Roller Coaster simulation - Lost momentum?

Hello,

As a first step to simulate roller coaster designs, I try to make a point follow a curve using Kangaroo, applying mass and gravity. It seems to work going downhill, but for some reason the momentum gained doesn't seem to help going up the next hill.

I feel like I'm missing some fundamental aspect here, but cannot figure out what that would be. I have searched the forums but cannot find anything that I think applies in this case.

So, why does my point stop in the valley? Anyone?

/Ola Jaensson

Attachments:

JaenssonRC_20130906.3dm, 32 KB
JaenssonRC_20130906.gh, 11 KB

Replies to This Discussion

Permalink Reply by Victor Leung on September 6, 2013 at 11:23am

It is the amazing friction of real world.

(Or at least I guess it is)

Try look into the Settings Input. It will take in a Kangaroo Setting Object which contains a friction setting. Check that out, if it doesn't work, yell again.

Permalink Reply by Ola Jaensson on September 6, 2013 at 6:26pm

Thanks, it wasn't the friction this time, but I'll have to deal with it eventually.

Permalink Reply by Daniel Piker on September 6, 2013 at 11:29am

Hi Ola,

When the Strength input of the CurvePull component is set to some positive number, it acts much like the other force components - applying an acceleration at each iteration to pull the point back towards the curve whenever it moves away.

However, when it is set to 0 as you have it, it acts as a hard constraint, and its position gets directly projected back onto the curve. Such hard constraints as they are in Kangaroo at the moment do not preserve momentum, as you have found. (When I first added it, I was thinking mainly of things like restraining the edges of a membrane to a boundary curve, where the sliding momentum is not important, but I guess for roller coasters it very much is!)

I am working on a more consistent treatment of constraints, but for now I can suggest using a high strength (start with 2000) for the CurvePull strength, and maybe lowering the drag in the settings (try 0.5).

Permalink Reply by Ola Jaensson on September 6, 2013 at 6:22pm

Thank you!

A positive CurvePull strength was what I needed.

Now I just need a real world roller coaster with known properties to calibrate against...

RSS