Vibration analysis is a diagnostic process used to detect, monitor, and prevent mechanical failures in machinery. Engineers perform vibration analysis to examine the vibration signal patterns in a system and find anomalies or changes. They spot any irregularities that may signal upcoming problems, especially for heavy-duty machinery like motors, compressors, and gears [1].
By examining these vibrations, one can learn more about the condition of the machine or structure and decide what maintenance is necessary. Stress, displacement, velocity, eigenmodes, and more can all be calculated and visualized using vibration analysis tools like SimScale, a cloud-native solution that offers seamless workflow integration, handling multiple physics simulations directly in your browser, particularly Finite Element Analysis (FEA), the numerical solution for analyzing vibration.
Thanks to vibration analysis, the lifespan of the machine or structure can be increased. And according to rules set forth by international standardization bodies, a lot of products are required to pass physical vibration testing before they can be approved for safe use. Thus, vibration analysis using SimScale FEA helps reduce the risk of structural failure during the actual vibration testing by providing accurate virtual testing and minimizing the need for physical vibration testing.
Fundamentals of Vibration
Parameters of Vibration
Vibration may be caused by various factors, including but not restricted to recurring forces, imbalance, and misaligned machine parts. Displacement, velocity, and acceleration are the three primary vibration measurement variables. These are measured in terms of their magnitudes (or amplitudes), where displacement magnitude is usually measured in millimeters or micrometers, velocity magnitude in millimeters per second, and acceleration magnitude in meters per second squared (or in g’s).
Types of Vibration
There are primarily two categories of vibrations:
Natural vibration: A system vibrates at a natural frequency when it experiences natural vibration, also referred to as resonance. If the operating frequency and the machine’s natural frequency are the same, this type of vibration may be problematic since it may cause excessive vibrations [1]. Using SimScale’s frequency analysis, users can examine the natural frequencies of their structures to assess the overall rigidity of their design structure as well as the local rigidity in specific places.
Forced vibration: This happens when a system vibrates because of an external force. Examples are the vibrations that a running motor or pump causes [1]. These external forces or imposed motion excitations can take the form of harmonic, periodic, non-periodic, or random motion excitations and can provide energy for vibration [2].
There are many different sources of excitation force for equipment, like roll imbalance, where the center of mass is not aligned with the axis of rotation. Nonlinear processes like vibration suppression and deterministic chaos have also been studied [3]. A versatile instrument, vibration analysis is essential for quality assurance, maintenance, and diagnostics.
The Method of Vibration Analysis
Steps for Vibration Analysis
Physical vibration analysis involves a systematic process to monitor and identify faults in machinery. The steps include [1]:
Preliminary Data Collection: This involves collecting details about the equipment, such as its operating conditions and the necessary sensors for measurements.
Vibration Measurement: Different tools like accelerometers, velocity sensors, and displacement sensors are utilized to measure the vibrations.
Data Analysis: The collected data is processed using algorithms and software to find vibration patterns and trends, which are then compared to industry standards.
Results Interpretation: Analysts interpret the findings to identify particular vibration frequencies associated with mechanical problems. The fundamental frequency, harmonics, and sidebands are significant frequency types that can detect flaws, such as misalignment or bearing problems.
Design Change or Maintenance: In the design stage, if anomalies are found, a change in design would follow the interpreted results. In the operation stage, anomaly detection leads to alarm thresholds being set. These can be absolute, trending, or statistical thresholds. When vibrations surpass these thresholds, timely action is mandated, which might include maintenance or further investigation.
In engineering design, while physical testing is vital, it can be quite costly and time-consuming, especially when multiple tests for different variables are required. This is where engineering simulation plays a significant role.
With SimScale’s cloud-native simulation, engineers can run multiple simulations in parallel, setting up varying real-world scenarios of vibration. This enables them to minimize the testing time significantly while maintaining high-quality data analysis using FEA solvers. As a result, the steps for vibration analysis in a simulation environment would take the following shape:
CAD preparation and upload: Import/upload the CAD file of the component into SimScale and prepare it for simulation.
Simulation setup: Select the FEA simulation that enables you to analyze the vibrations on the component (frequency analysis or harmonic analysis), and follow the steps in the simulation tree on your workbench to set up your settings, geometries, boundary conditions, meshes, and others.
Post-processing: Analyze the data and extract the relevant information plots, tables, 2D and 3D visualizations.
Informed Design Decision: Following the resulting information from your simulations, you can make more informed decisions on how the design needs to be configured in order to maintain the part’s structural integrity against vibration.
Types of Vibration Analysis
A variety of techniques can be utilized to analyze vibration data. These can be defined as the following:
Time-Domain Analysis: This technique evaluates raw vibration signals from waveforms. Key data points like peak amplitude and RMS are extracted. It identifies transient events, tracks vibration levels, and sets operational limits. Exceeding these limits suggests machine wear or defects [4].
Frequency-Domain Analysis: This technique typically utilizes the Fast Fourier Transform (FFT) to convert time-domain signals into frequency-domain. This reveals specific frequencies tied to mechanical faults and is adept at spotting abnormal vibration patterns. An example is a crack that has developed on a roller bearing outer race that leads to periodic collisions with bearing rollers, which may be masked in time waveforms [4].
