ME 4501 Vibrations & Controls Laboratory

PID Control of Rigid Body Motion

Due dates: 1 week

The experiment will be conducted as a team. Then each individual will do the work on the
sheet, and submit your work in D2L (scanned or documented).

SAFETY

Be sure to stay clear of the mechanism before when implementing the algorithm or
Executing the Trajectory on the ECP computer control software. If there is an instability
or large control signal, the motor may react violently. If the system appears stable after
implementing the controller, first displace it with a light, non-sharp object (e.g. a plastic
ruler or pen) to verify stability prior to touching the sliding carriage.

Objectives

1. Understand how a PID closed-loop system controls the position of a rigid body.

Equipment

▪ ECP Model 210 Rectilinear apparatus arranged with no springs, and carriage
connected to the motor via the rank-and-pinion. Run PID control with the ECP control
software and hardware. Feedback sensor is the optical sensor for the carriage 1.
Changes made in electronics, i.e., PID parameters.

▪ Rigid Body PD & PID Control Background and key Equations

This experiment demonstrates some key concepts associated with proportional plus
derivative (PD) control and subsequently the effects of adding integral action (PID).
This control scheme, acting on plants modeled as rigid bodies finds broader application
in industry than any other. It is employed in such diverse areas as machine tools,
automobiles (cruise control), and spacecraft (attitude and gimbal control). The block
diagram for forward path PID control of a rigid body is shown in Figure 1 where friction
is neglected.1

1The student may want to later verify that for the relatively high amount of control damping in the scheme
that follows – induced via the parameter kd – that the plant damping is very small.
RJR & DK Page 1 of 10
 ME 4501 Vibrations & Controls Laboratory

Figure 1. Rigid Body PID Control – Control Block Diagram for Feedback Control

The closed loop transfer functions is:

𝑘ℎ𝑤
𝑋(𝑠) (𝑚 ) (𝑘𝑑 𝑠 2 + 𝑘𝑝 𝑠 + 𝑘𝑖 )
𝑇(𝑠) = =
𝑅(𝑠) 𝑏 + 𝑘ℎ𝑤 𝑘𝑑 2 𝑘ℎ𝑤 𝑘𝑝 𝑘 𝑘𝑖
𝑠3 + 𝑠 + 𝑚 𝑠 + ℎ𝑤
𝑚 𝑚

We will assume khw = 12800 N/V (newtons per volt). For the first portion of this
exercise we shall consider PD control only (ki = 0). For the case of kd in the return
path the transfer function reduces to:

𝑘ℎ𝑤 𝑘𝑑 𝑘ℎ𝑤 𝑘𝑝
𝑋(𝑠) 𝑠 +
𝑇(𝑠) = = 𝑚 𝑚
(1)
𝑅(𝑠) 𝑏 + 𝑘 𝑘
ℎ𝑤 𝑑 𝑘ℎ𝑤 𝑘𝑝
𝑠2 + 𝑠 +
𝑚 𝑚

By defining:

𝑘ℎ𝑤 𝑘𝑝
𝜔𝑛 = √ (2)
𝑚

𝑏 + 𝑘ℎ𝑤 𝑘𝑑 𝑏 + 𝑘ℎ𝑤 𝑘𝑑 𝑘ℎ𝑤 𝑘𝑑

𝜁= = ≈ (3)
2𝑚𝜔𝑛 2√𝑚𝑘ℎ𝑤 𝑘𝑝 2√𝑚𝑘ℎ𝑤 𝑘𝑝

we may express:
𝜔𝑛2
𝑇(𝑠) = (4)
𝑠 2 + 2𝜁𝜔𝑛 + 𝜔𝑛2

The effect of kp and kd on the roots of the denominator (damped second order
oscillator) is studied in the work that follows.

RJR & DK Page 2 of 10

 ME 4501 Vibrations & Controls Laboratory

Procedure

Proportional & Derivative Control Actions

1. Modify the Model 210 with 3 thick plate (assume each plate weighs 500g) + 1 thin
plate (250g mass) on the first mass carriage with no springs or damper attached.
You may neglect friction. Include the mass of the empty carriage (0.59kg) and the
mass of the rack and pinion (0.34 kg).

2. Set-up the plant in the configuration described in Step 1. There should be no springs
or damper connected to the first carriage and the other carriages should be secured
away from the range of motion of the first carriage.

3. From Eq (3) determine the value of kp (kd=0) so that the system behaves like a 1.5 Hz
spring-mass oscillator (need to convert Hz to rad/s). Each individual, then each group
should calculate this (assumed that 𝜔𝑛 ≈ 𝜔𝑑 )

Hardware Natural frequency

Mass (m) Gain kp
Gain (khw) fn wn
Exp 1 12800
Unit kg N/V Hz rad/s

4. Set-up to collect Encoder #1 and Commanded Position information via the Set-up Data
Acquisition box in the Data menu. Set up a closed-loop step of 0 (zero) counts, dwell
time = 3000 ms, and 1 (one) rep (via Trajectory in the Command menu).

