2DOF Gear System, addition of torque causing directional issues - what is Simulink doing?

Hello,

I have a two degree of freedom gear system and have made a Simulink model to test the dynamics of it (Figure 1).

I have included both the torsional stiffness (kmb) and damping between the gears (qmb), and tested the dynamic response which seems accurate (pinion and gear rotate in opposite directions if the pinion is given a small initial displacement - Figures 3 and 4).

However, when I add an input torque to excite the system, the direction of the gears seem to go the same way (Figure 5), and I'm not sure why this is.

My Simulink model and equations of the system are shown below.

(Figure 1 - Diagram of System)

(Figure 2a - Equations)

(Figure 2b - Simulink System)

(Figure 3 - Angular Displacement vs. Time (No input torque just theta_p_0 = 1e-3))

(Figure 4 - Angular Velocity vs. Time (No input torque just theta_p_0 = 1e-3))

(Figure 5 - Angular Velocity vs. Time for step input of Torque)

Any pointers would be appreciated!

Answer 1

[Gillian Rosen]

Hi Tyler, 

What happens when you try different torque inputs? For example, different magnitudes of step, different functions (ramp, impulse, etc.)? This could help provide insight into what is happening in your system. 

Another option for you could be to try reducing your system to a simpler system (for example, simplifying or removing the spring-damper portion of the system) and testing the reduced system with your step input. This could help isolate the part of your system that is not behaving as expected.