~$ Sandip Gautam
BACK TO PROJECTS
$ cat ./projects/bachelor_thesis.md |

Fabrication of Particle Image Velocimetry Setup at Low Reynolds Number

Mechanical DesignCATIA

In this project, we designed and fabricated a Particle Image Velocimetry (PIV) setup to study low Reynolds number flow over a flat plate (chord length 76.33 mm) in the range of Re = 2,000–10,000. The system utilized a Class IIIB laser for particle illumination, a Chronos 2.1 high-speed camera, and a smartphone camera for PIV and flow visualization. Velocity fields were obtained through cross-correlation of sequential images using PIVlab (MATLAB). The experimental data were analyzed to identify key flow features such as laminar separation bubbles, reattachment points, turbulent mixing, and trailing-edge vortex shedding. Results were further validated through CFD simulations conducted under similar conditions, and potential sources of experimental and processing errors were systematically evaluated.

Fabrication of Particle Image Velocimetry Setup at Low Reynolds Number

Engineering Design and Structural Validation

The facility consists of a 2 m glass tank, a rigid structural frame, and a gantry-driven flat plate actuation mechanism. Blockage ratio constraints were imposed to maintain aerodynamic fidelity below 5%, ensuring minimal wall interference on flow development. Structural integrity of the tank under hydrostatic loading was verified using finite element analysis, leading to a 10 mm glass thickness selection with a safety factor of approximately 15. The actuation system was powered by a bipolar stepper motor controlled via ESP8266 and L298N driver circuitry, enabling precise velocity control and repeatable Reynolds number targeting. This ensured dynamic similarity and controlled transient flow initiation.

Velocity Triangle at Rotor Midspan

Fig: FEA stress & deformation contour of tank.

Velocity Triangle at Rotor Midspan

Fig: CAD assembly of frame and gantry

Optical System and Flow Seeding Strategy

Planar laser illumination was generated using a Class IIIB 5 mW laser to produce a thin light sheet intersecting the measurement plane. Flow tracing was achieved using fine glass powder particles (less than 190 µm), selected after evaluating relaxation time and sedimentation velocity to ensure adequate flow-following capability under low-speed conditions. High-speed image acquisition was performed using a Chronos 2.1 camera (100 fps), while additional frame extraction was performed using a secondary imaging system for lower frame rate validation.

Image Processing and Velocity Reconstruction

Velocity fields were reconstructed using PIVlab (MATLAB-based FFT cross-correlation). Pre-processing included Contrast Limited Adaptive Histogram Equalization (CLAHE), high-pass filtering, and intensity normalization to enhance particle visibility and minimize noise contamination. Vector validation and spurious vector removal were applied to ensure reliable cross-correlation peak detection. The processing pipeline enabled accurate recovery of instantaneous velocity vectors and vorticity contours, even under low particle density conditions.

Velocity Triangle at Rotor Midspan

Fig: Image Pre-processing Flowchart

Experimental Investigation and Flow Physics

Two test configurations were investigated: a NACA 0012 airfoil (Re ≈ 3,270) and a flat plate (Re ≈ 6,040). The velocity vector fields revealed leading-edge vortex formation and laminar separation bubble behavior consistent with low Reynolds number aerodynamic theory. Vorticity contours showed coherent vortex shedding patterns downstream of the flat plate, indicating successful resolution of unsteady flow structures. Spanwise velocity variations and near-wall flow distortion highlighted three-dimensional effects and the limitations of quasi-two-dimensional assumptions. Experimental results were compared against ANSYS Fluent simulations to assess qualitative agreement in separation location and velocity distribution trends.

Velocity Triangle at Rotor Midspan

Fig: Velocity (m/s) field around aifoil and vorticity contour at α = 12 degree, airfoil moving from left to right (mask applied at shadowed region)

Velocity Triangle at Rotor Midspan

Fig: Velocity (m/s) vector field and vorticity contour at α = 12deg, flat plate moving from left to right (mask applied at shadowed region)

Velocity Triangle at Rotor Midspan

Fig: Vorticity contours around the flat plate, flat plate moving from left to right (mask applied at shadowed region)

Velocity Triangle at Rotor Midspan

Fig: Correlation coefficient contour plot, flat plate moving from left to right (mask applied at shadowed region)

Uncertainty and Error Analysis

Measurement uncertainty was assessed based on particle size, interrogation window size, image resolution, and cross-correlation sensitivity. Primary error sources included illumination non-uniformity, particle sedimentation, optical distortion, and low signal-to-noise ratio at high displacement gradients. Despite these limitations, velocity trends remained consistent with theoretical and numerical predictions.

Technical Significance

This project demonstrates the ability to:

  • Design and fabricate experimental fluid mechanics infrastructure
  • Perform structural validation via FEA
  • Implement optical flow measurement systems
  • Conduct low Reynolds number aerodynamic experiments
  • Apply digital image processing and FFT-based cross-correlation
  • Validate experimental results against CFD simulations
  • The fabricated system now serves as a reusable research platform for future investigations in unsteady aerodynamics and vortex-dominated flows.