Search

This work presents the implementation, verification and the validation of an incompressible Eulerian multi-fluid model for polydisperse flows. The proposed model uses a novel monolithic, i.e. implicitly coupled phase continuity equation for an arbitrary number of fluids, where the breakup source and sink terms are handled implicitly in the block-system. The implemented model is tested for an upward bubbly flow inside a large vertical pipe. The selected flow conditions exhibit both breakup and coalescence. The grid refinement study is conducted on four structured grids with varying levels of refinement. In the validation section, the numerical results are compared to the TOPFLOW experimental measurements. The last presented test examines the performance of the novel implicitly coupled phase continuity equation to the corresponding segregated formulation and the standard segregated formulation. The performance is evaluated by comparing the conservation error over the non- linear iterations. The presented model exhibits good agreement with the experimental measurements and gives stable results on various grids with different levels of refinement. Moreover, the implicit coupling reduces the conservation error during the calculation.

In finite-volume-based flow simulations with free-surface waves, wave reflections at the domain boundaries can cause substantial errors in the results. ‘Implicit relaxation zones’ can be used to minimize these reflections, but only if the relaxation zone’s case-dependent parameters are optimized. This work proposes an analytical approach for optimizing these parameters. The analytical predictions are compared against results from 2D-flow simulations for different water depths, wave steepnesses, flow solvers, and relaxation functions, and against results from 3D-flow simulations with strongly wave-reflecting bodies subjected to nonlinear free-surface waves. The present results demonstrate that the proposed approach satisfactorily predicts both the optimum parameter settings and the upper-limit for the corresponding reflection coefficients C R . When optimized as proposed, simulation results for C R were mostly below or equal to the analytical predictions, but never more than 2.2% larger. Therefore, the proposed approach can be recommended for engineering practice. Furthermore, it is shown that implicit relaxation zones can be considered as a special-case of ‘forcing zones’, a family of approaches which includes absorbing layers, damping zones and sponge layers amongst others. The commonalities and differences between these approaches are discussed, including to which extent the present findings are applicable to these other approaches and vice versa.

Computational Fluid Dynamics (CFD) simulations based on the Finite Volume Method are often used to predict self-propulsion characteristics and to describe shippropeller interaction. Only recently, this kind of simulations has become feasible, as it requires substantial computing resources and a complete set of solver capabilities. Accurate representation of the rotating propellers and capturing of the free surface results in complexity and cost, which are still limitations for wider adoption in a preliminary design stage. Therefore, to reduce overall computational time while achieving reasonably accurate integral and local characteristics, a method using double-body model and a partially rotating grid is proposed.

In this paper, partially rotating grid method for predicting self-propulsion characteristics using a double body ship model approach is presented. Partially rotating grid method combines a full- discretized temporal-domain rotating propeller approach with locally introduced non-inertial equation system to reduce overall computation time. Verification and validation of the method are performed in this paper. Numerical self- propulsion simulations are conducted for a single- screw KRISO container ship both in model and full- scale. Obtained results are analyzed in terms of integral (propulsion factors) and local (velocity field) flow characteristics with a focus on ship- propeller interaction and scale effects. To better understand propeller performance and ship- propeller interaction effects, variation of a single blade KT, KQ, and KT/KQ values during one period of rotation are analyzed. The introduced partially rotating grid method proved to accurately predict the integral and local flow characteristics while reducing overall computation time, at cost of reduced accuracy.

Validation of a contact model which is sensitive to the elastoplastic deformation of asperities and rheological changes to the lubricant properties is presented. The contact model was derived using a two-dimensional variant of the Finite Volume Method, the Finite Area Method, and was implemented in foam- extend, an OpenFOAM fork. The contact lubrication model combines a modified Reynolds equation incorporating a cavitation algorithm, a deterministic contact model and a thermal contact model. The final model can be used to analyse lubricant flow and cavitation, changes of lubricant properties, including changes due to temperature variations and non-Newtonian effects, and asperity contact of rough surfaces. The cumulative validation of the aforementioned models was performed on two ball-on-disc test cases, which use two different types of oil (Turbo T9 and T68 oils). The numerical results were validated against experimental measurements for ball-on-disc tribological experiments found in the literature and available numerical data.

