Computational Fluid Dynamics - Science topic

What should be the Capillary Number obtained for water flow inside a silicone microchannel so that we can ignore the Capillary Effect in this study?

My study investigated forced flow with Reynolds numbers between 125 and 1300.

If our criterion for the capillary number is 1, and we consider the capillary effect non-negligible for low values of 1, according to the capillary equation, the capillary effect cannot be neglected in many conditions and cases. For this reason, I think the value of 1 is not a critical value.

Also, the denominator of the capillary number equation is related to the surface tension parameter. Is the value of this parameter equal to 0.0726 N/m, which is the surface tension between water and air, or should we put the surface tension between water and a solid wall (silicon)? In many research studies, authors have used the value of 0.0726 N/m.

Ca=μ*U​/σ

"How do advanced computational modeling techniques, such as finite element analysis or computational fluid dynamics, aid in the precise characterization and optimization of thermal bridging phenomena within complex building assemblies?"

In CFD simulations, determining the appropriate inlet and outlet boundary conditions is crucial for accurately modeling recirculation phenomena in both two-dimensional (2D) and three-dimensional (3D) scenarios.

For recirculation simulations, the inlet boundary condition typically involves prescribing the flow properties entering the domain. This may include specifying the velocity profile, temperature, turbulence characteristics, and any other relevant parameters. In 2D simulations, the inlet boundary condition can be defined as a 2D plane through which fluid enters the computational domain. In 3D simulations, this boundary condition extends to a full 3D volume or surface. In both 2D and 3D simulations, accurately representing the inlet and outlet boundary conditions is critical for capturing the complex flow dynamics associated with recirculation zones. Properly defined boundary conditions ensure that the simulated flow field closely matches the real-world behavior, thus enhancing the reliability and accuracy of the CFD predictions.

hello Everyone,

if I want to study, coal combustion in different atmospheres for example in O2/N2 and O2/CO2. I obtain the kinetics of both atmospheres using Hetrogenous models. ( Shrinking core model ) and Random pore model.

I was wondering if it’s possible in CFD to study particle profile. Does the heat transfer affect in CFD will calculated based on the Gas composition input or I should add something in UDF file.

FYI, the reaction models are based on conversion so I am not really sure how CFD will identify the differences in Atmospheres.

further, I wish If I found a sample UDF file that been used for Hetrogensous models.

Ahmad

Hello Everyone,

I have some queries about if its possible using heterogeneous models reactions to model particle Temperature profile in CFD ?

I performed Kinetic analysis of my reaction (coke oxidation for example) my only concern what should I do in CFD. Also Let's say I am performing the reaction into two atmosphere, how the CFD will identify the reactant since heterogeneous model are based on conversions.

This is a code block from nutWallFunction library in OpenFOAM where in, effective kinematic viscosity ($\nut_w$) at the wall is calculated using resolved field(in case of LES)/ mean field(in case of RANS) and $y^+_p$ (wall normal distance of the first cell center). this allows to set a new viscosity value as boundary condition at the wall using log law. Considering the first cell center is in the logarithmic layer of the universal velocity profile.

Now, in this code block of member function defined as nutUWallFunctionFvPatchScalarField::calcYPlus()

There has been iterations done for the yPlus value to reach convergence with maximum of 10 iterations. Why are these iterations needed? and why is the maximum number of iterations 10. I have given a reference of the code below;

tmp<scalarField> nutUWallFunctionFvPatchScalarField::calcYPlus

(

const scalarField& magUp

) const

{

const label patchi = patch().index();

const turbulenceModel& turbModel = db().lookupObject<turbulenceModel>

(

IOobject::groupName

(

turbulenceModel::propertiesName,

internalField().group()

)

);

const scalarField& y = turbModel.y()[patchi];

const tmp<scalarField> tnuw = turbModel.nu(patchi);

const scalarField& nuw = tnuw();

tmp<scalarField> tyPlus(new scalarField(patch().size(), 0.0));

scalarField& yPlus = tyPlus.ref();

forAll(yPlus, facei)

