emnewton-like system

[Victorcg]

Hi all. Just trying to make particles moving around some orbit, just like here
http://www.mootzoid.com/html/XsiCorner/emNewton.html

I don't want to use sin,cos functions and want to make it using forces.
So I have tangent force, but particles moving not by the circle traectory, but just like spiral away from center.
I tryed to use the force, to center (F=m*v*v/R), but it semms not ot work.

Any suggestions?
Thank You
Sorry for my English.

[owei]

Hi..!

Well, difficult to tell...but have you took a look at this video:



It might give you some hints...

Perhaps attach your scene, so one could look over it..

cheers,
oliver

[Victorcg]

Thank You for answer. Yes, I watched that video. But there is the problem: all particles are aimed to the same radius, and drag force is not the best solution, I guess... I've achieved some results and will upload scene tomorrow. I've used that formula for centrifugal force (F=m*v*v/R), but it should be divided by 2 more, by my experience (I don't know why), and the velocity is tangent to radius. But at the fast speed some particles still leave their orbit. I forgot my scene at another place, so I will upload it a little bit later. Thank You.

[Victorcg]

Here is the scene. Approximately at 200th frame particles start to leave the orbit.

[owei]

Hi..!

Just a short look over it.
When you check the closest position on geometry, don´t use the cut-off distance. This causes the particles to fly away.
As far as I can see, you consider the mass of the particles, but not the attraction force of your gravitational center. The faster the particles get, the bigger gets the centrifugal force that drives them away, so they will of course leave the orbit over time. You have to build in a force, that compensates this.
Adding simple pointforce might solve this....

EDIT:
And set the velocity BEFORE you set the forces...!!

just some thoughts.. ;)

oliver

[Victorcg]

Thank You. I'll give it a try. I just don't understand what do you mean under "compensate force". I thought that the centrifugal force is the compensate force. Am I wrong?

[owei]

...yes...I just thought of some forces for better control of this effect...but changing the order of giving the force and setting velocity is the biggest part, I think.. ;)

[Victorcg]

OK. I've tryed to disable cutoff distance and replace velocity with forces. Particles stopped to fly away. But I can't get constant orbit for each particle

[owei]

..oK, so I don´t understand what you are after. As all particles travel at the same speed around the center, the particle near the center have to travel a shorter distance, so they will orbit "faster" than the particles on the outer orbit...I´ve attached a modified scene...

Perhaps try to explain once more the effect you try achieve..

What do you mean with constant orbit..? The distance between particle and center (per definition)..?

cheers,
oliver

[Victorcg]

Yes! Exactly! the distance between particle and center. Again sorry for my english.
So if I set the speed about 500 in your scene the orbit changes

[owei]

..ok..that can´t work that way, I think. Your are doing a simulation. That means, that this system has to find it´s balance. By setting the velocity and using this in the next step, the system won´t get in balance. Considering, that the tangent/cross product operation just delivers an aproximation to something like a perfect circular orbit. Those flase values add up and cause the system to expand endlessly...so, you might think of an other solution...
If I got time, I´ll think about that further ;)

[Victorcg]

Thank You! I think I should do it trigonometrically maybe