Stringent rules and regulations to reduce ship environmental footprint and continuous aim to improve the propulsion efficiency stimulate development of different technical solutions. Typical technical solution that reduces ship total power needs (particularly for the propulsion) is the Energy saving device (ESD), i.e. special hydrodynamic appendage installed on the ship aft body. In this paper, the hydrodynamic loadings on circular duct (one of typical ESDs) of a bulk carrier are analysed in calm water and waves by CFD. To reduce the CPU time, actuator disk model is used instead of the direct propeller simulation. Effect of the actuator disk on the pressure distribution is investigated and discussed. Results are compared to available experimental measurements in calm water (wake field, resistance, self-propulsion) in order to validate the proposed simplified model. Additional simulations are performed in waves and a detailed evaluation of duct loadings is presented along with the duct performance in waves. Simulations are performed by means of OpenFOAM. Findings display a significant effect of the propeller suction on the ESD pressure distribution indicating its importance when assessing the structural reliability of the ESDs both in calm water and waves.

In this paper, a detailed numerical analysis of global, and local flow characteristics of a self- propelled KCS model is presented. Global flow characteristics are integral values such as thrust and torque, while the local flow characteristics are represented by the velocity components. Three different numerical propeller representations are considered: fully discretized propeller, actuator disc model which models the thrust effect on the fluid flow only, and actuator disc model that models both thrust and torque effects. Verification is performed using method based on Richardson extrapolation, and the approach is validated with publicly available experimental data, both for global and local flow characteristics. A procedure for accurate prediction of global and local self-propulsion characteristics using a combination of actuator disc and discretized propeller methods is presented. The hybrid procedure was proved to provide a good agreement of global flow characteristics with experimental data. The fully discretized propeller model using overset grid showed good accuracy of predicting local flow characteristics, observed by comparing axial and in-plane velocity components behind the working propeller.

Advancements in internal combustion technology, such as efficiency improvements and the usage of new complex fuels, are often coupled with developments of suitable numerical tools for predicting the complex dynamic behavior of sprays. Therefore, this work presents a Eulerian multi-fluid model specialized for the dynamic behavior of dense evaporating liquid fuel sprays. The introduced model was implemented within the open-source OpenFOAM library, which is constantly gaining popularity in both industrial and academic settings. Therefore, it represents an ideal framework for such development. The presented model employs the classes method and advanced interfacial momentum transfer models. The droplet breakup is considered using the enhanced WAVE breakup model, where the mass taken from the parent droplets is distributed among child classes using a triangular distribution. Furthermore, the complex thermal behavior within the moving droplets is considered using a parabolic temperature profile and an effective thermal conductivity approach. This work includes an uncertainty estimation analysis (for both spatial and temporal resolutions) for the developed solver. Furthermore, the solver was validated against two ECN Spray A conditions (evaporating and non-evaporating). Overall, the presented results show the capability of the implemented model to successfully predict the complex dynamic behavior of dense liquid sprays for the selected operating conditions.

This work restricts the term multiphase only to disperse flows, where one of the phases is present in the form of particles, droplets or bubbles, which are suspended within the continuous phase. The dispersed elements can vary in size. The proposed method uses the classes method in the Euler-Euler framework to handle the flow's polydisperse nature. With this approach, every droplet/bubble/particle class is treated like a different phase in the calculation, i.e. every size class has its continuity and momentum equation. However, the pressure is shared among all phases. The derived model is tested for various polydisperse flows, which display the developed model's capability to predict such complex dynamic behaviour. These test cases include complex bubbly flows and dense spray (where droplet sizes vary significantly).

In the present numerical study, implicit large eddy simulations (LES) of non-reacting multi-components mixing processes of cryogenic nitrogen and n-dodecane fuel injections under transcritical and supercritical conditions are carried out, using a modified reacting flow solver, originally available in the open source software OpenFOAM®. To this end, the Peng-Robinson (PR) cubic equation of state (EOS) is considered and the solver is modified to account for the real-fluid thermodynamics. At high pressure conditions, the variable transport properties such as dynamic viscosity and thermal conductivity are accurately computed using the Chung transport model. To deal with the multicomponent species mixing, molar averaged homogeneous classical mixing rules are used. For the velocity-pressure coupling, a PIMPLE based compressible algorithm is employed. For both cryogenic and non-cryogenic fuel injections, qualitative and quantitative analyses are performed, and the results show significant effects of the chamber pressure on the mixing processes and entrainment rates. The capability of the proposed numerical model to handle multicomponent species mixing with real-fluid thermophysical properties is demonstrated, in both supercritical and transcritical regimes.

