A place to discuss everything related to Newton Dynamics.
Moderators: Sascha Willems, walaber
by Neo » Mon Jun 18, 2012 5:30 am
Sorry to post a similar quistion...but this time for the hinge joint control.
In my last post, i said I found it difficult to turn the hinge join's pin vector via material callback, but later on it works all right when I change this line:
- Code: Select all
NewtonMaterialSetContactFrictionCoef(material,1.0f,1.0f,0);
NewtonMaterialSetContactFrictionCoef(material,1.0f,1.0f,1);
NewtonMaterialSetContactTangentAcceleration(material,speed,0);
into this:
- Code: Select all
NewtonMaterialSetContactFrictionCoef(material,1.0f,1.0f,0);
NewtonMaterialSetContactFrictionCoef(material,0.5f,0.5f,1);
NewtonMaterialSetContactTangentAcceleration(material,speed,0);
Does the difficulty to rotate the parent body of the hinge joint is related to secondary tangent vector? What's the difference between primary tangent vector and secondary tangent vector? Thx in advance.

-
Neo
-
- Posts: 127
- Joined: Fri May 11, 2012 12:29 pm
- Location: China,Sichuan
by Julio Jerez » Mon Jun 18, 2012 6:52 am
a contact is a force, tha can be decompose into three components.
one is the normal, the other is the tangent at the contact point, the third is the cross product of the first two.
the third is the secundary tangent vector, by rotating the tangent vector around the normal, you can achive a almost perfect columb kenetic frictiom model,
but you can also rotate thne for other porposes like to emulate a tire friction, wher one romal has diffrent frcition than the other
-
Julio Jerez
- Moderator

-
- Posts: 12426
- Joined: Sun Sep 14, 2003 2:18 pm
- Location: Los Angeles
-
by Neo » Mon Jun 18, 2012 7:16 am
Julio Jerez wrote:a contact is a force, tha can be decompose into three components.
one is the normal, the other is the tangent at the contact point, the third is the cross product of the first two.
the third is the secundary tangent vector, by rotating the tangent vector around the normal, you can achive a almost perfect columb kenetic frictiom model,
but you can also rotate thne for other porposes like to emulate a tire friction, wher one romal has diffrent frcition than the other
I see..thx.
So the secundary tangent vector is the one vertical to the direction of the acceleration.
And that's why it's hard to rotate the parent body just by accelerating the child body, when the tangent friction is set to 1.0 .
So in order to turn the vechicle by differ the velocity of each side(they are joints, and both are the child body of the hinge joint), the friction of second tagent vector must be very low, right?

-
Neo
-
- Posts: 127
- Joined: Fri May 11, 2012 12:29 pm
- Location: China,Sichuan
by Julio Jerez » Mon Jun 18, 2012 8:12 am
not very low but lower.
the thread has to be able to produce longitudinal force that is campable to over power the side friction of the same contact point, otherwise it will not turn.
If you look at Thread vehicle, they have groves on teh thread to minimize the size friction
-
Julio Jerez
- Moderator

-
- Posts: 12426
- Joined: Sun Sep 14, 2003 2:18 pm
- Location: Los Angeles
-
by Neo » Mon Jun 18, 2012 2:00 pm
Julio Jerez wrote:not very low but lower.
the thread has to be able to produce longitudinal force that is campable to over power the side friction of the same contact point, otherwise it will not turn.
If you look at Thread vehicle, they have groves on teh thread to minimize the size friction
OK, I think I got the point, thx, Julio.
-
Neo
-
- Posts: 127
- Joined: Fri May 11, 2012 12:29 pm
- Location: China,Sichuan
Return to General Discussion
Who is online
Users browsing this forum: No registered users and 1 guest