Control Systems

Lab #7

Proportional and Derivative

Dr. Gary Lowe

Lawrence Technological University

Topics Covered:
• Servo Position Control
• Proportional and Derivative Controller
• Designing a controller to meet specifications

Required:
• Complete Lab#4 - Empirical Modeling
• Complete Lab#5 - First Principles Modeling
• Complete Lab#6 - Second Order System Response
1 Background
1.1 Servo Model
In the First Principles Modeling lab the Qube Servo voltage to position
transfer function was found to be:
Θ(s) K1
G(s) = = (1.1)
Vm (s) s(τ s + 1)
The gain, K1 , and time constant,τ , were found to be: K1 ≈ 23.8 rad/V-sec
and τ ≈ 0.1 seconds. You will use this model to design a proportional and
derivative controller for the Qube Servo system that meets some specific
criteria.

1.2 PD Control
Proportional and derivative, or PD, control is a subset of Proportional,
Integral and Derivative, or PID, control. PID control is usually expressed as
a function of the error signal in a feedback control system. The input to the
system to be controlled is u(t). The error signal is the difference between
the commanded input r(t) and the output from the system y(t). This is
expressed as e(t) = r(t) − y(t). The PID controller is given by:
Z t
de(t)
u(t) = kp e(t) + ki e(τ )dτ + kd (1.2)
0 dt
In block diagram format, the feedback control system is shown in Figure 1.1
The portion of the block diagram enclosed by the dashed block is the

Figure 1.1: Block Diagram of PID Control System

implementation of the PID controller. For this lab, you will not be using

2
the integral term only the proportional and derivative terms.
Consequently,the controller, Gc (s) is represented by:
U (s)
Gc (s) = = kd (s + kp /kd ) = kd (s + a) (1.3)
E(s)
Where a = kp /kd . The PD controller just adds a zero and some extra gain
to the open-loop system, G(s). This results in a closed-loop system transfer
function of:
Y (s) Gc (s)G(s) K1 kd (s + a)/τ
T (s) = = = 2 (1.4)
R(s) 1 + Gc (s)G(s) s + (K1 kd /τ + 1/τ )s + K1 kd a/τ

2 Pre-Lab Work

This section must be completed before starting the lab. You will have
to do this portion of the lab on your own time. No time is allocated for
this during the lab time.
2.1. Design a PD controller using the Root Locus technique with the
system transfer function G(s) modeled in Equation 1.1 using
K1 = 23.8 and τ = 0.1. The closed-loop system must meet the
following specifications:
- The peak overshoot must be less than or equal to 8%
- The closed-loop system has a maximum settling time of 0.2
seconds to a step input.
2.2. Simulate the closed-loop system (controller and the system
transfer function G(s) using a Simulink model. Do not use a single
closed-loop transfer function. You may use a transfer function
block for Gc (s) and another for G(s). Your Simulink model should
look similar to the block diagram in Figure 1.1.
2.3. Ensure that the simulated closed-loop system meets the desired
requirements. If it doesn’t meet the requirements, redesign the
controller until it does meet the requirements.
2.4. Include a screenshot of the Simulink model and the input and
output plots in your lab report.

3
3 Lab Exercises

Construct the Simulink model shown in Figure 3.1. This model

Figure 3.1: Simulink Model for the PD Control System

3.1. Set kp = 2.5 and kd = 0 initially. Build, connect and run the model
with the Qube Servo system. Capture the position and motor
voltage plots for use in your lab report. What do you notice about
the response?
3.2. Vary kp between 1 and 4. Note: you do not have to rebuild the
model for this. Capture position and motor voltage plots at the
extreme gain values for use in your lab report. What do you notice
about the relationship between the output position response and
the magnitude of kp ?
3.3. Set kp = 2.5 and vary the derivative gain, kd , between 0 and 0.2.
Capture position and motor voltage plots at the extreme values of

4
 the gain for use in your lab report. What is the effect on the
output position response when increasing the derivative gain, kd ?
3.4. Use the values for kp and kd that you designed for the PD
controller in the Pre-Lab work Item 2.3 for the proportional and
derivative gains in your Simulink Model. Run the model and
capture the position and motor voltage plots for use in your lab
report.
3.5. Do these values satisfy the requirements given in Item 2.1? If
these values do not satisfy the requirements, adjust kp and/or kd
until the requirements are satisfied. How do the computed values
for the gains compare to the actual values for the gains?
3.6. Run the model you used to design the controller in Item 2.3 in
parallel with the actual system as shown in Figure 3.2 . Do the
position traces of the modeled system and the actual system
match? If they do not, why do you think there is a discrepancy?

Figure 3.2: Closed-loop Model and System Running in Parallel