In this paper RANS based CFD calm-water resistance and propulsion simulations of two benchmark cases (container ship KCS and bulk carrier JBC) are presented. Two-phase flow is handled using the Volume-of-Fluid method, while the discontinuities at the free surface are resolved using the Ghost Fluid Method. Propeller is modeled using an idealized pressure-jump actuator disk model, allowing large time step size suitable for naval hydrodynamics with the preservation of accuracy and CPU time efficiency. Computations are performed with the Naval Hydro Pack developed in FV framework of foam-extend, a community driven fork of OpenFOAM. Model scale KCS resistance and wave pattern is evaluated with a detailed analysis of the bare hull wake field following ITTC recommendations. Self propulsion simulation is carried out for the JBC case comparing the global features of the flow with the experiment.

An implicitly coupled pressure–velocity transient scheme for the solution of the Eulerian multi-fluid model with any number of phases (MIC) was developed and implemented in OpenFOAM®. The numerical methodology is based on the phase-intensive momentum equations for the phase velocities and a pressure equation using a deferred-correction approach of the Rhie–Chow interpolation. An extension of the compact momentum interpolation method (CMI) to a system with any number of phases was developed. The developed methodology was tested against a segregated multiphase solver (MS) and a steady-state single phase solver, in cases considering up to four phases. The proposed method has proven to be more robust and accurate than the segregated counterpart, being able to converge in cases with large drag coefficients, providing time-step independent steady-state solutions.

In this work, a block coupled algorithm for the solution of laminar, incompressible viscoelastic flow problems on collocated grids is presented. The inter-equation coupling of the incompressible Cauchy momentum equations and extra-stress tensor constitutive equation is obtained by deriving a pressure equation in a procedure similar to a SIMPLE algorithm with the Rhie-Chow interpolation technique, and a special treatment of the diffusion term in the momentum equations added by the improved both-sides diffusion (iBSD) technique, recently developed for the finite volume method. Additionally, the velocity field is considered implicitly in the extra-stress tensor constitutive equation by expanding the convective term with a second order Taylor series expansion. All the equations, comprising the continuity, momentum and extra-stress tensor constitutive equations are block-coupled into a single system of equations. The implicitly coupled system of equations is solved by an algebraic multigrid algorithm, which allows to accelerate the calculation process. The performance improvements obtained with the new solver are studied for three 2D viscoelastic flow case studies, and are quantified in terms of number of iterations and CPU time required to reach the predefined convergence criteria. The presented algorithm has been implemented into the open-source computational library foam-extend, a community driven fork of the OpenFOAM software., This work is funded by UID/CTM/50025/2013 - LA0025, with the financial support of FCT/MEC through national funds and when applicable by FEDER co-funded, within the partnership agreement PT2020. The authors would like to acknowledge the Minho University cluster under the project Search-ON2: Revitalization of HPC infrastructure of UMinho, (NORTE-07-0162-FEDER-000086), co-funded by the North Portugal Regional Operational Programme (ON.2-0 Novo Norte), under the National Strategic Reference Framework (NSRF), through the European Regional Development Fund (ERDF).

The present work addresses the numerical simulation of rising droplets with clean interface. An interface-tracking methodology was applied to simulate the particular behavior of water-immersed toluene drops. The obtained transient velocity profiles were compared with experimental data, confirming the methodology’s capability to represent this two-phase flow. Numerical simulations enable scrutinizing many flow details which are not accessible experimentally. In this regard, the elements linked to the velocity oscillatory behavior of droplet sizes between diameters of 2.2 and 3 mm are discussed. It could be verified that velocity oscillations were directly related to variations in shape, internal circulation and flow separation. Simulations indicated that three-dimensional shape asymmetries are associated with vorticity accumulation in preferential regions of the drop, which disturb the internal circulation pattern. In the oscillatory size range, flow separation causes periodic vortex detachment, enabling formation of a double-thread wake. Evaluation of non-dimensional parameters further disclosed the relation of the transitory behavior and steady state drop shape.

