Hi
I am trying to simulate movement of these gears to see if they match correctly but i am not able to maintain position of the slave gear…it moves under pressure of master gear
this is what happens
can anyone tell me what i am doing wrong ? here is the script GEAR SIMULATION.gh (40.7 KB)
thanks
As the axle of the lower gear is not at the centre of mass of the curve, you need to input this as the corresponding frame to the curve collision component.
Then the anchor at this point will keep the translation of the curve fixed, only allowing it to rotate about the axle.
Like this:
GEAR SIMULATION.gh (42.6 KB)

(also, I changed the slider from integers to floating point, so it moves smoothly instead of jumping by whole degrees)
4 Likes
many thanks for your quick help Daniel !! i really appreciate !!
can i ask how “frame” should be used ? …is it meant to define/maintain the consistancy of a set of points within respect to others on a plane?
thanks
A Frame or Plane input is used in several of the Kangaroo goals that work with rigid bodies.
For example Rigid Body, Rigid Point Set, and the curves in the CurveCollide component.
A rigid body has a position and an orientation, which both get updated at each iteration in the solver. Their other geometry (such as the shape of the curve) is defined relative to this frame and moves with it.
Some rigid bodies can have other particles attached to them besides this frame, but they don’t have to - the curves in the CurveCollide component are kept as NURBS curves, and the solver doesn’t need to calculate updates for the vertices of the curve as particles, it just updates the frame which carries the rest with it.
By placing an Anchor coincident with the frame in your file, we were able to fix the curve in that position while leaving it free to rotate.
1 Like
many thanks Daniel , all clear
one last question : as i see at a closer look the curves and not really colliding , there are 2 scenarios :
-
if i offset the curves to create some cleranace , they are getting closer one another and interacting but there is not real contact between the 2 gears.
-
if i work on ideal involute curves ( no offset ) curves should be matching at any time on the contact point but in this case when simulation with K2 i have slight interference between curves
as i can see the behavior depends and how fast i change the slider that controls the angle of the master gear.
QUESTION : is there a parameter ( example nr of crv points , tolerance etc ) to reduce the behavior above and say to get closer to real contact between curves without approximations described above ?
thanks.