Objective
A numerical program has been developed which takes into account hydrodynamic behaviour of out-of-shape and misaligned bearings under dynamic loading. The program supplies misalignment values and can used previously determined misalignment values. Results are friction torque, oil film thickness, oil pressure distribution, shaft trajectories etc.
Preliminary friction measurements using single cylinder engines were compared with results from the industrial partners own programs. Discrepancies that could not be accounted for were discovered. The focus was placed on the 2 sub-assemblies where major mechanical friction losses occur during engine operation, namely the crankshaft bearing assembly and the piston ring liner assembly.
An extensive experimental program was completed with friction rigs being studied, as well as single and 4 cylinder engines. The results were analysed and the computed results compared.
A computer program was developed capable of taking into account the different aspects which govern lubrication problems. The piston ring liner investigations made possible a better understanding of the complex lubrication phenomena.
A numerical program has been developed to take into account the hydrodynamic behaviour of out of shape and/or misaligned bearings under dynamic loading. It supplies the misalignment values, and can also use previously determined misalignment values. Results are friction torque, oil film thickness, oil pressure distribution, shaft trajectories, etc. Agreement between computed and measured values is satisfying up to 3000 revolutions per minute. Computed friction torque values confirm the existence of the friction peak near top dead centre.The program is restricted to a 4 cylinder in line engine. It does not take into account bearing elastic deformations, gyroscopic and thermal effects. It requires measurements or a preliminary determination of the crank shaft deflections. The program provides indications on running conditions of the bearings in terms of wear and fatigue. An extensive analysis of the piston rings and piston skirt lubrication, including piston dynamics has been developed, taking into account ring lift in their grooves, oil control twist, 3-dimensional bore distortions, rings to bore conformity. It supplies friction power, friction forces, oil film thickness, oil flow, interring gas pressure, etc considering mixed friction, fully flooded and starved lubrication, cylindrical or barrelled skirt profile. In spite of an extensive experimental programme, a satisfying agreement could not be obtained between experiments and computations due to difficulties in ring friction and skirt friction measurements.
Main limitations of the piston lubrication program originate in the restricted possible ring profiles, in the lack of information about the extent of starvation, in an inadequate definition of the lubricated area of the skirt, in convergence problems and numerical instabilities for some specific input data.
The objectives of the crankshaft bearing investigations were to calculate crankshaft deformations during engine operation, taking into account bearing block stiffness and oil film thickness, in order to determine crankshaft bearing misalignment angles; to apply the hydrodynamic theory to the computation of oil film thickness, pressure distribution and friction forces taking into account out of shape and/or misaligned bearings by dynamic loading and finally to develop a complete program coupling the crankshaft deformation code and the misaligned shaft oil film behaviour.
Experiments were performed to measure the geometric data needed by the lubrication analysis, to compare experimental results and computed values and to determine the influence of various parameters (oil temperature diametral bearing clearance, etc).
The mathematical models developed resulted in a confirmation of the existence of a friction peak near combustion top dead centre (TDC) and an explanation of this peak due to an asymmetric pressure distribution. They also allowed possible predication of oil film thickness, pressure distribution and friction torque. At present the analysis is restricted to a 4 cylinder inline engine type and requires experimental measurements of the crankshaft deformations.
The objectives of the piston ring liner investigations were to develop computer programs for the calculation of the overall piston ring and piston skirt friction including mixed friction of the ring pack and hydrodynamic lubrication of the piston skirt. The investigations would requires to take account of inter ring gas pressure, liner temperature and mechanical and thermal distortions of the bore.
Experiments were performed in order to determine the geometrical values needed for the lubrication analysis, to assess the significance of various parameters (ring elastic pressure, speed load) and to compare computed and experimental values.
These investigation resulted in analyses of the piston ring pack lubrication and of the piston skirt dynamics and lubrication. The corresponding computing programs, configured so as to use one common input data file, are able to evaluate: inter ring gas pressure; friction power losses and friction forces in the ring pack, specifying each ring distribution; oil film thickness and oil flow through the ring pack; piston friction force; piston transverse velocity and finally piston eccentricity. The programs take into account lubricating conditions (fully flooded or starved), ring left and inertia effects, bore distortions as well as barrelled or cylindrical skirt profile.
Correlation problems still exist between experimental and computed values even though the present analyses contribute to a much better definition and understanding of the lubrication problems.
MAJOR SOURCES OF FRICTION LOSSES IN INTERNAL COMBUSTION ENGINES FROM PISTON-RING-LINER ASSEMBLY AND FROM CRANKSHAFT-BEARINGS. UP TO NOW, THE EXISTING MATHEMATICAL MODELS WHICH ARE USED TO EVALUATE THESE FRICTION LOSSES ARE LIMITED, BEING RESTRICTED TO PROBLEMS SUCH AS TO AVOID EARLY DAMAGE IN PROTOTYPE TESTING OR AS A HELP TO EVALUATION OF FAILURE DURING TESTING. IN NO CASE CAN THEY PREDICT THE BEHAVIOUR OF THE ENGINE DURING RUNNING.
TO DEVELOP PREDICTIVE MODELS WHICH COULD BE USED AT AN EARLY STAGE OF THE DESIGN IN ORDER TO IMPROVE ENGINE RUNNING, TO DECREASE FRICTION LOSSES AND TO REDUCE FUEL CONSUMPTION IS THE MAIN OBJECTIVE OF THIS PROJECT.
THE FOLLOWING STEPS ARE TO BE INVESTIGATED:
- TO CHECK THE VALIDITY OF THE MATHEMATICAL MODELS AVAILABLE IN THE PARTICIPATING COMPANIES USING EXPERIMENTAL DATA,
- ANALYSIS OF THE RESULTS IN ORDER TO ESTABLISH THE REASONS FOR THE DIVERGENCE BETWEEN CALCULATED VALUES OBTAINED FROM THE DIFFERENT CODES AND FOR THE VARIANCE BETWEEN COMPUTATION AND EXPERIMENT,
- TO DEVELOP IMPROVED OR NEW CODES,
- TO VALIDATE THESE CODES THROUGH A SERIES OF EXPERIMENTS.
Fields of science
Topic(s)
Data not availableCall for proposal
Data not availableFunding Scheme
CSC - Cost-sharing contractsCoordinator
75116 PARIS
France