0

### Research Papers: Flows in Complex Systems

J. Fluids Eng. 2018;140(6):061101-061101-9. doi:10.1115/1.4038761.

The paper describes the modeling and the experimental tests of a variable displacement vane pump for engine lubrication. The approach used for the simulation has involved three-dimensional (3D) commercial tools for tuning a zero-dimensional (0D) customized model implemented in the LMS Amesim® environment. Different leakage paths are considered and the axial clearances are variable to take into account the deformation of the pump cover, calculated through a finite element analysis with ANSYS. The vane tip clearances are calculated as function of the dynamic equilibrium equation of the vanes. The displacement control takes into account the internal forces on the stator due to the pressure in all variable chambers and to the contact force exerted by the vanes. The discharge coefficients in the resistive components have been tuned by means of a complete 3D transient model of the pump built with PumpLinx®. The tuned 0D model has been proved to be reliable for the determination of the steady-state flow-speed and flow-pressure curves, with a correct estimation of the internal leakages and of the pressure imposed by the displacement control. The pump has been also tested using a simplified circuit, and a fair agreement has been found in the evaluation of the delivery pressure ripple.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2018;140(6):061102-061102-7. doi:10.1115/1.4038986.

The pressure drop across $90deg$ sharp-angled miter elbows connecting straight circular pipes is studied in a bespoke experimental facility by using water and air as working fluids flowing in the range of bulk Reynolds number $500. To the best of our knowledge, the dependence on the Reynolds number of the pressure drop across the miter elbow scaled by the dynamic pressure, i.e., the pressure-loss coefficient $K$, is reported herein for the first time. The coefficient is shown to decrease sharply with the Reynolds number up to about $Re=20,000$ and, at higher Reynolds numbers, to approach mildly a constant $K=0.9$, which is about 20% lower than the currently reported value in the literature. We quantify this relation and the dependence between $K$ and the straight-pipe friction factor at the same Reynolds number through two new empirical correlations, which will be useful for the design of piping systems fitted with these sharp elbows. The pressure drop is also expressed in terms of the scaled equivalent length, i.e., the length of a straight pipe that would produce the same pressure drop as the elbow at the same Reynolds number.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2018;140(6):061103-061103-10. doi:10.1115/1.4039089.

Flexible electricity demand and variability of the electricity produced by wind turbines and photovoltaic affect the stable operations of power grids. Pump-turbines are used to stabilize the power grid by maintaining a real-time electricity demand. Consistently, the machines experience transient conditions during the course of operation, such as start-up, load acceptance, load rejection, and shutdown, which induce high amplitude pressure pulsations and affect operating lifespan of the components. During the closure of the wicket gates, the transient flow characteristics is analyzed for a Francis-type reversible pump-turbine in generating mode by three-dimensional (3D) numerical simulation with a moving mesh technique and using detached eddy simulation (DES) turbulent model. Mesh motion is carried out in the region of wicket gates during the load rejection by a moving, sliding mesh, which makes dynamic flow simulation available, instead of building various steady models with different guide vanes angles. The transient flow characteristics are illustrated by analyzing the flow, torque, and pressure fluctuations signals by frequency and time–frequency analyses. The flow field analysis includes the onset and strengthening of unsteady phenomena during the turbine load reduction. The flow pattern in return channel maintained a quite stable flow field, whereas the flow pattern in the runner and draft tube emphasized its instability with the flow rate decreased. Influence of 3D unsteady flow structures on runner is determined, and its evolution is characterized spectrally during fast closure of wicket gates.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2018;140(6):061104-061104-12. doi:10.1115/1.4039235.
OPEN ACCESS

This paper presents large eddy simulations (LESs) of symmetric and asymmetric (cambered) airfoils forced to undergo deep dynamic stall due to a prescribed pitching motion. Experimental data in terms of lift, drag, and moment coefficients are available for the symmetric NACA 0012 airfoil and these are used to validate the LESs. Good agreement between computed and experimentally observed coefficients is found confirming the accuracy of the method. The influence of foil asymmetry on the aerodynamic coefficients is analyzed by subjecting a NACA 4412 airfoil to the same flow and pitching motion conditions. Flow visualizations and analysis of aerodynamic forces allow an understanding and quantification of dynamic stall on both straight and cambered foils. The results confirm that cambered airfoils provide an increased lift-to-drag ratio and a decreased force hysteresis cycle in comparison to their symmetric counterparts. This may translate into increased performance and lower fatigue loads when using cambered airfoils in the design of vertical axis turbines (VATs) operating at low tip-speed ratios.

