Explicit Coupled Multi-Physics Piston Cooling Simulation

Explicit Coupled Multi-Physics Piston Cooling Simulation Serdar Güryuva Technical Specialist CAE/CFD Powertrain Engineering - Ford OTOSAN Page 1 of 29

Outline Introduction Piston Cooling Piston Cooling Modelling Problem Definition Modelling Details & Methodology Results and Verification Further Opportunities Q&A Page 2 of 29

Introduction Piston Cooling Diesel Engines Specific Power Rating(SPR) increases by high pressure fuel injection and higher boosted turbo-charging. Increased SPR is limited thermal durability of Head, Liner(Bore), Exhaust Manifold & Piston due to increased thermal and mechanical loads. Engine Emission also depends on contradicting thermal requirements. HC gets reduced with higher Temperature while NOx gets higher. For Pistons, Increased Temperature would result in degradation of material properties particularly of aluminium ones and deterioration of lubricant quality. Lubricant coking Excessive Wear or Scuff Thermal Crack on Piston Carbon and debris deposition Ring sticking and micro welding It is necessary to accurately guess the piston temperatures to understand and consider the effect on both durability, lubrication and emissions. Page 3 of 29

Problem Definition Piston Cooling Lubricating oil serves as a secondary cooling medium for pistons to limit the temperature. Single Jet Multi Jet Oil is directed Under the Piston (Crown) To the Piston gallery Both. Piston Cooling Gallery Is closer to critical heat sources, Has higher heat transfer area, Has longer heat transfer duration. Cocktail Shaker Effect(CSE) enhances heat transfer inside the gallery and reduces the temperatures up to 300 C. Unconstrained movement of oil should be modelled to see CSE on Piston Temperature. REF:[1] Page 4 of 29

Piston Telemetry Tests Active Engine Dynamometer for accurate torque and speed control. Oil Temperature & Coolant Temperature are controlled by external heat exchangers. Negative Temperature Coefficient Thermistor (NTC) with inductive data transfer. NTC resistance declines with increased temperature NTC s are cemented into drill holes, intimate contact. Sensor Calibration is necessary for signaltemperature relation. REF:[2] Page 5 of 29

Front Piston Sensor Locations Rear Anti-Trust Trust Cyl1 Center Cyl2 Top Land Trust Cyl4 Bowl Rim Front Cyl4 Bowl Rim Rear Cyl1 Bowl Rim Anti-Trust Cyl4 Bowl Rim Trust Cyl1 Center Cyl1 Bowl Rim Trust Cyl1 Bowl Base Front Cyl3 Pin Boss Rear Cyl3 Top Ring Groove Trust Cyl3 Skirt Page 8 of 29

Piston Cooling Modelling General Information No Cylinder to Cylinder Variation No Phase Change, No Oil Vapour Steady State Working Conditions Constant Heat Transfer from Rings to Bore Surface Constant Heat Transfer From Piston Side Surfaces and Pin to Oil No Connecting Rod Motion No Crank-Shaft Oil Splash Constant Temperature Oil Properties Solid-Fluid Explicit Coupling by Co-Simulation Preferably. Last Cycle Mean Convective Heat Transfer From Fluid to Solid Data Exchange at Each 2 Cycles. Final Temperature at Surface From Solid to Fluid Data Exchange at Each 100s. Page 9 of 29

Piston Cooling with Moving Overset Piston motion is set by using velocity input vs CAD. Total of 4 Cycles ran. (Starting from better temperature distribution.) Page 10 of 29

Piston Cooling with Moving Overset Piston motion is set by using velocity input vs CAD. Total of 4 Cycles ran. (Starting from better temperature distribution.) Convection BC surface data is exported based on reference coordinate system Page 11 of 29

Piston Cooling without Motion Hard to Set-Up, Easy to Run. Relative motion of piston injection target w.r.t CAD is included by adding relative velocity of target point motion. Discrete Multiphase Analysis with VOF Mesh Requirements: Size of Oil Jet Inlet Time Step Requirements (Adaptive) CLF <0.5 at iso-surface Total Max Cell Count (Overset): 1.0M Total CPUhr req. For One Cycle: ~200CPUhr Page 15 of 29

Piston Cooling without Motion Piston Acceleration is used to consider the effects of moving piston. Piston motion is set by using acceleration input vs CAD. Total of 16 Cycles ran. (Starting from uniform initial temperature distribution.) Surface temperature is mapped using tabular data mapper for consistency. Final Temperature is obtained after 16th cycle, but this was due to trial of different thermal BC settings to be able to match the experimental data. Page 20 of 29

Comparison of Two Cases Fluid Flow Moving Piston with Overset Motionless Piston 2nd Cycle 2nd Cycle Page 21 of 29

Comparison of Two Cases Fluid Flow Moving Piston with Overset Motionless Piston 3rd Cycle 3rd Cycle Page 22 of 29

Comparison of Two Cases Fluid Flow Moving Piston with Overset Motionless Piston Page 23 of 29

Comparison of Two Cases Fluid Flow Moving Piston with Overset Motionless Piston Page 24 of 29

Comparison of Two Cases Fluid Flow Moving Piston with Overset Motionless Piston Page 25 of 29

Comparison of Two Cases Fluid Flow Moving Piston with Overset Motionless Piston Page 26 of 29

Test Results & Correlation Piston temperatures are adequately matching with telemetry results. RMS error for temperature is in ±15 Piston wall temperatures would effect the convective thermal loads and heat transfer from rings to block, so itis necessary to run CHT analysis for complete head and block system. Page 27 of 29

Q&A Thanks to: Anıl Boz, Can Çerkezoğlu, Cenk Gören, Erhan Doğruyol Page 28 of 29

References 1. David C. Luff, Theo Law, Paul J. Shayler, Ian Pegg, The Effect of Piston Cooling Jets on Diesel Engine Piston Temperatures, Emissions and Fuel Consumption, SAE-2012-01-1212 2. Norman Thiel, Hans-Joachim Weimar, Hartmut Kamp, Herbert Windisch, Advanced Piston Cooling Efficiency: A Comparison of Different New Gallery Cooling Concepts, SAE-2007-01-1441 Page 29 of 29