{

scalar kappaRe = kappa_*magUp[facei]*y[facei]/nuw[facei];

scalar yp = yPlusLam_;

scalar ryPlusLam = 1.0/yp;

int iter = 0;

scalar yPlusLast = 0.0;

do

{

yPlusLast = yp;

yp = (kappaRe + yp)/(1.0 + log(E_*yp));

} while (mag(ryPlusLam*(yp - yPlusLast)) > 0.01 && ++iter < 10 );

yPlus[facei] = max(0.0, yp);

}

return tyPlus;

}

My doubt is concerning the do-while loop at the end for yPlus iteration.

I need to use computational fluid dynamics software to obtain wake data for LHA ship, and I would like to get your help.

I am new to limiters in FVM CFD so pardon me if its a blunder. Was curious about Venkatakrishnan's limiter (first image) that is industry standard in many commercial CFD solvers. He came up with the final form as given below and further modified the Δ_ term in order to avoid division by numerical values close to zero. Why not eliminate Δ_ term by taking out one from the numerator(second image)? Any advantage of writing in this form(first image)? ref: 10.1006/jcph.1995.1084

ref:

Low Prandtl number fluids like air, Pr <1, takes very long endTime for simulation to get the fully developed convection cells pattern and thermal boundary layers. What is the possible reason for this? Is it solely because of the dependency on kinematic viscosity (numerator of prandtl) being very low? Or is this related to simulation case setup.

Dear Friends,

I am looking for a faculty or Engineers or others who are interested in working on CFD of multiphase flow in chemical engineering reactors for collaboration research. Please contact me on Facebook messenger or text and call on +9647713171293.

Associate Prof Haidar Taofeeq

Can we do it on COMSOL or Ansys (CFD)?

Please can you clarify about this question.

computational approach

I am in search of commercial Computational Fluid Dynamics (CFD) software tailored for modeling hydraulic structures, particularly for applications related to stormwater, small dams, flood control structures, channels, and similar scenarios. The primary focus will be on studying free-surface flows. Affordability is a crucial consideration in the selection process. While I am already familiar with Flow-3D, Ansys Fluent, and Ansys CFX, I am open to exploring additional software options. I would appreciate any recommendations for alternative applications, and it would be beneficial to receive insights on various features to facilitate a comprehensive comparison for making an informed decision.

Hello everyone, I'm working on a CFD analysis to compute wind pressure coefficients (Cp) on a building facade. My objective is to use these Cp values in EnergyPlus AFN to evaluate the impact of natural ventilation on thermal comfort. While reviewing various studies that have undertaken similar work, I've encountered conflicting information regarding the parameter Uref. I encountered some sources that recommend default values of 10 m/s for Uref and 10 m for Zref. Nevertheless, some research papers suggest that the dynamic wind pressure (Pd) necessary for Cp calculations should be determined at the building height. Moreover, one source suggests setting Uref as "the largest freestream velocity at the top of the [wind tunnel] modeling domain". My analysis involves a parametric design, exploring different building heights. So I suppose I need to establish a consistent reference height for all simulations. I've already obtained weather data containing wind speeds and directions and have calculated average wind speeds at a height of 10m for 12 different directions. My question is whether these averaged values should be directly input as boundary conditions for the cylindrical domain or if I should stick with a fixed value for Uref, like the mentioned 10 m/s. Any guidance to clarify this matter would be greatly appreciated.

I wonder if there is a document like tutorial that is useful to simulate an Archimedes Screw

Hello, the Research community,

I'm currently working on a project involving heat transfer between a solid domain and a liquid domain using Ansys Fluent, and I'm facing some challenges. Here's a brief overview of my setup:

I aim to simulate and analyze the heat transfer process between the solid domain walls and the flowing liquid. I'm seeking guidance on the further steps to solve this problem effectively. Specifically, I'm looking for advice on:

I would greatly appreciate any insights, tips, or step-by-step guidance from those with experience in conducting heat transfer simulations in Ansys Fluent. Your assistance will be invaluable in helping me advance my project.

