your sim and can you make it, say, ten or hundred times less grainy (increase the accuracy of calculations) and then run the program again with the new limits?

[Jubjub]

@ Ericdt

It's a common "error" in max with the "explosions". I would suggest that you try the new 2010 version.
I believe they have been working to fix that problem. (I'm not sure if they actually FIXED it or just made it better - it has never been too much of a problem for me)
Also you should read the new help-file regarding reactor as they are adressing this issue.

But in general take a look at your values and specifically distance-tollerances and your substeps/key. If the substeps are not high enough and the motion in the scene is too fast, objects could be interpenetrating with the result that everything explodes when the object try to move away from each other.
Also check for the difference between Havok1 and 3 in regards to what type of simulations you are trying to run.

Hopes this helps 


@ Omnibus

Quote:"So, what you're showing us is a natural behavior of the device, just like the the natural fall of the stone let go from some height w/ respect to ground?"

That is indeed the case 

I believe this whole "D"-idea with the permanent "out-of-balance-wheel" is pretty old. And my guess is that there has been a lot of people trying to build it over the years (centuries). Actually I built one with magnets and what not several years ago but never got it to run.

When I read on this forum about the idea it surfaced again in my memory and I just thought "What the hell, I'll just try and make a 3d-model and see if it works"

First I tried with just one wheel with dumbbells, but it came to a stop after a few revolutions. So I made a copy of that wheel, rotated it 15 degrees and attached it to the second one and voila! It just kept on running.

Sometimes it slows down a bit, but picks up speed again - I believe this is due to both:
1. The way my "sliding holes" are cut and the way the dumbbells are hitting the barrier.
2. The number of substeps I'm simulating. (Default is 10, I used 15 - Higher is better, but more timeconsuming)

But overall I think that the sliding holes position and shape is of minor importance - It will ofcourse help the machine to run better if they are optimized and moved around to "the sweet spot". What this place is I don't actually know though

[Omnibus]

@All,

Check this out (see attached) at default values (air resistance off etc.) which I take to represent ideal conditions. The CCW rotation is observed also with much less grainy accuracy conditions, so it isn't an issue.

[Omnibus]

@Jubjub,

I don't exclude the possibility that such machines had been in existence centuries ago but have been severely suppressed. That's beside the point here, however. What you're saying and presenting deserves utmost attention. First off, that 2D wheel making several full turns isn't to be ignored at all. Never mind it stops eventually.  What are the conditions for that and can you make a vid (sorry to be so pushy but, as you can imagine, I'm really curious to see that)? Also, my version is Autodesk 3ds Max 2009. Do you think it's the right version for what you're showing. If so, probably I should start with your 2D model.