Cairo University Engineering Measurement MDP3610

Faculty of Engineering Year 2022 – 2023

Mechanical Design and Production Department 1st Semester

Dynamic Signal Analysis Lab

Objectives
− Measure the vibration of a mechanical system.
− Use accelerometer to measure the vibration.
− Analyze the frequency content of the measured vibration signal.

Dynamic Signals
Dynamic signals are the physical quantities of a system which are varying with time. They are
measured using sensors and fed to a computer through a data acquisition system for purposes
of data analysis or control.
Sensors are the measuring devices which convert the physical quantities (temperature, pressure,
displacement, force, strain, …) into a measurable analog voltage which could be detected.
Data acquisition is the collection of hardware and software responsible to sample and process
the measured analog signals into digitized data saved and processed by computers.
Data acquisition converts the real world continuous analog signal into digitized data through
the analog to digital A/D converter and vice versa using digital to analog converter D/A.

Signal Discretization
The original measured signal is discretized due to the sampling effect. The sampling frequency
should be suitable to the original signal frequency to avoid loss of signal data.
If the sampling frequency is too low, this introduces a distortion in the discrete reconstructed
data called aliasing.

1/6
To avoid aliasing, the sampling frequency must be at least twice the maximum frequency that
needs to be monitored. If '𝑓𝑛𝑦𝑞𝑢𝑖𝑠𝑡 ' is the maximum frequency component to be detected in the
measured signal. The sampling time '𝑇𝑠 ' must be led to at least a sampling frequency twice the
maximum frequency to be detected.
1
𝑇𝑠 <
2𝑓𝑛𝑦𝑞𝑢𝑖𝑠𝑡
1
𝑓𝑛𝑦𝑞𝑢𝑖𝑠𝑡 =
2𝑇𝑠
𝑓𝑛𝑦𝑞𝑢𝑖𝑠𝑡 is called the Nyquist frequency.

The signal must be sampled at a rate at least twice the highest frequency in the signal. In fact,
experience indicates that 2.5 samples per cycle is a better choice.
𝑠𝑎𝑚𝑝𝑙𝑖𝑛𝑔 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 ≥ 2.56 𝑚𝑎𝑥𝑖𝑚𝑢𝑚 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦

Time Domain and Frequency Domain

A harmonic signal could be transferred form time domain into frequency using the Fourier
Transform. Fast Fourier Transform is the form implemented to a discretized signal to find its
frequency and amplitude.

Periodic Signal is composed of multi harmonic signal, each having its amplitude and phase.
x(t) = A1 sin(ω1t+1) + A2 sin(ω1t+3) + A3 sin(ω3t+3) + ...

The Fast Fourier Transform breaks down the signal to its harmonic signals such that
𝑁−1
1 2𝜋𝑘𝑟
𝑋𝑘 = ∑ 𝑥𝑟 𝑒 −𝑖( 𝑁 )
𝑁
𝑟=0

2/6
In case of discrete measurement sampling x1, x2, …, xN at given time values t1, t2, …, tN
respectively, where N denotes an even number of equidistant points over one period time T =
NTs.

∞ ∞
𝑎0 2𝜋𝑘𝑡𝑖 2𝜋𝑘𝑡𝑖
𝑥(𝑡𝑖 ) = + ∑ 𝑎𝑖 cos + ∑ 𝑏𝑖 sin
𝑁 𝑇 𝑇
𝑖=1 𝑖=1
𝑁
2
𝑎0 = ∑ 𝑥𝑖
𝑁
𝑖=1
𝑁
2 2𝜋𝑘𝑡𝑖
𝑎𝑖 = ∑ 𝑥𝑖 cos 𝑘 = 1, 2, …
𝑁 𝑇
𝑖=1
𝑁
2 2𝜋𝑘𝑡𝑖
𝑏𝑖 = ∑ 𝑥𝑖 sin 𝑘 = 1, 2, …
𝑁 𝑇
𝑖=1

Frequency vector (Hz)

1 1 2 1
𝑓𝑖 = [1, 2, … , 𝑁] = [ , ,…, ]
𝑁 𝑇𝑠 𝑁 𝑇𝑠 𝑁 𝑇𝑠 𝑇𝑠
Sampling frequency (Hz)
1
𝑓𝑠𝑎𝑚𝑝𝑙𝑖𝑛𝑔 =
𝑇𝑠
Frequency resolution (Hz)
1
∆𝑓 =
𝑁 𝑇𝑠
Nyquist frequency (Hz)
𝑓𝑠𝑎𝑚𝑝𝑙𝑖𝑛𝑔
𝑓𝑁𝑦𝑞𝑢𝑖𝑠𝑡 =
2

3/6
Example:
For a measurement of 20 ms sampling time and 256 samples, calculate the FFT maximum
frequency which could be measured, frequency resolution and frequency vector.

𝑇𝑠 = 20 × 10−3 𝑠 𝑁 = 256
1 1
𝑓𝑠𝑎𝑚𝑝𝑙𝑖𝑛𝑔 = = = 50 𝐻𝑧
𝑇𝑠 20 × 10−3
Maximum frequency to be measured
𝑓𝑠𝑎𝑚𝑝𝑙𝑖𝑛𝑔 50
𝑓𝑁𝑦𝑞𝑢𝑖𝑠𝑡 = = = 25 𝐻𝑧
2 2
Frequency resolution
1 1
∆𝑓 = = = 0.1953 𝐻𝑧
𝑁 𝑇𝑠 256 × 20 × 10−3
Frequency vector (Hz)
1
𝑓𝑖 = [1, 2, … , 𝑁] = [0.1953, 0.3906, … ,25, … , 50]
𝑁 𝑇𝑠

Structure Setup
A vibratory mechanical is to be selected by each students group consisting of mass and spring
(mass-spring, vibrating table, beam element, …).
The mechanical system shall be subject to different vibration excitations, under free and forced
vibrations excitations.
In the free vibration, the subject is to be excited by initial excitation. The free vibration is
measured and captured. Then, the measured vibration shall be analyzed and determined.
The vibration is measured due to external force excited due to motor unbalance. The measured
signal frequency content could be measured and analyzed.

Vibration Measurement
The system vibration is measured using a mechatronic accelerometer sensor (ADXL 345). It is
connected to the computer using Arduino controller. The measured vibration is captured, saved
and analyzed using a LabView code.

4/6
LabView Code

5/6
Experiment Procedure
1- Put the ADXL 345 accelerometer sensor on the structure you want to measure the vibration.
2- Connect the accelerometer to the Arduino Board.
3- Run the Arduino and LabView codes to run the measurements.
4- Record and analyze the vibration.
5- Submit a group technical report for the experiment. [objective, experiment setup picture
and explanation, results plots, data analysis, discussion]

6/6