This paper describes a self-contained parallel fluid-structure interaction solver based on a finite volume discretisation, where a strongly coupled partitioned solution procedure is employed. The incompressible fluid flow is described by the Navier-Stokes equations in the arbitrary Lagrangian-Eulerian form, and the solid deformation is described by the Saint Venant-Kirchhoff hyperelastic model in the total Lagrangian form. Both the fluid and the solid are discretised in space using the second-order accurate cell-centred finite volume method, and temporal discretisation is performed using the second-order accurate implicit scheme. The method, implemented in open-source software OpenFOAM, is parallelised using the domain decomposition approach and the exchange of information at the fluid-solid interface is handled using global face zones. The performance of the solver is evaluated in standard two- and threedimensional cases and excellent agreement with the available numerical results is obtained.

The Harmonic Balance method applied to fully nonlinear, viscous, temporally periodic free surface flows is presented in this paper. The Harmonic Balance approach is used to project a transient periodic two phase flow into multiple coupled steady–state problems discretising the periodic time interval. The numerical framework is based on Finite Volume method for Computational Fluid Dynamics in the open–source software foamextend. The disability of the Harmonic Balance method to simulate flows with zero or close to zero mean velocity is overcome by coupling the Harmonic Balance steady–state equations implicitly, presenting a novelty in the field. Von Neumann stability analysis is performed for the governing Harmonic Balance equations to mathematically prove that zero mean velocity can be simulated using the proposed implicit coupling. The method is validated and verified on a periodic free surface flow over a ramp and regular surface wave propagation with current, both including grid convergence studies and spectral resolution sensitivity studies. All results are compared to fully transient computations. A detailed study is performed for wave propagation without current, confirming that convergence of free surface elevation can be achieved without mean velocity. The approach proved accurate and applicable for twophase flows, opening a new research field.

An extensive verification and validation for green sea load simulations is presented. The calculations are per- formed using the Naval Hydro pack, a library based on foam–extend, which is an open source Computational Fluid Dynamics software. The geometric Volume of Fluid method is used for interface advection, while the Ghost Fluid Method is employed to discretise the free surface boundary conditions at the interface. Pressure measured at the deck of a fixed structure is compared to experimental data for nine regular waves. Verification is performed using four refinement levels in order to reliably assess numerical uncertainties. A detailed uncertainty analysis comprises both numerical and experimental data. Comparable uncertainties are exhibited in simulations and experiments, with good agreement of results.

In Fluid Structure Interaction (FSI) problems encoun- tered in marine hydrodynamics, the pressure field and the velocity of the rigid body are tightly coupled. This coupling is traditionally resolved in a partitioned manner by solving the rigid body motion equations once per nonlinear correction loop, updating the position of the body and solving the fluid flow equations in the new configura- tion. The partitioned approach requires a large number of nonlinear iteration loops per time–step. In order to enhance the coupling, a monolithic approach is proposed in Finite Volume (FV) framework, where the pressure equation and the rigid body motion equations are solved in a single linear system. The coupling is resolved by solving the rigid body motion equations once per linear solver iteration of the pressure equation, where updated pressure field is used to calculate new forces acting on the body, and by introducing the updated rigid body boundary velocity in to the pressure equation. In this paper the monolithic coupling is validated on a simple 2D heave decay case. Additionally, the method is compared to the traditional partitioned approach (i.e. “strongly coupled” approach) in terms of computational efficiency and accuracy. The comparison is performed on a seakeeping case in regular head waves, and it shows that the monolithic approach achieves similar accuracy with fewer nonlinear correctors per time–step. Hence, significant savings in computational time can be achieved while retaining the same level of accuracy.

A novel two-way coupled Eulerian-Eulerian CFD formulation is presented, to simulate drifting snow based on a new viscous model of the snow phase. This approach allows explicit resolution of the saltation layer without resorting to empiricism, unlike other Eulerian-Eulerian approaches based on mixture formulations and one-way coupling. Initial validations carried out against detailed snowflux, airflow, and turbulence measurements in a controlled drifting snow wind tunnel experiment show the present approach is capable of accurately predicting snowflux, airflow profiles, as well as snow transport under developing and nearly developed flow conditions, when polydispersity is taken into account.