In addition, kindly guide me through the necessary steps to create an effective heat transfer interface between the solid and liquid domains in Ansys Fluent. Any insights, tips, or tutorials to help me set up this heat transfer simulation would be greatly appreciated.

Thank you in advance for your assistance.

Thank you in advance for your support.

Best regards,

Sudeep N S

I would like your expertise and guidance regarding a challenging Computational Fluid Dynamics (CFD) problem I currently have. The project involves species transport in Ansys Fluent, focusing on simulating a steam methane reforming process.

The issue I am encountering pertains to the Chemkin files that I have uploaded into the Ansys Fluent software. Unfortunately, I am encountering persistent errors, and it seems the software is not properly considering the uploaded Chemkin files. I have thoroughly reviewed the inputs, but the problem persists. Despite my best efforts, I have been unable to resolve this issue.

Furthermore, in this simulation, a key aspect involves maintaining a wall at an elevated temperature of 2000 degrees Celsius. The objective is to facilitate heat transfer to the methane and steam feed mixture to initiate the steam methane reforming reaction.

Regrettably, I have observed that the heat is not effectively transferring to the feed mixture, resulting in the absence of the desired chemical reaction. This poses a significant setback to the project, as the core objective is to study the reaction kinetics and product distribution in this specific environment.

Your support in this matter would be immensely important to my research progress.

Thank you in advance for considering my request. I am eager to learn from your expertise and am open to any suggestions or instructions you may have.

Sincerely,

Sudeep N S

FORTRAN in CFD Solvers

Hello dear all,

We want to set-up a CFD -Discrete Particle Modelling case for a bubble column where liquid is eulerian and gas bubbles and solid particles are discrete phases (three-phases total holdup <10 %) as given in attached diagram. In this case only gas bubbles are flowing with velocity UG <1 cm/s while liquid and solid particles have no initial velocity (means batch mode). Boundary condition (DPM) for gas phase at outlet is pretty straight forward "escape" but we want to retain second discrete phase "particles" with in column volume. However, Boundary condition panel in Ansys Fluent do not show "escape" or "wall" or "reflect" for an individual DPM (injection) phase but one for all DPMs.

We want to model the effect of gas bubble induced flow behavior (in one-way coupling) on liquid and solid phases, so we need to keep liquid in batch mode and solid phase (one-time injection). Any suggestion on how to set-up DPM-BC for two discrete phases (two different injections) separately in Ansys Fluent?

How can we optimize the flow & consumption of fuel in Aerospace sector using CFD ?

Detailed suggestions are needed with tutorial link etc.

Dear CFD Researchers,

Since AI tools are currently very popular, I am wondering if anyone use them to choose turbulence model for a CFD case.

So if you did, please share your experiences. Due to the answer, we can extend the boundaries of this discussion.

Thank you for your comments.

Kind regards,

Guven

Hello all,

I'm doing a 2D simulation of flow beneath a partially-submerged rectangular bluff body. The problem's geometry is shown in the figure attached. I'm using a fully structured hexahedral mesh with a y+ of below 1. When the draft (i.e. t in the figure) is small, the model predicts the reattachment length (i.e. Lr) quite accurately. However, it leads to excessive reattachment lengths for cases of large drafts.

As the draft increases, I wonder what turbulent mechanisms/characteristics become influential that SST fails to capture properly. Could it be the increasing curvature of the streamlines or probably turbulence anisotropy? I appreciate any insights.

Regards,

Armin

Hi everyone. I have some problems about evaporation - condensation. (This is very long question :) )

I am trying to simulate water evaporation for different designs but actually I can't.