Commentary by Dr. Valentin Fuster

### Research Papers: Fundamental Issues and Canonical Flows

J. Fluids Eng. 2018;140(6):061201-061201-7. doi:10.1115/1.4038730.

Cavitation leads to rapid degassing of fluids. Up to date, there is a lack of model approaches of cavitation-induced degassing. The aim of the present study is to gain a more thorough knowledge of the process. Therefore, the relation between cavitation intensity and air release is investigated experimentally for an orifice flow as function of cavitation number. For this, shadowgraphy imaging is used to visualize regions of steam and air volume downstream of the orifice. Analysis of the images shows a strongly nonlinear scaling behavior for both cavitation intensity and air release as a function of cavitation number. Three distinct regimes could be identified for cavitation-induced gas release. While an exponential scaling was found at high cavitation intensities, degassing rates appear to be nearly constant in the intermediate cavitation number range. Empirical scaling laws are given here that may serve as first model approach for the prediction of cavitation induced air release behind flow constrictions.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2018;140(6):061202-061202-10. doi:10.1115/1.4039032.

Two-dimensional (2D) numerical simulations of multiphase flows past a circular cylinder close to free surface waves were performed to investigate an interaction of vortex formation around a cylinder with periodic waves by utilizing waves2Foam. The lock-on responses of lift coefficient at low frequencies were lower than those of the natural response due to the existence of “out-of-phase” between the fluctuations of lift coefficient by vortex formation and the vertical force fluctuations by wave motions. The resonant effect of an external excitation by the periodic waves on the lift coefficient fluctuations was not significant despite the occurrence of lock-on.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2018;140(6):061203-061203-8. doi:10.1115/1.4038668.

Numerical simulations of laminar pipe and channel flows were carried out: (i) to understand the effect of inlet conditions, viz., flat inlet and streamtube inlet, on entry lengths, and (ii) to investigate the flow development in radial/transverse locations. Results show that hydrodynamic entry lengths from the streamtube inlet simulations are significantly lower compared to the entry lengths from the flat inlet simulations for low Reynolds numbers. Moreover, results from the current study (Newtonian flow with no-slip) as well as the results from the literature (non-Newtonian flow with no-slip) showed that for many flow situations, the slowest development of axial velocity in the transverse location happens to be very near to the wall. For the above cases, the existing entry length criteria (centerline as well as global entry length) are not appropriate to define the entry length. We have proposed a new entry length criterion based on the displacement thickness which is an integral measure of the velocity profile. A new entry length correlation using the displacement thickness criterion is proposed for Newtonian flows in pipe and channel based on simulations with the streamtube inlet condition.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2018;140(6):061204-061204-10. doi:10.1115/1.4039120.

Transient pressure peak values and decay rates associated with water hammer surges in fluid lines are investigated using an analytical method that has been formulated, in a previous publication, to simulate pressure transients in turbulent flow. The method agrees quite well with method of characteristics (MOC) simulations of unsteady friction models and has been verified with experimental data available for Reynolds numbers out to 15,800. The method is based on the formulation of ordinary differential equations from the frequency response of a pressure transfer function using an inverse frequency algorithm. The model is formulated by dividing the line into n-sections to distribute the turbulence resistance along the line at higher Reynolds numbers. In this paper, it will be demonstrated that convergence of the analytical solution is achieved with as few as 5–10 line sections for Reynolds numbers up to 200,000. The method not only provides for the use of conventional time domain solution algorithms for ordinary differential equations but also provides empirical equations for estimating peak surge pressures and transient decay rates as defined by eigenvalues. For typical sets of line and fluid properties, the trend of the damping ratio of the first or dominate mode of the pressure transients transfer function is found to be an approximate linear function of a dimensionless parameter that is a function of the Reynolds number. In addition, a reasonably accurate dimensionless trend formula for estimates of the normalized peak pressures is formulated and presented.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2018;140(6):061205-061205-8. doi:10.1115/1.4039513.