Surf-riding phenomenon occurs when a ship sailing in following waves is accelerated to wave celerity. In this paper, surf-riding occurrence is studied for the parent hull of the Systematic Series D by 3 degrees of freedom (DOF) CFD simulations. Simulation cases are chosen following IMO recommendations for a wavelength to ship length range and following IMO criterion for ship speed cases. Initial conditions for simulations are obtained from 1-DOF surge motion equation solved by bifurcation analysis. The aim is to define surf-riding limits for ship operability with a less conservative approach, taking into account force changes in waves compared to the one used in the 1-DOF model, which considers calm water approximations. After defining surf-riding boundaries in terms of wave heights for different wavelengths and ship speeds, calculated forces are analysed to evaluate the magnitude and trend of non-linear effects due to wave elevation and ship-wave interaction. A detailed analysis is conducted on resistance, thrust, and wave forces in surging and surf-riding conditions, comparing the applied methodologies. The introduction of wave velocity field influence in the thrust force and the calculation of Froude-Krylov wave-force up to actual wave profile have been discussed and identified as possible improvements of 1-DOF approach.

This chapter covers the fundamental aspects of Computational Fluid Dynamics simulation tools and introduces the terminology and principles of the second order accurate Finite Volume Method with polyhedral mesh support, as implemented in OpenFOAM (Weller et al. 1998). The first part is dedicated to types and properties of computational meshes, followed by a description of the Finite Volume discretisation for a generic scalar transport equation. Algorithms for pressure-velocity coupling for single–phase incompressible Newtonian fluid flow, as described by the Navier-Stokes equations are presented, supported by an overview of algorithms for the solution of the resulting linear system of algebraic equations. Application of the polyhedral FVM solver for real engineering problems is illustrated on some practical examples.

The paper considers the side launching of ships from a horizontal berth by tipping tables. The launching process is split into the six phases and a system of nonlinear differential equations of motion is established. The system is solved using the finite difference method. The numerical calculations of the side launching of a large tanker are performed and the results are compared against the model test results. The generation of a high surface wave is simulated with CFD as a 2D problem. The numerical simulation captures the physics of the side launching with high accuracy. The advantages of building the floating structures on a horizontal berth are pointed out.

Ongoing development of a CFD framework for the simulation of primary atomization of a high pressure diesel jet is presented in this work. The numerical model is based on a second order accurate, polyhedral Finite Volume (FV) method implemented in foam-extend-4.1, a community driven fork of the OpenFOAM software. A geometric Volume-of-Fluid (VOF) method isoAdvector is used for interface advection, while the Ghost Fluid Method (GFM) is used to handle the discontinuity of the pressure and the pressure gradient at the interface between the two phases: n-dodecane and air in the combustion chamber. In order to obtain highly resolved interface while minimizing computational time, an Adaptive Grid Refinement (AGR) strategy for arbitrary polyhedral cells is employed in order to refine the parts of the grid near the interface. Dynamic Load Balancing (DLB) is used in order to preserve parallel efficiency during AGR. The combination of isoAdvector-GFM-AGR-DLB presents a unique framework for diesel jet atomization. The developed numerical framework is preliminarily tested on the Engine Combustion Network (ECN) Spray D geometry and conditions. The unstructured, mostly hexahedral grid is used with the base cell size of 40 micrometres. Four refinement levels are used in the close proximity of the interface in order to attempt to resolve break-up of droplets. The finest cells near the interface have the size of 2.5 micrometres. Part of the nozzle is also considered in the simulation in order to capture the developed jet profile at the entry into the combustion chamber. The temporal evolution of the jet is presented, along with the preliminary comparison of droplet statistics with available results.

Results from Blind Test Series 1, part of the Collaborative Computational Project in Wave Structure Interaction (CCP-WSI), are presented. Participants, with a range of numerical methods, simulate blindly the interaction between a fixed structure and focused waves ranging in steepness and direction. Numerical results are compared against corresponding physical data. The predictive capability of each method is assessed based on pressure and run-up measurements. In general, all methods perform well in the cases considered, however, there is notable variation in the results (even between similar methods). Recommendations are made for appropriate considerations and analysis in future comparative studies.

The present paper and its companion present a solution and domain decomposition model for general two-phase, incompressible and turbulent flows encountered in naval hydrodynamics. The mathematical and numerical model are derived within the framework of polyhedral finite volume method. Interface capturing is obtained with implicitly redistanced level set method derived from the phase field equation. This approach removes the need to redistance the level set field, thus saving CPU time and increasing numerical stability. A modified, Spectral Wave Explicit Navier–Stokes Equations method is introduced and used for wave modelling, where the solution is decomposed into incident and perturbation fields. The incident field is readily available from potential flow theories, and only the perturbation component is solved within the non-linear equation set of the free surface flow model. The domain is decomposed with implicit relaxation zones used to prevent wave reflection by forcing the perturbation fields to vanish in the far-field. A second-order, collocated finite volume method is used to discretise the equations. All equations are solved implicitly, which enables the use of higher Courant–Friedrichs–Lewy numbers compared to explicit scheme. The algorithm is implemented in foam-extend-3.1, a community driven fork of the OpenFOAM CFD software. Verification and validation of the model is presented in the accompanying paper.