5. Enter the Control Algorithm box under Set-up notice that the digital sampling rate
(Ts=0.00442 s), and select Continuous Time control OR Discrete Time (one or the
other works!). Select PID and Set-up Algorithm. Enter the kp value determined above
for fn = 1.5 Hz oscillation (kd & ki = 0, do not input values greater than kp = 0.08) and
select OK.
Place the first mass carriage at approximately the -0.5 cm (negative is toward the
motor) mark. Or set/leave at 0 cm is OK.

RJR & DK Page 3 of 10

 ME 4501 Vibrations & Controls Laboratory

In this and all future work, be sure to stay clear of the mechanism before doing the
next step. Selecting Implement Algorithm immediately implements the specified
controller; if there is an instability or large control signal2, the plant may react
violently. If the system appears stable after implementing the controller, first displace
it with a light, non sharp object (e.g. a plastic ruler or pen) to verify stability prior to
touching plant.

Select Implement Algorithm, then OK.

6. Make sure encoder #1 is at or very close to 0. If not you can “zero position” under
Utility menu. Then Enter Execute under Command. Prepare to manually displace the
mass carriage roughly 2.5 cm. Select Run, displace the mass approximately 2.5 cm
and release it. Do not hold the mass position for longer than about 1 second as this
may cause the motor drive thermal protection to open the control loop so that motor
does not overheat.

Plot encoder #1 output using the MATLAB. FYI: 1mm == 227 counts. Determine the
frequency of oscillation from the time period from the chart. (You may want to
consider several cycles, get the time at start and end, as you did in previous labs.)
Compare this (using % difference) with the predicted frequency in step. Given the
uncertainty in your measurements, is your % difference acceptable?

Natural frequency (fn)

Gain kp
Predicted Experiment Difference
Exp 1 1.5
Unit V/m Hz Hz %

RJR & DK Page 4 of 10

 ME 4501 Vibrations & Controls Laboratory

7. What will happen when proportional gain, kp, is doubled? kp =

8. Repeat Steps 5 - 7 and verify your prediction. (Again, for system stability, do not
input values greater than kp = 0.06). Calculate the new value for the natural frequency
(wn or fn). Plot encoder #1 output with kp doubled. Then calculate the % difference
between the natural frequency predicted and the natural frequency calculated above.

Natural frequency (fn)

Gain kp
Predicted Experiment Difference
Exp 2
Unit V/m Hz Hz %

RJR & DK Page 5 of 10

 ME 4501 Vibrations & Controls Laboratory

PD Control Design
9. From design controllers (i.e. find kp & kd) for a system natural frequencies fn = 4 Hz,
and three damping cases:
1)  = 0.2 (under-damped),
2)  = 1.0 (critically damped)
3)  = 2.0 (over-damped).

These results are used in the PD controller implemented in step Reponses in the
following steps. Show your work in space below this table.

 kp kd
Exp 3 0.2
Exp 4 1
Exp 5 2

RJR & DK Page 6 of 10

 ME 4501 Vibrations & Controls Laboratory

Step Response
10. Implement the underdamped controller (via PID) and set up a trajectory for a
2270 count (approx. 1 centimeter) closed-loop Step with 1000 ms dwell time and 1
rep.

11. Execute this trajectory and plot the commanded position and encoder position. (Plot
them both on the same vertical axis so that there is no graphical bias.)

12. Plot the output using MATLAB (Need to include the plot here). Does this make sense
for Underdamped system step response and why? Now calculate the % overshoot
based on the peak output, steady-state value, and rough estimate of the Settling Time
(seconds) based on the graph. (Include the data points in the plot used using the date-
tip)
Overshoot Steady-State Settling time
 kp kd
(%) value (s)
0.2

RJR & DK Page 7 of 10

 ME 4501 Vibrations & Controls Laboratory

13. Repeat Steps 10 - 12 for the critically damped case. Plot the output using the
MATLAB. Does this make sense for Critically Damped system step response and
why?
Overshoot Steady-State Settling time
 kp kd
(%) value (s)
1.0

RJR & DK Page 8 of 10

 ME 4501 Vibrations & Controls Laboratory

14. Repeat Steps 13 & 14 for the over-damped case. Plot the output using the MATLAB.
Does this make sense for Over-Damped system step response and why?
Overshoot Steady-State Settling time
 kp kd
(%) value (s)
2.0

RJR & DK Page 9 of 10

 ME 4501 Vibrations & Controls Laboratory

Questions to answer (in Discussion)

1. Why does the P-control system behave like a mass-spring-damper system, even
though no spring exists?

2. How does the kd variable effect the system? Does it improve the response? (i.e.,
does it add to “damping”? does it affect the settling time? Does it affect
overshoot?)

RJR & DK Page 10 of 10