I'm a relatively newbie in liquid-gas two-phase simulations, especially in Fluent. I'm trying to simulate evaporation and condensation that occurs at every temperature, not boiling. I can't get the right result when I run my models by trying the VOF and mixture models.

The methods I have tried so far and the methods I did not get results:

I defined three phases (air, water-liquid and water-vapor).

1. I created a piecewise - linear graph for saturation temperature by choosing water to steam as the mass transfer. There was no evaporation. Also, if I choose steam to water, I got a negative latent heat error.

2.While researching, I tried creating source terms for cell zone, liquid and vapor. Simply, if t>t_sat was evaporation, if t<t_sat was condensation mechanism, they were udfs. I added define properties(saturation_temp) in udf, but I don't know if I made a mistake in Fluent, I couldn't Run the simulation. (Even if evaporation-condensation started somehow, the mass transfer rates were not correct.)

3. Lastly, I watched solar still videos on YouTube and tried to reproduce it. I tried to do exactly the same things, but what people do either doesn't work or even if the simulation works, evaporation at low temperature doesn't start.

To conclude; I need tips, a step-by-step guide, or the right UDF files to get to the right model. I am open to any help. 😊

If I don't get a result, I will try with OpenFoam, but my priority is Fluent because of my colleagues' Fluent experience and habits. (Another reason is that Fluent seems to be the easiest way for us to transferring this experience to our students when necessary.)

Turbulence structure is described by the formation of eddies. Size of eddies vary based on the turbulence length scale ( (Kolmogrove, Reynold and Hinze). Smaller eddies describes intense turbulence, larger eddies display low turbulence zone.

Recently, I came across a book by Bertin and Cummings, which says there are no physical laws to describe detailed turbulence structure pattern.

Are there any ?

Hi All,

Question: How do I get Mach No contour for CFD Post-processing?

What could I try?

Any help is appreciated.

Thank You,

Alin

I'm currently facing negative cell volume detected error because of weird mesh deformation, ig I might have made mistake in the dynamic mesh setting but I'm unable to pin point my mistake. Also the simulation occurs for a few time steps with ridiculous values of lift and drag coefficients. I'm attaching below the pictures of deformed mesh, lift, drag coefficient curves, my smoothing and remeshing parameters too. Also I'm using a udf to give motion to the cylinder body. The frequency defined in the udf is 1. I had created a total thickness inflation mesh around my cylinder which is of 5 layers, growth rate of 1.2 and maximum thickness of 0.025. Hence, I first separate the cylinder along with the 5 layered inflation from the surface body, then I give the udf motion to both the cylinder and the 5 inflation layers that I had separated. Kindly help me regarding this issue as I've been trying to solve this issue since past 3 weeks, took reference of changing the time step from various articles but still I get the same error.

I'm running a surface reaction. I'm wondering how does ANSYS Fluent calculate the surface deposition rate?

Thank you.

I'm trying to simulate VORTEX INDUCED VIBRATIONS in Flow Around A Rigid Cylinder. The diameter (D) and mass of the cylinder are 0.5 m and 35.78 g, respectively. The spring stiffness is 69.48 kN/m, the damping coefficient is 0.0039 Ns/m and Re=200. I've been trying various UDF since past 2 weeks but everytime m not able to simulate any oscillation so it would be of great help if you kindly provide the UDF.

This research aims to investigate the dynamics and properties of accretion disks around black holes using Computational Fluid Dynamics (CFD) simulations. Accretion disks play a crucial role in astrophysics, and understanding their behavior is essential for studying the physics of black holes and their associated phenomena. The proposed research will employ CFD techniques to model the complex fluid flow within the accretion disk, considering factors such as viscosity, magnetic fields, and relativistic effects near the event horizon. The simulation results will be analyzed to gain insights into the disk's structure, energy transport mechanisms, and radiation emissions. The research findings will contribute to advancing our understanding of black hole accretion processes and their impact on astrophysical phenomena.

