Search for: All records

We present the development and laboratory evaluation of RPiPIV, an underwater particle image velocimetry (PIV) system controlled by a Raspberry Pi. Designed specifically to measure bubble characteristics and bubble‐induced flow in natural hydrocarbon seeps, RPiPIV comprises three primary pressure enclosures, housing a consumer‐grade laser for particle illumination, a Gig‐E camera for image capture, a Raspberry Pi for system control, and essential supporting electronics for voltage conversion, battery management, and remote connection. Operating on 24–36 V DC power, the RPiPIV system can be deployed tethered onto a remotely operated vehicle or self‐contained for extended duration measurements. Comparing the RPiPIV and a laboratory high‐speed camera system, we conducted assessments of bubble imaging in a bubble stream and PIV measurements in a water jet, bubble‐chain flow, and single‐orifice bubble plume. Laboratory assessments revealed that bubble diameter estimates differed by approximately 5%. In PIV measurements, mean axial velocities exhibited differences of approximately 5%, while turbulent normal and shear stresses showed variances within 10–30%. Dissipation rates of turbulence kinetic energy differed by approximately 60%. These findings underscore the system's potential for reliably quantifying complex multiphase flow characteristics in deep‐sea environments.

We conducted a spectral analysis of the turbulence kinetic energy (TKE) budget in a bubble plume using particle image velocimetry with fluorescent particles. Our findings confirmed the hypothesis of an inverse energy cascade in the bubble plume, where TKE is transferred from small to large eddies. This is attributed to direct injection of TKE by bubble passages across a wide range of scales, in contrast to canonical shear production of TKE in large scales. Turbulence dissipation was identified as the primary sink of the bubble-produced TKE and occurred at all scales. The decomposition of velocities using the critical length scale of inter-scale energy transfer allowed us to distinguish between large- and small-scale motions in the bubble plume. The large-scale turbulent fluctuations exhibited a skewed distribution and were likely associated with the return flow after bubble passage and the velocities induced by the bubble wake. The small-scale turbulent fluctuations followed a Gaussian distribution relatively well. The large-scale motions contributed to over half of the Reynolds stresses, while there were significant small-scale contributions to the normal stresses near the plume center but not to the shear stress. The large-scale motions in the vorticity field induced a street of vertically elongated vortex pairs, while the small-scale vortices exhibited similar sizes in both horizontal and vertical directions.