Aerodynamic forces on bicycle racing wheels were investigated experimentally in a wind tunnel facility at the University of Calgary. The main geometric parameters investigated were the tire sidewall width, ($21.40 mm≤T≤26.15 mm$), rim depth, ($55 mm≤D≤90 mm$), rim width, ($19.28 mm≤W≤25.75 mm$), and angle of attack, ($0deg≤ψ≤12.6deg$). A total of six wheels and five tires were tested. Coefficient of drag area ($CdA$) and coefficient of side force area ($CsA$) versus angle of attack (ψ) were measured using a multi-axis force transducer. Wind tunnel conditions were set to simulate 2.94 m/s (6.6 mph) wind speeds and 13.4 m/s (30 mph) cyclist speeds. The performance of the wheel tire combinations was assessed using a wind-averaged drag method. The results show that the $CdA$ of aerodynamic wheels is highly dependent on the wheel-tire combination. There is a strong linear correlation between wind averaged drag ($CdA¯$) and the ratio of the tire and rim width $(T/W)$. The $CsA$ of wheels is primarily controlled by the rim depth (D) and only weakly correlated with tire width (T).

Commentary by Dr. Valentin Fuster

### Research Papers: Multiphase Flows

J. Fluids Eng. 2018;140(6):061301-061301-13. doi:10.1115/1.4038753.

A twin-fluid atomizer configuration is predicted by means of the two-dimensional (2D) weakly compressible smooth particle hydrodynamics (SPH) method and compared to experiments. The setup consists of an axial liquid jet surrounded by a high-speed air stream (Ug ≈ 60 m/s) in a pressurized reactor, which is operated at up to 11 bar. Two types of liquid are investigated: a viscous Newtonian liquid (μl = 200 mPa·s) consisting of glycerol/water mixture and a viscous non-Newtonian liquid (μ1,apparent. ≈ 150 mPa·s), which is a carboxymethyl cellulose solution. Three-dimensional (3D) effects are taken into account in the 2D code by introducing: (i) a surface tension term, (ii) a cylindrical viscosity operator, and (iii) a modified velocity accounting for the divergence of the volume in the radial direction. The numerical results at high pressure show a good qualitative agreement with experiment, i.e., a correct transition of the different atomization regimes with regard to pressure, and similar dynamics and length scales of the generated ligaments. The propagation velocity of the Kelvin–Helmholtz (KH) instability is well predicted, but its frequency needs a correction factor to be globally well recovered for the Newtonian liquid. The Sauter mean diameter (SMD), calculated from the spray size distribution, shows similar trends of the reactor pressure dependency. The simulation of the non-Newtonian liquid at high pressure shows the same breakup regime with finer droplets compared to Newtonian liquids, and the simulation at atmospheric pressure shows an apparent viscosity similar to the experiment.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2018;140(6):061302-061302-14. doi:10.1115/1.4039130.

The hydrodynamic entrance length, pressure drop analysis, viscosity, and fully developed velocity profile in horizontal pipe for crude oil with and without water and surfactant were studied in a 2 in ID horizontal pipe of length 2.5 m experimentally. Hydrodynamic entry length and fluid characteristics have been examined by varying temperature, water fraction, and flow rates. Temperature was varied by 25–40 °C, flow rates 40–60 LPM, and water 0–15% v/v and Madhuca longifolia from 500 to 2000 ppm. Triton X-100 was mixed with water to increase the emulsion capability during crude oil–water flows. The results showed significant influence of water, flow rate, and temperature on the hydrodynamic entry region length, pressure drop, viscosity, and velocity profiles along with natural surfactant. Pressure drop was reduced by 93.75%, 94.18%, and 93.02% with 15% water and 2000 ppm surfactant at 40 °C for 40 LPM, 50 LPM, and 60 LPM, respectively. Viscosity of the crude oil during flowing is greatly influenced by water and addition of surfactant. After addition of 2000 ppm surfactant and 15% water at 40 °C, viscosity reduced by about 94%. Hydrodynamic entry region length increased from 0.0354 to 0.2014 m, 0.0368 to 0.2336 m, and 0.0384 to 0.2641 m during transportation of crude oil after addition of 2000 ppm surfactant and 15% water at 40 °C for 40 LPM, 50 LPM, and 60 LPM flow, respectively.

