Defining the to and fro motion of the link about an axis of rotation
Hello Everyone,
I am trying to make a MBS model of a system. I need to give the to and fro motion of the link (Please refer the attached picture) as input to my model. For example for the first 0.1 second, the link rotates in anti-clockwise direction about the axis of rotation into the paper and reaches the extreme left end position (from Start position). For the rest 1.9 seconds the link starts rotating in the clockwise direction about the axis of rotation into the paper and reaches the extreme right end position, which is 'a' degrees wrt the extreme left end position.
As far I understand this should be not very difficult, but somehow I was not able to acheive the same. I tried to use control points for the motion definition through curves, but since I am new to this software, was not able to acheive the same. So, if someone can help me with an example to build the motion through expression or curves, it would be great.
Thanks for your help in advance!
Warm regards,
Uday
<?xml version="1.0" encoding="UTF-8"?>
Answers
-
Hi Uday,
You can do this by applying displacement/velocity on the joint and using a curve or expression,
I will try to create a sample model for you, Meanwhile, you can check the available tutorials on MotionView user guide.
0 -
Hi Prakash,
Thanks for all the help! I solved it. Hopefully, I will have more questions in future :-)
Regards,
Uday
0 -
Hi Prakash,
So, I tried simulating the To and Fro motion of the link. Please refer to the attached figure. I am simulating 2 seconds. Now, when I simulate only backward motion (To motion) towards extreme left position from start position, my simulation converges and I get results for whole 2 seconds. The same is true, when I simulate only forward motion (Fro motion) towards extreme right position from start position. Now, when I try to simulate the full motion i.e. from start position to extreme left end position and then to the extreme right end position, the simulation does not complete to 2 seconds and gets stuck at around 0.68 sec.
I attached the control points (X amd Y co-ordinate points) in excel, which I am specifying as my input to the motion defined at the axis of rotation of the link.
I did not change anything else in the model apart from defining the full motion. I do not understand, what is the problem.
If you can help, would be great!
Warm regards,
Uday
<?xml version="1.0" encoding="UTF-8"?>
0 -
0
-
Hi Prakash,
Please find attached the log file.
Warm regards,
Uday
0 -
Hi Uday - If the motion is sinusoidal in nature, you can define the motion using an expression
e.g: `60d*SIN(2*PI*TIME)`
Here the body will move 60 degrees in one direction and then back to mean position and again 60 degrees in the opposite direction in one second. (1 hz)
If you want to change the frequency of this osscilation, change the '2*PI' portion. e.g. `60d*SIN(4*PI*TIME)`. Now this oscillation is 2 Hz
Is that what you are looking for ?
Attached is a simple pendulum model that uses such an expression. Run this model, review the animation, and see if it matches your requirement.
Regards
Praful
0 -
Hi Everyone,
@Praful, Thanks for replying and the example model!
I tried to run the simulation with Sine expression. It works better than defining the motion through cuves. However, the convergence problem is recurrent. Though, it happens much later in time around 1.92 seconds than at 0.68 sec. At 1.92 sec the convergence becomes impossible i.e. the calculation of time step takes huge physical time to calculate.
I am simulating the locking of one body into a gear and if the locking is achieved, the velocity of rotating gear becomes zero and two bodies are in the locked position for the rest of the simulation. With all the simulations, I was able to achieve the functionality. Though the simulations does not complete for the specified time.
I have no idea what might be the problem.
Warm regards,
Uday
0
Altair Employee