Four bar linkage odd behaviour? (Simscape Multibody)

  • Thread starter Thread starter Rodders
  • Start date Start date
  • Tags Tags
    Linkage multibody
Click For Summary

Discussion Overview

This discussion revolves around the unexpected behavior of a four bar linkage simulation in Simscape Multibody, comparing it with results from Autodesk Inventor and MATLAB. The focus is on the dynamics of the mechanism under specific conditions and the impact of initial positions on its behavior.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested

Main Points Raised

  • The original poster describes a discrepancy between the expected behavior of a four bar linkage and its simulation results in Simscape Multibody, noting that it behaves as a pendulum in Autodesk Inventor.
  • Some participants suggest that the issue might stem from a minor imbalance in forces affecting the simulation results.
  • The original poster refutes the idea that initial positions could cause such a drastic change in dynamics, citing that removing initial positions leads to expected oscillatory behavior around the equilibrium position.
  • Further simulations in MATLAB yield results similar to those in Simscape Multibody, indicating a potential attractor at a specific angle, which the original poster believes is incorrect.
  • The original poster expresses frustration over the situation, highlighting the complexity and challenges of physical simulations.

Areas of Agreement / Disagreement

Participants do not reach a consensus on the cause of the unexpected behavior, with some suggesting force imbalances while the original poster maintains that initial positions should not have such a significant effect.

Contextual Notes

There are unresolved questions regarding the assumptions made in the simulations, particularly concerning the influence of initial conditions and the accuracy of the simulation models used.

Rodders
Messages
3
Reaction score
0
Dear community,

I posted this thread in Mathworks Simscape Community two days ago or so. And I have decided to record a video and post all the information in this forum. This is my first time, so hello and thanks for reading!

Last month I was playing around with Simscape Multibody and I realized something. They have a "four bar linkage" preset to simulate its physical behaviour, so I changed a bit the initial positions and I found this:



I have simulated the same mechanism in Autodesk Inventor and It behaves as I expected (as a pendulum). I supposed that the simmetric position in this case should be the equilibrium position, the attractor.

Then, I decided to create a new simulation in Matlab and this very weird situation popped up. But I do not see why, how, where... I mean, It goes far beyond my understanding.

Some data are:

-Connector Link (Yellow bar):
-Density: 1000 kg/m^3
-Volume: 10x2x0.5 cm^3

-Rocker Link (Red bar):
-Density: 100 kg/m^3
-Volume: 25x2x0.5 cm^3

-Crank Link (Blue bar):
-Density: 100 kg/m^3
-Volume: 25x2x0.5 cm^3

Conditions:
-No friction.
-No external foces.
-Just gravitation acting.
-State targets (initial positions): specified in the video. Sorry about the size. Please watch full screen in 1080.

I would appreciate your comments on this issue, please, I really need it...

Thanks again!
 
Engineering news on Phys.org
So, what exactly is the issue? Looks to me like the simulation is not reflecting reality due to some minuscule unbalance in forces.
 
Thanks for your feedback Mech_Engineer, really appreciate it.

I have refused the possibility you mention because, when initial positions are eliminated (see 1:28 in the embedded video) the linkage behaves as expected. It oscillates around the equilibrium position (the attractor), the simmetric position. I cannot believe that just by changing its initial position, the physical behaviour could vary in such a different way...

I have simulated dynamically the linkage in Autodesk Inventor, and it behaves as shown in the video below:



It definitely looks better... I refuse to believe that a change in initial position might affect (in this way) the dynamics of the mechanism.

To double-check the dynamics of this mechanism, I have simulated in Matlab the mechanism and I have come across the following result:



My results are quite similar to Simscape Multibody (similarly wrong?).
As you can see (adjusting the resolution to 1080p), it seems there is an attractor when the angle on the upper-left is set to 0.7 Rad, the mechanism shows a minimal displacement. It seems that this position is its equilibrium position, but this is definitely wrong, as it is supposed to be the simmetric position.

I feel quite lost in this regard... Sometimes physical problems are quite frustrating and mind blowers. This one is blowing my self-esteem! XD

Thanks a lot for your time, hope anyone can help me...
 
Solved in Mathworks Community (link to the thread).

Thanks to everyone
 

Similar threads

Four-Bar Parallel Linkage Pendulum
  • · Replies 62 ·
3
Replies
62
Views
6K
Understanding the four bar linkage DOF
  • · Replies 3 ·
Replies
3
Views
5K
Looking to create a better frame for my e-motorcycle
  • · Replies 10 ·
Replies
10
Views
2K
Instantaneous centre of four bar linkage
  • · Replies 1 ·
Replies
1
Views
2K
Analytical Vector methods for a four bar linkage
  • · Replies 2 ·
Replies
2
Views
2K
Calculating the driving force in a double-acting cylinder using air
  • · Replies 4 ·
Replies
4
Views
5K
Is the Bar of Gold Real? Using Buoyancy to Determine Density
  • · Replies 15 ·
Replies
15
Views
2K
MATLAB Solving Freudensteins Equation for Four-Bar Linkage in Matlab
  • · Replies 2 ·
Replies
2
Views
8K
Find Mass Flow Rate and the Exit Flow Area
  • · Replies 3 ·
Replies
3
Views
6K
Center of Mass Calculation: x-bar, y-bar
  • · Replies 6 ·
Replies
6
Views
17K