Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Help  Pulley with Obi rope
#1
So i'm trying to simulate a pulley using the Obi rope asset.

The cubes are rigidbody with a mass of 100 (it seems that the greater the mass the better the dynamic result though I have no idea why). They are attached to the rope using Obi particle attachment (dynamic).

The pulley is made with 3 cyllinders: rigidbody and Obi collider so that the rope can collide with them.

It seems that the rope bounces against the pulley and then it all goes crazy. Any idea how to solve that??

Or is it another better way to simulate a pulley??
Reply
#2
(22-07-2020, 11:59 AM)adrianaa97 Wrote: So i'm trying to simulate a pulley using the Obi rope asset.

The cubes are rigidbody with a mass of 100 (it seems that the greater the mass the better the dynamic result though I have no idea why). They are attached to the rope using Obi particle attachment (dynamic).

Mass determines what percentage of the force is applied to each object during a pair interaction (collision, constraint of any kind). If you had an object with infinite mass and a object with very small mass collide with each other, the object with infinite mass would be considered fixed (won't move) and the object with small mass would retain all the energy it had previous to the collision (assuming perfectly elastic collision). This is what happens in real life, too.

(22-07-2020, 11:59 AM)adrianaa97 Wrote: The pulley is made with 3 cyllinders: rigidbody and Obi collider so that the rope can collide with them.

It seems that the rope bounces against the pulley and then it all goes crazy. Any idea how to solve that??

Particles should not bounce against anything, as restitution in Obi is always zero. Also, should not go crazy in any way, unless you've attached particles inside a collider that's also set to collide with the rope (as explained here:http://obi.virtualmethodstudio.com/tutorials/pinconstraints.html) Can you share a video of this?

(22-07-2020, 11:59 AM)adrianaa97 Wrote: Or is it another better way to simulate a pulley??

There's better approaches to pulleys than using a traditional lumped-mass iterative constraint based solver like Obi, if we're talking about all existing algorithms. See for instance our other cable simulator, Filo: https://assetstore.unity.com/packages/to...tor-133620. But as far as Obi goes, this is the best way to simulate a pulley.
Reply
#3
(23-07-2020, 12:18 PM)josemendez Wrote: Mass determines what percentage of the force is applied to each object during a pair interaction (collision, constraint of any kind). If you had an object with infinite mass and a object with very small mass collide with each other, the object with infinite mass would be considered fixed (won't move) and the object with small mass would retain all the energy it had previous to the collision (assuming perfectly elastic collision). This is what happens in real life, too.


Particles should not bounce against anything, as restitution in Obi is always zero. Also, should not go crazy in any way, unless you've attached particles inside a collider that's also set to collide with the rope (as explained here:http://obi.virtualmethodstudio.com/tutorials/pinconstraints.html) Can you share a video of this?


There's better approaches to pulleys than using a traditional lumped-mass iterative constraint based solver like Obi, if we're talking about all existing algorithms. See for instance our other cable simulator, Filo: https://assetstore.unity.com/packages/to...tor-133620. But as far as Obi goes, this is the best way to simulate a pulley.

Soo, i tried to make it again. And a few days before it worked better. But now i have exactly the same problem. I have know idea why the rope goes flying as soon as a play the simulation instead of staying in the pole.
Reply
#4
(24-08-2020, 02:37 PM)adrianaa97 Wrote: Soo, i tried to make it again. And a few days before it worked better. But now i have exactly the same problem. I have know idea why the rope goes flying as soon as a play the simulation instead of staying in the pole.

Can you describe how you have set it up? For the pulley, a simple rigidbody with a hinge joint works just fine right away, so I'm not sure what could be causing trouble.
Reply
#5
(24-08-2020, 03:28 PM)josemendez Wrote: Can you describe how you have set it up? For the pulley, a simple rigidbody with a hinge joint works just fine right away, so I'm not sure what could be causing trouble.

The cylinders of the pulley are rigidbodies (and also have Obi colliders) and there are (Obi) collisions between the Obi rope and all three cylinders that form the pulley.

My idea was to have the cylinders off the pulley completely fixed and the rope being the only moving object.

Thanks for the previous answer btw Sonrisa
Reply
#6
(24-08-2020, 07:33 PM)adrianaa97 Wrote: The cylinders of the pulley are rigidbodies (and also have Obi colliders) and there are (Obi) collisions between the Obi rope and all three cylinders that form the pulley.

My idea was to have the cylinders off the pulley completely fixed and the rope being the only moving object.

Thanks for the previous answer btw Sonrisa

Hi there,

If the cylinders do not rotate, remember that you will need to have zero friction for the rope to slide smoothly over them. In this case the setup is quite simple and should just work. If you have trouble with it, would it be possible for you to send your scene/project to support(at)virtualmethodstudio.com for us to take a look?

cheers!
Reply
#7
(25-08-2020, 08:03 AM)josemendez Wrote: Hi there,

If the cylinders do not rotate, remember that you will need to have zero friction for the rope to slide smoothly over them. In this case the setup is quite simple and should just work. If you have trouble with it, would it be possible for you to send your scene/project to support(at)virtualmethodstudio.com for us to take a look?

cheers!
 
I've just send it, thank you so much in advance
Reply