Vibration can be simply stated as a mechanical oscillation around an equilibrium point. In the first stage, vibrations can cause energy wastage and noise, but in later stages; they can even cause the machines to become unusable. This can lead to high maintenance costs, downtime, spare parts costs and even scrapping the equipment you spent tens of thousands of dollars. Please consider that; we are dealing with gearboxes, electric motors, or any machine element, their movements during operation generate vibration. This Vibration can be caused by the unbalance of rotating parts in motors, lubrication problems, gear cracks, and many other problems as well as the characteristics of the equipment itself. At this point, vibration is a parameter pointing out the problems. On the other hand, it is also a problem. Failures to detect defects can lead to catastrophic damages to the equipment. For this reason, vibration analyzes are performed in the design and quality control stages of products with moving parts. As a result, final products are produced. However, despite these design studies, many external factors such as improper assembly, insufficient lubrication, looseness in the field applications of the products cause vibration and interrupt system continuity especially in important places such as production lines, air conditioning, and energy systems. To avoid the destructive effects at this stage, vibration measurements and analyzes are made on the products within the scope of predictive/preventive maintenance activities. In this way, critical winnings are achieved in maintenance costs and there are no interruptions in production. In this way, savings are achieved by lowering the costs arising from production interruptions.
In the vibration analysis, the vibration data received from the sensors placed on the machine elements are processed and the failure occurrence is predicted. Vibration signals received from different points contain a complex waveform due to different forces and factors. Therefore, it is really difficult to determine the faults by examining the graphs in the time-wave form. The signals received in the time-wave form from the vibration sensors are created in the frequency band spectra using the Fourier Transform in the software. The frequencies in the spectrum indicate the type and source of the faults or defects, and the frequency amplitudes indicate the severity of the damage. Certain faults occur at specific frequencies and if these faults are not remedied, the amplitude of that frequency continues to increase.
Spectrum analysis is the most effective method used in the interpretation of the vibration signals. The most widely used is the Frequency-Amplitude graph. Frequency indicates how many times the vibration repeats per second. The unit of frequency is “1/second”, in other words, Hz, or “RPM”, that is, rev/minute units, especially for rotating equipment vibrations. Amplitude, on the other hand, refers to the power of vibration, which can be expressed in different units, it should be noted that these differences in different software or products will cause misinterpretation. Peak-to-Peak, Peak, RMS are commonly used amplitude display units.
Another mistake made while interpreting vibration data is to consider that vibration measurement has a unit of its own and ignore that it is a more likely to be an mathematical expression. Units measured in mechanical vibrations is in fact position, velocity or acceleration. Acceleration (mm/s2) at frequencies higher than 1000 Hz and velocity (m/s) in the range of 10 Hz ≤ f ≤ 1000 Hz are preferred. For vibrations smaller than 10 Hz, position(micron) is used. This situation cannot be called either true or false. The standards have been shaped in this direction due to the ease of reading/interpretation and measurement techniques. Since spectrum analysis is a field that requires expertise, in general, the monitoring of the vibration power can be done through the RMS value. Although it does not provide detailed information, the increase or decrease of the vibration density can be followed over this value. This value is the area under the Frequency-Amplitude (RMS) graph.