Sjack Abeling Gravity Wheel and the Worlds first Weight Power Plant

[Omnibus]

@Cherryman,

We should try to be as close to the patent as possible. So far @AquariuZ' seems to be the closest (see attached).  Seems the grooves have to be slimmer when compared to those in the patent. Also, the form of the guide (the barrier) is slightly different at the ends (no counter guide (barrier) to form that egg oval appears to exist in the patent). Wonder if we could make one model exactly as shown in Fig.2 of the patent?

[Omnibus]

By the way, how do you scale down the dimensions. @mondrasek was quite right, we have to take care of the dimensions of the rotor.Is @AquariuZ' rotor 200m indeed or I'm reading it wrong?

[mondrasek]

There is an issue with Wm2d that when you scale too small constraints break so much easier, this has to do with accuracy settings. It stops being fun when you get hit by integrator errors all the time.

I would try large scale first and then scale down to actual size, but that's just me

In the accuracy setting change the Integration Error size from the default of .001 to .0001 or .00001 (or lower if you have to).  The sim runs slower, but also much more accurate.  You definitely need the Integration Error size lower than the smallest interference you can handle.  When the sim calculates the move of an object it will allow it to enter other objects it is supposed to collide with a small amount (smaller than the Integration Error size I think).  If that interference object then also moves and increases the overlap of the two objects, that is when the sim pukes.

I don't know how much more I can stress that you need to be modeling at the proper scale!  Think of it this way:  When you scale down a wheel, the Centrifugal Forces also scale down.  But Gravity does not scale down!  So you cannot scale a Gravity Wheel design that uses CF and expect the same performance at all.  Designing in anything but the proper scale is mostly a wasted effort.

Keep in mind that I also draw my CAD models with a wheel around 1000 units in radius, since working within this range of numbers is easier for me.  But I scale the entire thing by .001 before importing to WM2D.  Then the wheel is only 2 meters in diameter.  It's that easy.

M.

[AquariuZ]

Morning, Afternoon, Evening all

I slept through my alarm  14:23 local

Abeling patent papers everywhere * Nightmares about OU bunnies  * Wife wants a divorce

What a mess

@Cherryman to reset velocities or change properties for a group of objects select those objects with left mouse and ctrl for multiples. Then double click on any selected to get a mixed materials property window. Enter zero in the last three velocity fields to reset Vx Vy and Vo (axials and rotational) or if you want to change the materials just do so and all objects will be under same material.

It is a pity that there is no glass material in wm2d but there is ice.

@m: you are right ofcourse, but the scaling down, with a correct integrator error is not something my system can handle, and it slows down to around one frame per minute, even with frame skipping. So... Even though your argument makes perfect sense, I use the large scales to determine the optimal paths and polygon smoothing, and will then throw it in here at the correct scale for you all verify.

@Omnibus: If you cannot select an object because it is blocked from view or overlap just select any object and go to the first field in the properties window which contains a multiple selector field. Look for the object there in the list and select it. Then use Object -> move to front to make it visible. Alternatively move the blocking objects to the back with the same function. Stationary objects should be anchored, not pinned.
The less polygons the better so combine when you can.

I think I will start all over again because I want a wheel as in FIG 8 (maybe 10 meters diameter because M is right) Going to look what Cherryman has been up to..

[AquariuZ]

By the way, how do you scale down the dimensions. @mondrasek was quite right, we have to take care of the dimensions of the rotor.Is @AquariuZ' rotor 200m indeed or I'm reading it wrong?

You are right. I stole the wheel from Cherryman and I really like it. But I will start again and make it a little smaller. I wish I had a desktop CRAY system. (Running Linux ofcourse )