The design of journal bearings is important for the development of internal combustion engines. They experience load that varies both in magnitude and direction, the load being caused by the pressure forces and inertial forces of the crank–slider mechanism. It has been found that the most common causes of premature failure in internal combustion engine bearings are fatigue, sliding surface wear, erosion due to cavitation phenomena and all these effects are directly or indirectly related to the shape of the oil pressure field in shaft-bearing interface. An accurate estimation of oil pressure distribution characteristics and measurement of oil film thickness in the shaft-bearing interface would allow better prediction of the bearing performance and reliability. The goal of this work is to develop a method for calculating the oil film thickness (OFT), oil film pressure (OFP) in the shaft-bearing interface and its validation by experimental results. The detailed analysis of the lubrication condition around the position of minimum oil film thickness conditions is based on finite difference method of integration of the Reynolds equation coupled to the elastic deformations determined by a detailed finite element model. Four non contact eddy current gap sensors mounted on the main bearing of a single cylinder engine are used to measure the oil film thickness and the journal orbit for variable speed and variable loading conditions.

OFP distribution corresponding to these conditions was calculated by a numerical solution of the Reynolds equation. the instantaneous eccentricity was used as a starting point along with the operating and physical parameters.Approach To arrive at the OFT. A correlation was established between the two. This was fed back to the instantaneous eccentricity and an iterative technique leading to the OFT under elastohydrodynamic conditions was developed by matching the Load Carrying Capacity at every step. The journal orbit was plotted using Mobility method. The loading information thus obtained was applied to a FEM model of the bearing shell in Hypermesh environment coupled to Nastran solver. and compared with the experimentally measured orbit. to determine its elastic deformation. leading to a validation of the elastohydrodynamic lubrication model developed. . The benefit of this model is that is somewhat accurately predicts OFT and the corresponding OFP for the real case of an elastic bearing shell.

Flowchart Representing Analysis Methodology .

Flowchart Representing Operation of Matlab Code .

Experimental Bearing Model Results Experimental Bearing Model – OFT Iteration 2 (Load = 3.11 kN) .

Pressure Variation for the Elastic Case Bearing Load = 17 kN (Pressure in MPa) Pressure Distribution along length of bearing Pressure distribution along width of bearing .

Pressure Force. Inertia Force and Total Force 1250 rpm and 3 Nm Polar Plots of Bearing Loading 1250 rpm and 3 Nm .

Results – Oil Film Thickness vs CAD 1250 rpm and 3 Nm .

Journal Orbit 1250 rpm and 3 Nm .Results – Oil Film Thickness vs CAD 1250 rpm and 3 Nm Results .

Indicated Elastic Deformation = 9 μm Results .Displacement Indicated by Finite Element Model 1250 rpm and 3 Nm .

Results – Comparison of Journal Orbit determined by Mobility Method with Experimental Result 1250 rpm and 3 Nm Mobility Method Experimental Result .

produce 'noise' as the elements roll over the raceways and rub against the internal cage and flanges. Most bearing faults occur with the rolling elements. This noise is generated at high frequency and low amplitude.Rolling Bearing Vibration Detection ABSTRACT Rolling bearing vibration detection. The bearing housing amplifies the noise to a point where a sensitive accelerometer can 'hear' them. The frequency of the fault has a direct relationship with the geometry of the bearing and the relative speed of each individual raceway. 4. Readings start to indicate amplitude changes and the noise moves into the range of human hearing. 1. This is the time to record its frequency 'signature' and normal operating acceleration and velocity values. By measuring the velocity. The technology is well established and well proven but requires a good understanding of equipment vibration characteristics before a user can draw good interpretations from the results. . the noise signal can be amplified and represented as a frequency series in real-time. At this point the raceway shows visible signs of surface failure. In the case of a raceway failure these would be the four progressive stages. Keywords: out of balance. However under the surface of the raceway sub-surface defects have started. By careful detection and filtering. vibration analysis. Software libraries of noises for bearings in ideal condition are available for the vast number of bearings. By comparing the actual noise to the ideal bearing signature it is possible to highlight problems with the in-service bearing. acceleration and frequencies emitted from a rolling bearing it is possible to tell its condition and the likelihood of imminent failure. More vibration frequencies appear and more velocity side bands develop. the overall base level noise has risen and the velocity spectrum (graph) has risen higher. 3. If the bearing is still in service everyone knows it is time to change it out because they can hear it. Vibration analysis permits the monitoring of the bearing's progression through each stage and to estimate when failure will actually occur. even those in perfect condition. The extent of the failure increases and grows with more metal coming off in minute sheets (delaminating). bearing life Method of Bearing Vibration Monitoring All bearings. The Four Stages of Bearing Failure A roller bearing progresses through four stages to failure. The bearing is new and has no defects. Roller bearing vibration measurement is one of the major condition monitoring tools in regular use. Within the background noise particular frequencies start to standout (side bands) and indicate failure is fast approaching. The velocity spectrum is much higher and much more background noise has developed. The frequency signature has changed. cage or raceways. The method of presenting the data is either by visually representing it on an oscilloscope screen (monitor screen) as waves or as a number on a digital display. If examined at this stage there would be no visible defects. 2.