Envelope Analysis: Also termed demodulation, its primary use is in the early detection of bearing defects by isolating high-frequency impact signals from overall vibration.
Vibration Modal Analysis: An advanced method that pinpoints a machine’s natural frequencies, mode shapes, and damping characteristics, aiding in understanding the machine’s dynamic behavior and potential structural or resonance issues. Depending on the specific aims and needs of the analysis, vibration modal analysis is an example of an FEA analysis type that can be performed in both the time domain and the frequency domain [5]. Within SimScale, modal analysis simulation offers a holistic view of machinery conditions, highlighting resonance-related issues. It outlines a system’s response limits to loads, and using this, design changes can be proposed to ensure system stability.
Vibration Analysis in SimScale
SimScale is a powerful vibration analysis tool for engineers as it offers structural mechanics simulation capabilities reinforced with cloud computing. This allows for cloud-native simulation capabilities that enable:
Running parallel simulations at the same time, which reduces analysis time significantly,
Accessing FEA simulations anytime, anywhere, simply in a web browser,
Saving costs by avoiding manual upgrades, installation, maintenance, and any associated fees.
One of the many capabilities of SimScale is to analyze vibration patterns and find irregularities in mechanical systems. SimScale’s FEA modal (frequency) analysis, powered by the Code Aster solver, enables the computation of natural frequencies of a structure and the corresponding oscillation mode shapes. It can help identify the eigenfrequencies (eigenvalues) and eigenmodes (mode shapes) of a structure under vibration. Furthermore, SimScale’s adaptability isn’t restricted to simple shapes; it also includes support for a variety of CAD file types and validation in typical engineering settings, such as the random vibration response of cantilever beams. For more information about why we need vibration analysis and why SimScale is the appropriate tool for that, check out our application page on Vibration Analysis Simulations.
Example of Vibration Analysis with SimScale
SimScale enables vibration analysis across various industries and application areas, including automotive, aerospace, consumer products, and machinery and industrial equipment. One example from the automotive industry is the vibration analysis simulation of an electric motor bracket using SimScale. Here, it is important to examine the bracket’s structural characteristics in order to confirm that its critical response frequencies fall within the designated operational range.
This simulation study’s design objective is to enhance the electric motor support bracket such that its natural frequencies stay outside of the motors’ running speeds, preventing potential part damage, bolt loosening, reduction in clearances between parts, and unwelcome noise. Through this examination, engineers can make sure that the electric motor bracket is stable and efficient for the duration of its operational life.
Figure 3 below shows the simulation workflow from CAD geometry to post-processed simulation results. Upon simulation, numerous statistical data would be available to further examine the bracket’s eigenmode number, eigenfrequency, Modal Effective Mass (MEM), Normalized Modal Effective Mass, and cumulative Normalized Modal Effective Mass (CNME). Furthermore, multiple CAD variants of the bracket can be visualized and simulated in SimScale simultaneously to find the optimal geometry that enables the furthest eigenfrequency value from the speed of the rotating shaft, thus minimizing the risk of possible failure due to vibration.
Vibration can be measured, analyzed, and controlled with the appropriate set of tools. Vibration analysis simulation can now be used as a predictive tool in your engineering toolbox for optimized designs, better structural integrity, and more efficient operation thanks to structural analysis tools in SimScale.
SimScale provides a user-friendly interface and a cloud-based architecture in addition to its extensive toolkit, enabling real-time collaboration and fast upgrades. It’s more than simply a platform; it’s an ecosystem that regularly develops and updates to match the needs of a technological environment that is rapidly changing.
Set up your own cloud-native simulation via the web in minutes by creating an account on the SimScale platform. No installation, special hardware, or credit card is required.
C. China, “What is vibration analysis and how can it help optimize predictive maintenance?,” IBM 2023. Available: IBM, https://www.ibm.com/blog/vibration-analysis/
G. Chen, “Handbook of Friction-Vibration Interactions”, Woodhead Publishing, 2014, Pages 9-70, https://doi.org/10.1533/9780857094599.9
J.J. Thomsen, “Vibrations and Stability. Advanced Theory, Analysis, and Tools”, Springer, 2021, ISBN: 978-3-030-68044-2, https://doi.org/10.1007/978-3-030-68045-9
TWI-Global, “What is Vibration Analysis and what is it used for?”, TWI-Global, 2023. Available: TWI-Global, https://www.twi-global.com/technical-knowledge/faqs/vibration-analysis
T.-H. Le, Y. Tamura, A. Yoshida, N. Dong, “Frequency Domain versus Time Domain Modal Identifications for Ambient Excited Structures”, International Conference on Engineering Mechanics and Automation (ICEMA 2010), Hanoi, July 1-2, 2010. Available: ResearchGate, https://www.researchgate.net/publication/285397109_Frequency_Domain_versus_Time_Domain_Modal_Identifications_for_Ambient_Excited_Structures