Dynamic Three-Dimensional CFD Simulation of Closed Circuit Torque Converter Systems

Event
SAE 2016 World Congress and Exhibition
Authors Abstract
Content
This paper details the capability of PumpLinx® and Simerics® in simulating both Steady-State (Multiple Reference Frame) and transient, three dimensional torque converter performance and predicting the coupling point in a closed torque converter system in automatic transmission. The focuses of the simulation are in predicting the performance characteristics of the torque converters at different turbine to impeller rotating speeds (speed ratios) for 7 different torque converter designs and determine the coupling point at 70°C temperature. The computational domain includes the complex 3D design of all the impeller, turbine and reactor blades, the path ways that the oil travels between the above three components and the leakage gaps between these components. The physics captured in the simulation include the turbulence in the flow field and the rigorous treatment of the Fluid Structure Interaction (FSI) for the one-way free wheel reactor in predicting coupling point. The one-dimensional rotating dynamic modeling of the reactor enables the simulation of the whole range of speed ratios starting from 0 to 0.99. The integrated values of the transient torque on all the rotating components are found out to determine the torque ratio, K-Factor and efficiency. The comparisons with the hardware measurements show less than 5% differences between the test and simulation results. The consistency of the numeric schemes used for the simulation combined with the extremely fast run times and close comparisons with the test measurements adds value to the use of PumpLinx as a tool for simulating full torque converter systems.
Meta TagsDetails
DOI
https://doi.org/10.4271/2016-01-1345
Pages
12
Citation
Srinivasan, C., Joshi, D., Dhar, S., and Wang, D., "Dynamic Three-Dimensional CFD Simulation of Closed Circuit Torque Converter Systems," SAE Int. J. Passeng. Cars - Mech. Syst. 9(1):289-300, 2016, https://doi.org/10.4271/2016-01-1345.
Additional Details
Publisher
Published
Apr 5, 2016
Product Code
2016-01-1345
Content Type
Journal Article
Language
English