Please, are there any CFD software applications that can be used to generate an O-grid mesh for airfoil flow simulations apart from pointwise?

Hi all,

I am currently working on simulating a water jet problem, where water is injected from the left boundary and exits the domain through the right boundary. However, I am facing a challenge of keeping both momentum and mass conservation at the same time. In order to ensure mass conservation in the scenario where a single two-dimensional jet evolves into multiple droplets, it is necessary to enforce the condition that the outlet flux, represented by the product of the cross-sectional area of the droplets (S2) and their velocity (u2), is equal to the inlet flux, represented by the product of the initial cross-sectional area of the jet (S1) and its velocity (u1). Since S2 is typically greater than S1, it follows that u2 must be smaller than u1 to maintain mass conservation. However, this approach alone does not guarantee momentum conservation.

ps: I am using my own code for the simulation and it is an incompressible flow solver， which has been validated in many benchmark cases. The physical properties in my code jump across the interface without any diffusion.

Best regards,

Min Lu

Hi guys

Can anyone model the water spray from the tiny holes in a tank exposed to a hot, high velocity external flame using CFD? What commercial software do you recommend?

thanks

It would make my essay easier to find information about this airfoil.

Thank you anyway!

#naca #aviation #engineering #airfoils #pilots #2430 #information #wind turbine #CFD

I've worked on simulating 3D VIV (a cylinder forced by Karman vortex street) and been stuck with settings. Considering the transition to turbulence due to the oscillation of cylinder (Re≤200, laminar upstream), "SST k-ω coupling with intermittency transition" is my scheme now.

However, I'm not sure whether I should enable "intermittency transition" since I don't fully understand the statement given in the user's guide, which says "The Transition SST model is not Galilean invariant and should therefore not be applied to surfaces that move relative to the coordinate system for which the velocity field is computed; for such cases the Intermittency Transition model should be used instead."

I don't understand the bold sentence in the statement especially. Does it mean the moving surface of cylinder (in my case)? Hope anyone can provide any guideline. Thank you so much.

Hello everyone!

I'm doing CFD propeller marine. But I am errored. Can you help me?

Error at Node 0: floating point exception

Error at Node 1: floating point exception

Error at Node 2: floating point exception

Error at Node 3: floating point exception

Error at Node 4: floating point exception

Error at Node 5: floating point exception

===============Message from the Cortex Process================================

Compute processes interrupted. Processing can be resumed.

==============================================================================

Error: floating point exception

Error Object: #f

Registering ReportDefFiles, ("C:\kp_files\dp0\FFF\Fluent\.\report-file-0.out")

Calculation complete.

So I usually use the COMSOL Multiphysics software to deal with hemodynamic problems with the CFD module of the software. i found OpenFoam as a opensource software but i don't have ideas about it.

is OpenFoam suitable for scientific reasearche in matter of medical field?

what are the major difference between Ansys Fluent, Comsol Multiphysics and OpenFoam?

are the results obtained with OpenFoam equivalent to thous obtained with the commercial software ?

thanks in advance

Hello,

I wish to solve the weak N-S eq (based on PDE's) in COMSOL.

By default, COMSOL applies the equation in all domain but I want to solve it in each mesh (not all domain).

I want to do this in order to add different porosity in each CFD mesh.

Is it possible?

Thanks in advance.

Which software is mostly preferred in research for preprocessing during CFD analysis and gives best results?

G'day,

I'm working on simulating 3D Karman Vortex Street, confused how to distinguish laminar and turbulence. As the pic shows (this is a frame before starting oscillation), there is downwash near the top of the cylinder, but the Reynolds number in this case doesn't exceed 200, namely, it should be laminar. ----------------update--->

Yesterday, I asked Perplexity Ai to find some info and it provided some reference talking about the wake structure due to the end of finite cylinder. So now I just want to collect your suggestions, since it seems no reference directly indicates "the downwash is inherent no matter the flow is laminar or turbulent". So if anyone knows relevant paper, please let me know, thank you so much!

