Control Systems

Lab #3

Stability Analysis

Topics Covered:
• Stable, Marginally Stable and Unstable Systems
• Open-loop position and speed responses of a Servo System
1 Background
1.1 QUBE Servo Model
As depicted in Lab #1 & #2, the QUBE Servo System can be modeled as shown in
Figure 1-1.

𝑅𝑅𝑚𝑚 𝐿𝐿𝑚𝑚
+
𝑇𝑇 = 𝐾𝐾𝑇𝑇 𝐼𝐼𝑚𝑚
𝐼𝐼𝑚𝑚 𝜃𝜃𝑚𝑚 (𝑡𝑡), 𝜔𝜔𝑚𝑚 (𝑡𝑡)
+
𝑉𝑉𝑚𝑚 𝑉𝑉𝑏𝑏 = 𝐾𝐾𝑏𝑏 𝜔𝜔𝑚𝑚 (𝑡𝑡)

− 𝐽𝐽ℎ
𝐽𝐽𝑚𝑚
−
𝐽𝐽𝑑𝑑
Figure 1-1 QUBE Servo DC motor and load
Using the differential equations derived from this model of the system (ignoring
𝐿𝐿𝑚𝑚 because it has an insignificant effect on the response of the system), the
following open-loop transfer function is obtained for the angular velocity, Ω𝑚𝑚 (𝑠𝑠),
versus the motor voltage, 𝑉𝑉𝑚𝑚 (𝑠𝑠).
Ω𝑚𝑚 (𝑠𝑠) 1/𝐾𝐾𝑏𝑏
=
𝑉𝑉𝑚𝑚 (𝑠𝑠) 𝐽𝐽𝑒𝑒𝑒𝑒 𝑅𝑅𝑚𝑚
𝑠𝑠 � 𝐾𝐾 𝐾𝐾 � + 1
𝑡𝑡 𝑏𝑏

This transfer function is of the form:

𝐾𝐾
𝐺𝐺(𝑠𝑠) =
𝜏𝜏𝜏𝜏 + 1

𝐽𝐽𝑒𝑒𝑒𝑒 𝑅𝑅𝑚𝑚
Where 𝐾𝐾 = 1/𝐾𝐾𝑏𝑏 and 𝜏𝜏 = .
𝐾𝐾𝑡𝑡 𝐾𝐾𝑏𝑏

The parameters from the QUBE User’s Manual are:

𝐾𝐾𝑏𝑏 = 0.042 V/rad/sec
𝐾𝐾𝑡𝑡 = 0.042 N-m/A
𝐽𝐽𝑒𝑒𝑒𝑒 = 2.0886 × 10−5 Kg-m2

𝑅𝑅𝑚𝑚 = 8.4 Ω
Likewise the open-loop position transfer function is obtained from the
relationship:

2
 𝑑𝑑𝜃𝜃𝑚𝑚 (𝑡𝑡)
𝜔𝜔𝑚𝑚 (𝑡𝑡) =
𝑑𝑑𝑑𝑑
This means that the transfer function for the angular position, Θ𝑚𝑚 (𝑠𝑠), versus the
motor voltage, 𝑉𝑉𝑚𝑚 (𝑠𝑠) is:
Θ𝑚𝑚 (𝑠𝑠) 1/𝐾𝐾𝑏𝑏
=
𝑉𝑉𝑚𝑚 (𝑠𝑠) 𝐽𝐽𝑒𝑒𝑒𝑒 𝑅𝑅𝑚𝑚
𝑠𝑠 �𝑠𝑠 � 𝐾𝐾 𝐾𝐾 � + 1�
𝑡𝑡 𝑏𝑏

This is of the form:

𝐾𝐾
𝐺𝐺(𝑠𝑠) =
𝑠𝑠(𝜏𝜏𝜏𝜏 + 1)

1.2 Stability
A simple definition of stability is Bounded-Input Bounded-Output (BIBO)
stability. This means:
1. A system is considered stable if every bounded input yields a bounded
output.
2. A system is considered unstable if any bounded input yields a unbounded
output.
We can also talk about Asymptotic Stability. Asymptotic Stability means that any
system with initial conditions close to an equilibrium point of the system will
eventually converge back to the equilibrium point as 𝑡𝑡 → ∞.
A special case of asymptotic stability is exponential stability. This means for any
initial condition or bounded input the system will converge exponentially to a
finite value. A system is exponentially stable if and only if all of its poles are in
the left half of the s-plane.
Thus the stability of a system can be determined from its poles:
• Exponentially stable systems have poles only in the left-half of the s-plane.
• Unstable systems will have at least one pole in the right-half of the s-plane
and/or poles of multiplicity greater than 1 on the imaginary axis.
• Marginally stable systems have one or two complementary poles on the
imaginary axis and the other poles in the left-half of the s-plane.

2 Lab Exercises
Using the Simulink model created for the Integration and Filtering Lab, modify it to
create a model that applies a 1 V step to the motor and monitors the angular position
and angular velocity of the output of the motor. This is shown in Figure 2-1 below.

3
 Figure 2-1: Simulink Model to Measure Angular Position and Angular Velocity

Note that the differentiator and the low pass filter has been replaced by a transfer
function that does both functions. The “s” in the numerator is the Laplace transform
for the derivative of a function and the remainder of the transfer function is simple the
low pass filter with a cut-off frequency of 𝜔𝜔𝑐𝑐 = 150 rad/sec. You will be using this cut-
off frequency for this experiment.

2.1 Stability of the Angular Velocity

In this section you will determine the stability of the angular velocity
mathematically and experimentally.
1. Calculate the location of the pole(s) using the transfer function in Section
1.1. Determine based on the location of the pole(s) of the velocity transfer
function if the system is exponentially stable, unstable or marginally
stable. Include this analysis in your lab report.

2. Use the Monitor & Tune button , on the Hardware tab, to Run the
Simulink model in Figure 2-1. Note that the Stop time has been changed
from infinity, “inf”, to 2.5 seconds. Capture the motor voltage and speed
scope traces. Based on the speed response to the step input is the system
exponentially stable, unstable or marginally stable? Include pictures of
the scope traces and explain the rational for your answer based on the
experimental data in your lab report.

4
2.2 Stability of the Angular Position
In this section we examine the stability of the angular position of the disc
mounted on the motor hub both mathematically and experimentally.
1. Calculate the location of the pole(s) of the angular position transfer
function in Section 1.1. Determine based on the location of the pole(s) of
the position transfer function if the system is exponentially stable,
unstable or marginally stable. Include this analysis in your lab report.

2. Use the Monitor & Tune button , on the Hardware tab, to Run the
Simulink model in Figure 2-1. Again, the Stop time has been changed from
infinity, “inf”, to 2.5 seconds. Capture the motor voltage and position
scope traces. Based on the position response to the step input, is the
system exponentially stable, unstable or marginally stable? Include
pictures of the scope traces and explain the rational for your answer based
on the experimental data in your lab report.
3. Is there a signal where the open-loop position response will be stable? If
so, modify your Simulink model to apply that input to the system to verify
that the position response is BIBO stable. Include the diagram of your
Simulation model and scope traces with your analysis in your lab report.
Hint: Try either a short pulse or a sine wave input. (A pulse can be
constructed by adding a positive step to a delayed negative step).

This completes the lab exercises. Remember to save your Simulink model and
any screen captures to your USB memory stick for use in completing your lab
report.