Date of Award


Document Type

Open Access Dissertation

Degree Name

Doctor of Philosophy in Mechanical Engineering-Engineering Mechanics (PhD)

Administrative Home Department

Department of Mechanical Engineering-Engineering Mechanics

Advisor 1

Darrell Robinette

Committee Member 1

Jason Blough

Committee Member 2

Carl Anderson

Committee Member 3

Gowtham S.


A torque converter was instrumented with 29 pressure transducers. The pressure transducers were located in multiple cavities. The instrumented cavities included, four transducers mounted on the impeller shell, on the channel between blades. Six transducers mounted on the pressure and suction sides on the middle streamline of a turbine blade. Another seven transducers mounted on the pressure and suction sides of the core, middle and shell streamlines of a stator blade. Seven transducers mounted on the torque converter clutch cavity. Finally, five on the cavity between the pressure plate and the turbine shell. The torque converter was part of a 6 speed front wheel drive transmission and differential, also instrumented with various pressure transducers, thermocouples and a flow meter. The transmission measurements were not in scope for the present work with the exception of the thermocouples, flow meter and torque converter clutch pressure, which approximated torque converter inlet pressure during early stages of the project.

A transmission lab was designed and built as part of the investigation. Acquisition of the torque converter pressure data was accomplished with a custom designed and built telemetry system developed for the present study by IRT Telemetrics located in Hancock Michigan.

A computational fluids dynamics model was developed using a commercially available software. The computer model was used to correlate with the torque converter measured torques and pressures. The computer model was optimized accuracy of predicted torques and for accelerated solution time. Solution times were reduced from 9 hours to under 40 minutes per speed ratio while the accuracy of torques error varied by up to 6% between tests and simulation. Accuracy of pressure simulated values varied widely depending on the cavity under study. The torque converter inlet flow worked best with 5% turbulence intensity while other cavities such as the toroidal ones were best modeled with a turbulence intensity set to 50%.

The computer model was able to predict pressure trends during the many tests completed as part of the investigation. Flow recirculation was seen on the turbine and stator blade passages on the low speed ratios. The recirculation region affected simulated and measured pressures on both sides of the turbine and stator blades as seen in previous investigations.

Further studies should be carried out using the model developed as part of this work as a starting point. Further improvements in accuracy and solution time are highly valued by the industry to help reduce costs associated with computer time and development costs associated with inaccuracies.