Immediate downtime due to failure of the FAG bearing model itself is rare, for example due to incorrect installation or lack of lubrication. Depending on the operating conditions, it can take a few minutes, and in some cases months, for a bearing to begin to fail until it actually fails. When selecting the type of bearing monitoring, the gradual deterioration of the condition should be based on the application of the bearing and the failure consequences of the bearing when it is running on the equipment. . 1.1 Subjective identification of failure In most bearing applications, if the operator finds that the bearing system is not running smoothly or has abnormal noise, it can be judged that the bearing is damaged, see Table 1. Bearing monitoring with technical equipment Precise and long-term monitoring of bearing operation is required when bearing failures can lead to dangerous events or lead to long-term downtime. Take, for example, the turbine of an engine and a paper machine. For monitoring to be reliable, it must be selected based on the expected failure type. Damage spread over large areas Sufficient and clean lubricant is the main prerequisite for trouble-free operation. Undesirable changes can be detected by: Monitoring the lubricant supply Oil sight glass Measuring oil pressure Measuring oil flow Detection of abrasive particles in the lubricant Continuous on-line detection of the number of particles flowing through temperature measurement generally with thermocouples 41 Abnormal operation means failure 1: Failure detected by the operator Damage to the ferrule or rolling elements Vibration saw: more shocks and vibrations at the connecting rod, reduced working accuracy, worn rings or rolling elements due to contamination or insufficient lubrication, clearance or preload change, lathe: further development of vibration lines on the workpiece Periodic surface defects, such as tensile deformation, segregation streamlines, etc. Unusual running noise: Rumble or irregular noise Horse neighing or rattling noise Gradual change in running noise Surface (for example due to contamination or fatigue) motor gear (because the noise of the gear is always submerged, so the noise of the bearing is difficult to identify) 2: The temperature change of the spindle bearing of the FAG machine tool. Test conditions: n dm u003d 750 000 min1 mm. 3: The temperature change of the disturbed floating bearing. Test conditions: n dm u003d 750 000 min1 mm. Bearing failures due to insufficient lubrication can be detected reliably and relatively simply by measuring the temperature. Normal temperature characteristics: A stable temperature can be reached during smooth operation, see Figure 2. Abnormal characteristics: A sudden increase in temperature may be caused by lack of lubrication or radial or axial over-preload of the bearing, see Figure 3. Uneven temperature changes and a continuous upward trend in temperature are usually due to a deterioration of the lubrication condition, such as the end of the grease life, see Figure 4. However, it is not appropriate to use temperature measurement to determine initial damage, such as fatigue, there. 4: The relationship between temperature change and time when the grease fails. Test conditions: n dm u003d 200 000 min1 mm. Local damage to the bearing, such as dents, static corrosion or fractures caused by the rolling elements, can be detected in time by vibration measurements. Vibration waves caused by pits under cyclic motion are recorded by path, velocity and acceleration sensors. These signals can be further processed in different ways, depending on the operating conditions and the desired confidence level. The most common are: Measurement of RMS value Measurement of vibration value Signal analysis by envelope detection Experience has shown that the latter is more reliable and applicable. With a special signal processing, even damaged bearing parts can be found, see Figures 5 and 6. For more information please refer to our TI No. WL 80-63 'Diagnosing Rolling Bearings with the FAG Bearing Analyzer'.