Scribd
Upload
37 views

Control Systems Lab Report 1

This lab report analyzes a simple and inverted pendulum system using a PCS rig. For the simple pendulum, damping ratio and natural frequency were calculated from oscillation amplitude data. For the inverted pendulum cart system, the useful voltage range and effect of gain on hysteresis were determined. Hysteresis decreased exponentially with increasing gain due to improved compensator precision.

Uploaded by

Niall King
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
37 views

Control Systems Lab Report 1

This lab report analyzes a simple and inverted pendulum system using a PCS rig. For the simple pendulum, damping ratio and natural frequency were calculated from oscillation amplitude data. For the inverted pendulum cart system, the useful voltage range and effect of gain on hysteresis were determined. Hysteresis decreased exponentially with increasing gain due to improved compensator precision.

Uploaded by

Niall King
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 7

Lab Report 1: Analysing a Pendulum

1: Introduction
This report sets out to accomplish two primary objectives. The first of these is to perform an analysis
on two different pendulum systems; simple and inverted. As extension to the first objective, the
second objective was to deduce and model the characteristics and transfer functions of the different
systems.

Both systems could be observed using a PCS rig. For ease of recording and displaying results, each
system would be observed in a separate experiment due to the differing methods of obtaining
necessary variables for each system. In the case of the carriage servo additional variables, such as
the gain, were also of significance when evaluating results in addition to shared variables.

2: Equipment
- PCS rig
- PCS control console
- An Oscilloscope and Signal Generator
- Coaxial cables
- 30cm Rule

3: Methodology
3.1: Pendulum Modelling
When analysing the simple pendulum:
- Record effective pendulum length (required for deriving time period).
- Re-attach pendulum to rig.
- Connect control console to oscilloscope.
- Displace pendulum by 20 degrees and release.
- Observe and record oscilloscope readings for amplitude (mk).

3.2: Cart Modelling


When modelling the cart:
- Remove pendulum mass and place rig in inverted position.
- Set up control console in analogue control mode (See Figure 3.1).
- Set gain to maximum and velocity feedback to 75%, adjust to obtain optimum control from
potentiometer.
- Set dial “a” to approximately 2.7 and gain to approximately 1.2, set gain switch to “-ve”.
- Plug in compensators (See Figure 3.2) and switch on power.
- Vary value of gain from 0 – 3 and observe and record results.
- Remove pendulum rod and set potentiometer to 0, switch on power.
- Move potentiometer in increments of 2 volts, record output voltage at each increment.
- Repeat down to zero, to -10, then back to 0.
- Disconnect servo input and connect to GND.
- Push carriage to the right and release, calculate hysteresis.
-Repeat to the left, then repeat again at values of gain for 0%, 25%, 60% and 100%.
Figure 3.1.

Figure 3.2.

4: Results & Discussion


4.1: Pendulum Modelling
The magnitude of overshoots was recorded in Table 1, where K is the oscillation number. The values
of ln(mk) were also calculated and then plotted against values of K, as shown in Figure 4.1. This
graph showed very strong, negative correlation between values of ln(mk) and K. Due to this heavily
linear relation, and the relation of ln(mk) to the damping ratio, ζ, it can be concluded that ζ remains
constant throughout the period of oscillations, and therefore working out a value for ζ using an
average value of α can be deemed accurate to a reasonable degree.
ln(mk)/mV
0.12
0.1
0.08
ln(mk)/mV

0.06
ln(mk)/mV
0.04
Linear (ln(mk)/mV)
0.02
0
-0.02 0 2 4 6 8 10 12 14 16 18 20
-0.04
K

Figure 4.1.

Equation 1 was used to find a value of α for values of K up to 16, as shown in Table 2, with an
average value of α then being taken from these results, worked out as 0.01508. From this value, ζ
was worked out by consulting Equation 2, with ζ being solved as 2.4x10^-3. Natural frequency, ωn,
was calculated using Equation 3 and the value for T obtained experimentally, which gave a value for
ωn of 2pi. The transfer function (G(s)) was then modelled as shown in Figure 4.2.

Table 1.
Table 2.

Figure 4.2.

Using the oscilloscope readings, the time period of oscillations could also be estimated through the
use of cursors, though it should be noted that the oscilloscope did not provide a high degree of
precision due to the limitations of the scale, as cursors could not be placed accurately. To work out
the time period, the peak-peak value of T, in seconds, was recorded as per the readings shown in
Figure 4.3, worked out at T = 1s. Through deriving a formula of T according to the calculation in
Figure 4.4, a value of T was calculated as 0.954s.

This small discrepancy can be attributed both to the low precision of the oscilloscope and to the
potentially low accuracy of determining a value for L, as this was carried out using a 30cm rule which
had mm as the highest level of precision. As a result, a degree of uncertainty of +/-0.5mm may have
applied to L, as well as potential human error in the placement of the rule alongside the pendulum
rod.
Figure 4.3.

Figure 4.4.
4.2: Cart Modelling
When adjusting gain to balance the pendulum, balance was observed when the gain was set to 1.2.
Additionally, the ability of the carriage to compensate for the oscillation seemed to decrease when
the value was lowered, being lost entirely at around a value of 0.5. When gain was changed to values
above 1.2, the carriage then began to overcompensate. At a value of 2.5, the carriage would reach
the end of its track before resetting, allowing oscillations to resume normally. However, when the
value was further increased to 3.0, the carriage would not return to the central position and
oscillation of the pendulum would stop entirely.

The useful voltage range of the carriage servo was then obtained using the values displayed in Figure
4.5, worked out as a range from -8 – 6.5 volts, beyond which the output voltage no longer matches
input voltage. Hysteresis in the positive voltage range was not observed, likely due to the change in
direction occurring outside of the carriage’s useful voltage range. However, in the negative voltage
range, hysteresis was recorded as a value of 0.3 volts for the input and output values at -8 volts (-7.2
when working down from 0 and -7.5 when working back towards 0).

When measuring hysteresis through moving the cart by hand hysteresis was recorded as 0.25 volts.
This compares favourably with the result obtained in the previous part of the experiment,
considering the significant lack in accuracy resultant from pushing the cart by hand, which is only
compounded further by a large amount of noise observed on the oscilloscope reading preventing a
more precise scale from being used.

As displayed in Figure 4.6, hysteresis exhibits exponential decay as gain is increased. This is because
servo gain is intended to affect the precision of the compensator; increasing the gain brings the ratio
of controller input to output closer to 1. Resultantly hysteresis, a discrepancy between input and
output, decreases as the compensator’s precision increases.
10
9
8
7
6
5
4
3
2
1
0
-1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
-2
-3
-4
-5
-6
-7
-8
-9
-10

Input Output
Figure 4.5.

0.35

0.3

0.25

0.2

0.15

0.1

0.05

0
0% 25% 60% 100%

Figure 4.6.

5: References
Tigertek.com. 2014. Servo Motor Compensation - Adjusting Servo Motor Compensation. [online]
Available at: <http://www.tigertek.com/servo-motor-resources/servo-motor-compensation.html>
[Accessed 25 November 2021].

You might also like