Determination of the flow field inside a Sonolator liquid whistle using PIV and CFD. (18th May 2017)

• Record Type: Journal Article
• Authors: Ryan, David J.; Simmons, Mark J.H.; Baker, Michael R.
• Is Part Of: Chemical engineering science. Volume 163(2017)
• Page Start: 123
• Page End: 136
• Publication Date: 2017-05-18
• Subjects: Sonolator -- Liquid whistle -- Turbulence -- Emulsification -- Orifice; Chemical engineering -- Periodicals; Génie chimique -- Périodiques; Chemical engineering; Periodicals; Electronic journals; 660
• DOI: 10.1016/j.ces.2017.01.035
• Languages: English
• ISSNs: 0009-2509

Abstract

Highlights: Experimental and numerical evaluation of flow in a liquid whistle was made. CFD average velocities were in excellent agreement with PIV measurements. Turbulent flow parameters from CFD were within estimates from PIV data. The presence of the blade was less influential than the size of the orifice. Abstract: PIV experiments and CFD simulations were carried out on a mimic of a pilot scale Sonolator Model A inline liquid whistle mixer (Sonic Corp. USA) for water in turbulent flow for Reynolds numbers at the orifice between 17, 500 and 77, 200. Three different sizes of orifice were used. The results from PIV were compared with the CFD simulations, with both global and local validations being performed. The former focusses on the pressure drop across the Sonolator and the latter was carried out by comparison of local values of velocity magnitude, turbulent kinetic energy and local specific turbulent energy dissipation rate. Velocity magnitude values were found to agree within 10% more than fifteen millimetres downstream of the orifice. A similar level of agreement was found at the orifice for lower flow rates and larger orifices. Factors which precluded this level of agreement for higher flow rates and smaller orifices were the appearance of cavitation and a minimum achievable laser pulse separation, limiting the maximum velocity measureable by the PIV. Agreement between PIV and CFD was also poorer for the turbulent parameters, although the PIV and CFD data had similar trends, the magnitudes were different. The reasons for these discrepancies include the fact that the oscillation period for the orifice jet could not be precisely identified and eliminated from the data and errors inherent in the methods used to estimate the local specific turbulent energy dissipation rate from the PIV data.