Regarding vibration detection of equipment
Thread Content
We commonly use vibration meters to monitor the vibration levels of equipment, such as those of pumps or motors. For centrifugal pumps or motors, what vibration level indicates that there is a problem with the equipment (such as damaged bearings)? Additionally, vibration values can be expressed in terms of acceleration, velocity, and displacement – which one is generally measured? Is it more appropriate to use that value to determine the fault? If the test results show that the displacement meets the requirements but the velocity does not, what is the reason for this? This post was last edited by fanli*n on 2009-2-23 12:22 ]
This issue has been discussed many times on forums, and in every single forum it’s been brought up more than once. The vibration standards for equipment should first be determined based on the specific conditions of that equipment. There are international standards, **industry standards, corporate standards, and manufacturer-specific standards, among others. If you have no idea where to start, you can first download the ISO 2372 or 10816 standards for use. As for which parameter to use, since it’s usually a handheld instrument, speed is a more appropriate choice; most standards are based on speed as well. If the standard involves displacement, then displacement should be used. Standards only tell you the general condition of the equipment; they cannot reveal the specific cause of the fault, which requires further vibration analysis, including an examination of the spectrum and similar factors.
Generally, what we measure is speed; I usually pay attention to how it changes, as the values vary from person to person. We generally focus on whether there is a trend of increase :)
The standards used per unit vary, and the focus of the measurements also differs; displacement is used more frequently.
Yes, we are looking at the displacement, that is, the amplitude. Vibration is generally classified into 4 levels: A: 0.71 μm, B: 1.8 μm, C: 4.5 μm, D: 11.2 μm. This applies to pump motors with a power of 15 kW or less; a level of D indicates failure to meet the standards. Last edited by lflhome1 on 2009-2-23 at 20:25.]
What kind of equipment is upstairs? The vibration level is too low. Looking at your values, they seem to represent speed; converting the units to mm/s should give a more accurate figure. The issue for the person asking the question is that they are starting measurements and need a reference value. In fact, the more international standards are used, the greater the differences may be from each factory’s specific conditions, as those standards are designed for general use, while each factory has its own unique requirements. Therefore, international standards can be referred to, and the equipment in the factory should be tested; the vibration values when the equipment is operating normally can be used as a reference, with 4 to 6 times these normal values serving as the alarm threshold. Make adjustments according to the specific circumstances. This post was last edited by mirook on 2009-2-24 09:49]
Theoretically, all types of displacement sensors can provide acceleration, velocity, and displacement outputs. An inertial accelerometer can have its acceleration value integrated twice to obtain displacement; if signal analysis is sufficient, the acceleration signal can be used directly, with the phase being 180 degrees different from that of displacement. The same is true in reverse. From a performance perspective, the choice mainly depends on the frequency and amplitude range you need. The most commonly used types are the acceleration (inertia) type and the eddy current type. The differences I can recall are as follows: 1. The acceleration type measures absolute vibration (absolute coordinates, including vibration of the base), while the eddy current type measures relative vibration (the surface of the shaft relative to the bracket). 2. Prices are generally on the higher side due to inertia. However, the installation location is quite flexible. Eddy current type is inexpensive; domestic ones cost a little over 1000, and a bracket is needed. 3. When using eddy current measurement, surface surface waves need to be deducted.
Measure velocity or acceleration signals. It is determined based on the effective value of the average speed, with national standards as a reference ； Acceleration is determined by its peak value
Typical handheld instruments are equipped with acceleration sensors, which can be used to calculate velocity and displacement; velocity sensors can also be used to calculate displacement. In all cases, integration is involved – there are no sensors that use differentiation of displacement to determine velocity, nor are there sensors that use differentiation of velocity to determine acceleration. The owner mentioned that the displacement was above the limit while the speed was not; this was likely measured using a handheld device. It is recommended to rely on the speed value instead. Since the displacement is relatively large at low frequencies, and handheld instruments generally provide inaccurate readings for signals below 10 Hz, the displacement data obtained after integration often contains significant low-frequency errors. Moreover, the magnitude of the test values is also related to the location where the measurement is taken. Since handheld instruments require the operator to select the measurement point, it is necessary to take measurements on the bearing cover, and the measurement points should be marked clearly. If it is not a magnetically mounted sensor, the force and direction with which the device is pressed also play a role. The person responsible for testing should receive some basic training and familiarize themselves with the relevant standards; it is best if the same person carries out the measurements.
Implementing differentiation in hardware is indeed a bit challenging; generally, post-processing can be used to take the derivative of the displacement signal at a certain characteristic frequency with respect to time, thereby obtaining the amplitude of the velocity and acceleration at that frequency. In essence, it’s just a matter of multiplying by a coefficient corresponding to that frequency. Its main function is to standardize different vibration results, for example, by expressing vibrations in terms of velocity.
I have only seen integrals; I have never seen derivatives.
Differentiation can be achieved using hardware circuits, but the interference it may cause is a major issue, and it’s hard to say anything about the high-frequency response of such circuits. It’s not easy to solve anyway. Using software methods, the velocity and displacement can be obtained by direct digital integration of the acceleration sensor signal. Conversely, real-time digital differentiation of the displacement sensor can also be achieved. Speaking of which, the algorithm is somewhat simpler in terms of differentiation, as integration involves dealing with an arbitrary integral constant. But maybe the derivative is more sensitive to interference. So it’s best to filter out the high-frequency components before differentiating. What we do isn’t differentiation either; it’s simply that after collecting the data, we convert the acceleration signals of the two or three main frequencies, while ignoring all the others. It’s relatively simple – just take the derivative and sum the sine signals. This post was last edited by RainWolf on 2009-3-6 23:24 ]
The experts are among the people. Rainwolf and Mirook are learned from control instruments, right!