[1]

Hello, I'm COMSOL newbie working on cluster computing.

I'm using distributed parametric sweep mode, in which each server node solves a single problem (ex. Node 1 - Inlet fluid velocity 1m/s, Node 2 - Inlet fluid velocity 2m/s, etc...)

In distributed parametric sweep mode,

(1) I want to monitor multiple convergence plots since multiple problems are solved simultaneously. But I can only monitor a single convergence plot... Please give me some advice.

(2) How can I monitor multiple probe plots (which are Probe parameter - Iteration number plots)? I can only monitor the accumulated probe table, which is the final converged value. But I want to check the probe value trend with respect to iteration number.

Thank you very much!!! Every comment is my energy please help me!

I am working on BioFluid Dynamics. Ansys CFD Post doesn't have an option to plot AWSS and OSI directly like wall shear stress(WSS). So I am unable to plot this. In hemodynamics, it is an important parameter. Can anyone help how to add AWSS and OSI in CFD Post?

Hi, I am modeling sloshing of a fluid inside a container using ANSYS Fluent. For this, I am using VOF model. I consider two accelerations. a_x=2 m/s² and a_z=9.81 m/s². The modeling continues for 2 sec. Physically, a container with a liquid inside under constant acceleration will have a steady inclined surface after sometime. However, the surface that I get fluctuates. As it is shown in the image (it is just an indication), the flat surface reaches to max inclination until t=0.8 sec. After that the surface starts returning to its original flat horizontal condition and again starting going up. Could you please let me know what is the problem? Thank you,

I'm doing a 2D transient simulation of turbulent flow past a sharp-edged rectangular bluff body. I'm using a version of k-omega model for turbulence modeling. My objective is the time-averaged flow field; I'm not interested in the instantaneous field. The problem is that I'm not sure how to choose an efficient timestep size.

In the literature, i have come across two methods in this regard:

(a) The timestep is adjusted repeatedly to ensure a max/rms Courant number, or

(b) based on an estimation of the Strouhal number (associated with the dominant frequency of vortex shedding) known from previous experiences, the timestep size is determined such as to resolve each vortex shedding cycle through n temporal increments.

The problem is that, i couldn't find any consensus in the literature over appropriate values for max/rms Courant number or n in either methods. I understand that this should be really addressed through a sensitivity study, but I have very limited computational power to afford that, especially given that it takes a long time for the simulations to reach a statistically converged solution. I am hoping there's some reliable rule of thumb or related studies to use, thereby avoiding such a sensitivity analysis.

I would appreciate your comments.

i have a problem with reaction modeling by Ansys Fluent. when i change the Arrenius parameters, nothing change and it makes me confuse. can anybody help me?

Dear ALL,

What project do you recommend for CFDer not software engineer to improve coding skills, Is leetcode a good choice?

In addition, what strategy do you recommend for reading others multiple-file code? Debug and see the stack?

Thanks in advance.

My combustion simulation is fully converged. The mass imbalance that Fluent calculates is 3×10^-9. However, if i calculate the mass imbalance from the average surface integrals, as (density×velocity)_avg xArea is the mass flux from inlet and outlet, the resulting mass imbalance is the order of 10^-7. What causes this deviation as its order is around 10^2. Thank you.

Hello everyone.

I have used the CCD to build a table of 25 runs (4 variables and 2 responses). The responses have been calculated based on CFD results. Thus the center point can not be repeated due to constant values in results.

to enhance the precision of models, 15 more runs have been added to the table of experiments (totally 40 runs).

Finally, after analysis, the R2 (adjusted and predicted) is more than 95% for both of responses.

But, I cant see any data about "lack of fit" in ANOVA and fit summery sections.