Turbomachinery simulations represent one of the most challenging fields in Computational Fluid Dynamics (CFD). In the recent years, the general CFD ca- pabilities of foam-extend have been extended by introducing and maintaining addi- tional features specifically needed for turbomachinery applications, with the aim of offering a high quality CFD tool for the study of rotating machinery. This work presents the implementation and validation of new capabilities for tur- bomachinery with foam-extend, a community driven fork of OpenFOAM. The for- mulation of energy equation more convenient for turbomachinery compressible ap- plications is the rothalpy equation. Rothalpy is a physical quantity conserved over a blade row, stator or rotor, but not over a stage, both stator and rotor. It is fun- damental to take into account that the value of rothalpy is not continuous across the rotor-stator interface due to the change of rotational speed between zones. The rothalpy equation has been derived for both relative and absolute frame of reference, showing that additional terms appear in absolute frame of reference. Moreover, ad- ditional functionality has been added to the rotor-stator interface boundary condi- tions General Grid Interface (GGI), partial overlapGGI and Mixing Plane Interface in order to account for the rothalpy jump. The development of these new capabili- ties and their validation are shown as well as industrial applications of compressible turbomachinery flows.

Climate change, energy security and air pollution are all motivators for the further development of fuel cells. A volume of fluid approach was used to investigate the impact of dynamic contact angle boundary conditions (Kistler model), mainly at the gas diffusion layer surface but also at the channel wall, of a polymer electrolyte fuel cell gas channel. From this study, it is clear that a dynamic contact angle boundary condition, with advancing and receding contact angles, influences the droplet detachment characteristics, for example, the detachment time and droplet size. Implementing dynamic contact angle boundary conditions for a thin channel causes the droplet, after being reattached to the wall on the side opposite the GDL, to flow very slowly when attached to the wall, until it is merged with a second droplet and they exit the channel (but remain attached to the wall) fairly quickly. Similar phenomena are not observed while using a static contact angle.

Four different types of breaking wave impacts on a vertical wall are simulated using a two-dimensional two-phase Computational Fluid Dynamic (CFD) model. Air is considered as an isentropic ideal gas without solving an additional energy equation and water is treated as an incompressible liquid. The discontinuity in fluid properties across the free surface is treated using the Ghost Fluid Method, which accounts for the jump in density and compressibility. The numerical results are compared with large-scale experimental data for five cases in terms of surface elevations, total forces and pressure distributions along the wall, and a reasonable agreement is obtained overall. The characteristics of impact pressures under different breaking wave conditions are discussed and compared to each other. The two largest total forces on the wall occur in the ‘flip-through’ and ‘large air pocket’ cases. The peak pressure of the flip-through impact is localized in both time and space. The pressure within the trapped air pocket is nearly uniform with a smaller peak value and a much longer duration than that of the ‘flip-through’ case. The compression and expansion of the air pocket results in pressure oscillations, which are overestimated in frequency and amplitude due to the inaccuracy in capturing the air escape. The broken wave case has the smallest total force, but the largest local pressure among the present numerical cases, which demonstrates the necessity to study all different impact types.

A novel two-way coupled Eulerian–Eulerian CFD formulation was developed to simulate drifting snow based on turbulent drag and a new viscous treatment of the drifting snow phase, derived from first principles. This approach allowed explicit resolution of the saltation layer without resorting to empiricism, unlike other Eulerian–Eulerian models based on mixture formulations and one-way coupling. Initial validations were carried out against detailed snow flux, airflow velocity, and turbulent kinetic energy measurements in a controlled experimental simulation of drifting snow in a wind tunnel using actual snow particles. The two-way coupled approach was found capable of simulating drifting snow fluxes in both saltation and suspension layers with reasonable accuracy. Recommendations were made to improve the accuracy of the method for air velocity and turbulent kinetic energy, and to allow simulating a drifting snow phase with a particle size distribution.