Commentary by Dr. Valentin Fuster

### Research Papers: Techniques and Procedures

J. Fluids Eng. 2018;140(6):061401-061401-5. doi:10.1115/1.4038802.

Based on computational fluid dynamics and the finite volume method, water droplet erosion in the last stage of an industrial steam turbine was researched and the trajectories of the water droplets were traced by using the Lagrange method. Under steady conditions, the influence of variant bowed vane designs was compared based on the distribution and movement trends of the secondary water droplets. In addition, the effects of the bowed blades on water movement at the vane surfaces and their impact areas and intensity on the blades were analyzed. The results showed that: (1) a negatively bowed blade can reduce the speed of the secondary water droplets at the mid span of the blade, which are also effective for water droplets on the surface of the vanes and (2) a negatively bowed blade improves the speed of the secondary droplets on the end walls of vanes, which is advantageous to the secondary droplets through blade passage and reduction of secondary droplet impulse on the blades.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2018;140(6):061402-061402-10. doi:10.1115/1.4038866.

This paper is concerned with the study of a kind of discrete forcing immersed boundary method (IBM) by which the loosely aero-elasticity coupled method is developed to analyze turbine blade vibration. In order to reduce the spurious oscillations at steep gradients in the compressible viscous flowing field, a five orders weighted essentially nonoscillatory scheme (WENO) is introduced into the flow solver based on large eddy simulation (LES). The three-dimensional (3D) full-annulus domain of the last two stages of an industrial steam axial turbine is adopted to validate the developed method. By the method, the process of grid generation becomes very simple and the unsteady data transferring between stator and rotor is realized without the process of being averaged or weighted. Based on the analysis of some important aerodynamic parameters, it is believed that hypothesis of azimuthal periodicity is not reasonable in this case and full-annulus passages model is more feasible and suitable to the research of turbine blade vibration. Meanwhile, the blade vibration data are also discussed. It is at about 65% of rotor blade height of the last stage that an inflection point is observed and the midspan region of the blade is the vulnerable part damaged potentially by the blade vibration.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2018;140(6):061403-061403-10. doi:10.1115/1.4039087.

The Spalart–Allmaras (SA) is one of the most popular turbulence models in the aerospace computational fluid dynamics (CFD) community. In its original (low-Reynolds number) formulation, it requires a very tight grid spacing near the wall to resolve the high flow gradients. However, the use of wall functions with an automatic feature of switching from the wall function to the low-Reynolds number approach is an effective solution to this problem. In this work, we extend Menter's automatic wall treatment (AWT), devised for the k–ω-shear stress transport (SST), to the SA model in our in-house developed three-dimensional unstructured grid density-based CFD solver. It is shown, for both momentum and energy equations, that the formulation gives excellent predictions with low sensitivity to the grid spacing near the wall and allows the first grid point to be placed at y+ as high as 150 without loss of accuracy, even for the curved walls. In practical terms, this means a near-wall grid 10–30 times as coarse as that required in the original model would be sufficient for the computations.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2018;140(6):061404-061404-15. doi:10.1115/1.4039158.

The parallel plate flow chamber (PPFC) has gained popularity due to its applications in fields such as biological tissue engineering. However, most of the studies using PPFC refer to theoretical relations for estimating the wall shear stress (WSS) and, hence, the accuracy of such quantifications remains elusive for anything other than steady laminar flow. In the current study, a laser Doppler velocimetry (LDV) method was used to quantify the flow in a PPFC (H = 1.8 mm × W = 17.5 mm, Dh = 3.26 mm, aspect ratio = 9.72) under steady Re = 990, laminar pulsatile (carotid Re0-mean = 282 as well as a non-zero-mean sinusoidal Re0-mean = 45 pulse) and low-Re turbulent Re = 2750 flow conditions. A mini-LDV probe was applied, and the absolute location of the LDV measuring volume with the respect to the wall was determined using a signal monitoring technique with uncertainties being around ±27 μm. The uniformity of the flow across the span of the channel, as well as the WSS assessment for all the flow conditions, was measured with the uncertainties all being less than 16%. At least two points within the viscous sublayer of the low-Re turbulent flow were measured (with the y+ for the first point < 3) and the WSS was determined using two methods with the differences between the two methods being within 5%. This paper for the first time presents the experimental determination of WSS using LDV in a small-scale PPFC under various flow conditions, the challenges associated with each condition, and a comparison between the cases. The present data will be useful for those conducting biological or numerical modeling studies using such devices.

Commentary by Dr. Valentin Fuster