# Drag problem

This topic is 4843 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

## Recommended Posts

Hi...first of all sorry that the thread subject wasn't terribly descriptive as I don't know how to describe it in a short way... Now I just deleted all of this because it seems that the angular velocity suddenly becomes really big, causing the quaternion filled with _very_ small values which goes to NaN when normalized. Now I think it's my drag function which is wrong, as I found that putting (0,0,14.8600006) as the angular velocity caused it to output (0,0,-706)!! Here's the naughty thing:
public static void CalculateSphereDrag(Vector3 velocity, Vector3 angularVelocity,
float linearDragCoefficient, float angularDragCoefficient,
out Vector3 linearDrag, out Vector3 angularDrag)
{
//Linear drag force - normalised drag vector * speed*speed * fluidDensity * LINEARDRAGCOEFFICIENT * radius*radius
Vector3 dragDir = -velocity;
float speed = dragDir.Length();
linearDrag = dragDir * speed * fluidDensity * linearDragCoefficient * sphereRadius * sphereRadius;

//Angular drag force -  - normalised drag vector * speed*speed * fluidDensity * ANGULARDRAGCOEFFICIENT * radius*radius
dragDir = -angularVelocity;
speed = dragDir.Length();
angularDrag = dragDir * speed * fluidDensity * angularDragCoefficient * sphereRadius * sphereRadius;
}

and I called it with: Physics.RigidBody3D.CalculateSphereDrag(this.Velocity, this.VelocityAngularBody, this.BoundingSphere.Radius, 10, 0.5f, 0.5f, out linearDrag, out angularDrag); Maybe 10 is too big a fluid density? I know that the drag coefficients have to be 0-1 but does the fluid density have to be within a certain range too? What are the typical values for the parameters? (for, say, air or water) Thanks [Edited by - DrGUI on July 12, 2005 4:47:43 AM]

##### Share on other sites
Also, if you know of any online lists of fluid densities, friction coefficients etc can I have the links please?
Thanks

##### Share on other sites
Hi,

I think the problem with your code is that force is too big for considering the model as linear.
The things you have to do fix it are:
- reduce time steps
- reduce drage force : if you speak in international units, water density is 1000 kg/m3 and air density is 1 kg/m3
- if it's not sufficient, you should check that resulting velocity has same direction as initial one. If not, set resulting velocity to 0.

##### Share on other sites
Thanks AP, I tried reducing my time step however that didn't make any difference, and I can't get access to the resultant velocity as that's in my numerical integrator and it seems like a bit of a hack anyway.

I'm sure it's something wrong with my drag equation...

Thanks again, I would rate you if you weren't anomynous!

1. 1
Rutin
26
2. 2
3. 3
4. 4
5. 5

• 9
• 13
• 19
• 14
• 9
• ### Forum Statistics

• Total Topics
632940
• Total Posts
3009328
• ### Who's Online (See full list)

There are no registered users currently online

×