How to understand apply_central_force

Scripting Issues
21

@serega-kkz Thanks for your kindly reply.

For the balance, I means the applied force is equal to the constraint resistance.

I suppose the applied force is a continous force, which is always applied on the ruler top in the whole simulation. Let’s have a look for a ruler, the constraint resistance is zero at the beginning, then the applied force is bigger than the constraint resistance, and the applied force make the ruler bending. When the ruler is bending, the constraint resistance increases and finally equal to the applied force. If there is dampling, the velocity of ruler top would turn to be zero, and the ruler would be static as an arc or parabola. Moreover, the ruler arc/parabola and the applied force should be in the same plane, which means if the force is applied in x-y plane, the arc/parabola should also happen in x-y plane. This is the process that takes place in reality. Am I correct?

However, if I apply a continous force (1, 1, 0), two weird things happen:

  1. the tube turn to be twisty, for example the following figure. It should not happen in reality.

    image

  2. Since the force is (1, 1, 0), I expect the arc/parabola should happen in the blue plane in the following figure. However, the arc/parabola swing around in any plane.

    image
    image759×600 6.06 KB

I don’t know how to fix the above two questions.

22

Well, one thing that occurs to me–and I don’t know that this is a significant part of the issue, I will admit–is that in the example situation that you describe, once the ruler is bent to satisfaction, the hand applying force to the ruler doesn’t apply a fixed, arbitrary force.

Instead, I would imagine that it applies a force that is equal-but-opposite to the force of the ruler “attempting” to straighten; more and it would bend the ruler further; less and the ruler would straighten.

As to the matter of twist, I’ll note that a ruler is not quite as simple as the model that you currently have: it’s flat, and thus has matter to either side of the bend, the internal bonds of which likely provide some resistance-force against twisting.

I doubt that these points fully explain what you’re seeing, but perhaps they contribute a little to figuring it out!

23

I think it’s worth noting that in the real world the ruler is fixed from swinging at the point of applied force. Since the finger is not mobile. This also needs to be accepted.

1 Like
24

If the tube is swinging, I think tube and the force should be in one plane. Currently, the tube would appear in any plane. That makes me confuzed.

25

@Thaumaturge @serega-kkz Finally, I decide to simulate the tube as a soft tetrahedral.

1 Like