Application of a Supersonic Fluidic Oscillator to Improve Performance of the Superplastic Forming Process

The Superplastic Forming (SPF) process involves gas injection from a variable pressure supply into a die, to form a heated metallic sheet over the surface of the die into a complex automotive body panel shape. The existing process involves excessive forming times to allow relaxation of residual stresses and avoid part cracking and tearing. Previous research shows that multi-pulse pressure forming has a favorable effect on overall material formability and properties hence, reducing the required manufacturing time.

A Supersonic Fluidic Oscillator (SFO), due to an absence of moving parts, is capable of reliably generating the required pressure fluctuations under the extremely high temperatures present in the SPF process.

Our research collaboration with AEM began over five years ago with a Natural Sciences and Engineering (NSERC) Engage grant. In this work a supersonic fluidic "load" oscillator, previously investigated for a different application, was studied using a commercially available finite-volume computational package (ANSYS Fluent). The predicted oscillation frequency compared reasonably well with certain, but not all, of the previously published results. Determination of the reasons for discrepancies lead to our additional studies. Another concern is the excessive computational time required to achieve the numerical results for just one set of design parameters.

Two subsequent studies which were supported by Ontario Centres of Excellence (OCE) VIP I grants. The first involved the design, construction and instrumentation of an experimental test facility to obtain reliable data to evaluate the accuracy of the computational fluid dynamic model of the SFO and make recommendations for improvements to the model. The second involved investigating methods of overcoming the extremely large computational times. This research led to certain refinements of the computational model which improved the accuracy however, the parallel implementation did not reduce the computational time to an acceptable level.

The observed ability of the SFO to provide suitable oscillations led to an investigation of the performance of the device while pressurizing a chamber under conditions closer to those found in the actual SPF process. A two-year OCE VIP II research grant supported this research which included three separate but related projects. The first resulted in a MASc thesis in which a high-pressure (approximately 4 MPa) cold model was designed, constructed and tested. The second also resulted in a second MASc thesis which a refined CFD simulation of the overall process which was compared with the experiments. The third project involved the development of a low-order mathematical model of the SFO application for use in the practical industrial design of the device for a specific implementation. This project was initiated by a PhD student who is currently being supported with a MITACS Accelerate Internship.