Why does my acceleration scope in an impact simulation accelerate to 1300 m/s^2 for the first .0004s of impact when the velocity is only 4 m/s?

First simulation so I am not too familiar with the inner workings of the model, but I have an impact simulation where a pneumatic cylinder impacts and accelerates a stationary sled on a track. The velocity scope looks good, but the acceleration jumps significantly right at impact. Why is the acceleration so large? (1300 m/s^2)

One of my ideas is regarding the solver. I use a variable step automatic solver at the moment, where there are more data points right at impact. I've tried switching to one of the fixed solvers but I receive a slew of errors.

Any ideas? If I am on the right track with the solver, what solver would you recommend and what additionally would I have to change to have it work?

Thanks in advance!

Answer 1

[David Goodmanson]

Hi Ryan,

Actually the acceleration could have been a lot larger. Suppose del_v = is the change in velocity in the time interval del_t. You know that del_v = a*del_t. With a = 1300, del_t = 4e-4, then delta_v is only .52, which leaves a lot more del_v to come from other time intervals.

Does the model have any details about when the two objects collide? If there is effectively a spring between the cylinder and the sled, then increasing the spring constant decreases the contact time and increases the max acceleration. There is no upper limit in a model like this for how large the acceleration can be if the spring constant is made large enough.

[Ryan DiNapoli]

David,

Thank you for your informative reply. I have a box to box force block parameter, where I have contact stiffness and damping values as inputs. Are there ways to increase the contact time? Even have an impact and stick scenario?

Appreciate your feedback

[David Goodmanson]

As far as impact and stick I don't know since I don't know anything about the blocks you are using. Generally, reducing the contact stiffness should increase the contact time.

[Ryan DiNapoli]

sounds good, thanks again!