Fluid Dynamic Study of the FDA Benchmark Model under Pulsatile Flow

Restricted (Penn State Only)
Kessler, Joshua C
Graduate Program:
Master of Science
Document Type:
Master Thesis
Date of Defense:
June 28, 2017
Committee Members:
  • Keefe Manning, Thesis Advisor
  • William Hancock, Committee Member
  • Computational Fluid Dynamics
  • CFD
  • Laser Doppler Velocimetry
  • LDV
  • FDA Benchmark Nozzle Model
  • FDA Critical Path Initiative
  • Hemolysis
  • Sudden Expansion
  • Conical Diffuser
  • Turbulence
Pulsatile flow through a simplified benchmark model was analyzed as a part of the FDA’s Critical Path Initiative, which intends to evaluate the accuracy of computational fluid dynamics (CFD) in the study of medical devices. The benchmark nozzle model mimics the flow characteristics of common medical devices. In one orientation, the model consists of a rapid contraction followed by a narrowed throat region and a conical diffuser. In the other orientation, the model comprises a gradual contraction followed by a narrow throat and a sudden expansion, from which flow exits as a jet with significant areas of recirculation. The in vitro data collected using the benchmark model provides a reliable method for validating CFD simulations. Of particular significance is the validation of Reynolds shear stresses, which are used to predict blood damage. Laser Doppler velocimetry (LDV) was used to collect two-component velocity data within the benchmark model under pulsatile flow at locations with pertinent flow characteristics. The measured velocities were used to calculate principal Reynolds stresses, maximum Reynolds shear stresses, turbulence intensities, viscous shear stresses, and wall shear stresses during mid-acceleration, systole, mid-deceleration, and diastole. Axial velocities during systole were compared with previous data taken using LDV. Shear stresses near the wall of the nozzle throat, at the boundary of the free shear layer within the conical diffuser and downstream of the sudden expansion, and within the recirculation regions during diastole exceeded accepted thresholds for the onset of hemolysis, indicating that the model geometry may induce blood damage under certain flow conditions. Combined with the results of previous studies of the benchmark model, the results presented in this study indicate that the model provides a useful comparison for computational simulations that seek to predict hemolysis in medical devices.