Could you help me about this problem?

Is it necessary to report the "lack of fit" data in this analysis?

It should be noted that the model predict the responses well.

I am working on wind interference effect in Ansys CFD. Even though the domain and geometry are symmetric, the results in the contour in CFX are not symmetric on Leeward side of the wind but the contour is symmetric in windward side of the building what could be the reason, what are the parameters should I check on?

I need a collaborator with experience in code development and, if possible, numerical analysis too.

I am currently developing an open source code in python that can be used to solve different kinds of Integral Equations.

In the last one and half years I have done some work in numerical Algorithms for integral equations with some papers already published. It has culminated to several python codes which I have used to produce the results.

The codes are all private but the results are published so I am inclined to make these codes publicly available so that others can use them at no cost and minimum effort.

I, therefore, need a fellow Researcher who has good skills in software engineering and numerical analysis to join me in this line.

A minimum requirement is the knowledge of git and python.

You can email me at [email protected]

Hello everyone.

I am doing a CFD simulation for waste heat boiler and simulation boiling in steam drum.

I used ANSYS Fluent to perform the simulation. However during the calculation section. I received this error

"reversed flow in 274 faces on pressure-outlet 18.

Stabilizing mp-x-momentum to enhance linear solver robustness.

Stabilizing k to enhance linear solver robustness.

Divergence detected in AMG solver: k temperature limited to 1.000000e+00 in 8519 cells on zone 2 in domain 3

turbulent viscosity limited to viscosity ratio of 1.000000e+05 in 18252 cells

Divergence detected in AMG solver: k

Error at host: floating point exception

Error at Node 0: floating point exception

Error at Node 1: floating point exception

Error: floating point exception

Error Object: #f"

May I know what are the possible cause of this error?

Any advice? Thank you

Hello, I want to publish a research paper on the assessment of CFD turbulence models, I need an airfoil geometry to start with. where can I find airfoil geometry or profile? and its experimental data, cd and cl values or cl alpha values? I have searched for experimental data, but unfortunately, I did not find any.

Can you please advise me about where to find airfoil geometry and its experimental data?

what type of airfoil should I choose for the assessment of different turbulence models?

I have used in my current paper the pressure-based solver for air distribution inside the control room, I would like to ask which governing equation determines the velocity field and also the continuity and momentum equations for which purpose work in CFD calculation if there is an explanation

Currently, I have completed my research study for CFD simulations, using Ansys Fluent as a student in Anglia Ruskin. Now, I cannot use any Ansys licence through university as I am not anymore a student.

So, I am looking for a good software CFD free of charge to be used for research without restrictions (Student or Educator Version). Ansys Fluent Student version has some limitations and cannot be used for research with refined mesh study. I am looking for alternatives, finding a free good CFD software.

I have looked on Autodesk website and found the Autodesk CFD Ultimate as a Student or Educator version, it is free of charge.

This software can be used for computational fluid dynamics simulations and solid body motion analysis.

However, I have not used this software for CFD and I do not know if there are any limitations.

Is anyone using this software for CFD simulation please let me know if this software has any limitations?

Any idea, suggestion or thought, I would much appreciate.

Thank you.

I wanted to couple two simulations that are calculated separately like in-cylinder combustion and cylinder block, in your opinion can we map boundary conditions of the cylinder with block surface transiently? If yes which tool do you suggest?

I am interested to the discontinuous Galerkin in my research, but after my modest research, I found two type of it Modal DG and Nodal DG. I wanted to know :

how effective is each one of them ?

What is the difference between them ?

Hello all,

I've got a 2D simulation case in which the flow separates from the sharp leading edge of rectangular bluff body and reattaches to the wall some distance downstream. The main goal is a accurate prediction of pressure distribution along the body's face parallel to the flow.

I'm doing a transient simulation using SST model in conjunction with gamma-Re transition model. The time- cord-averaged y+ is less than 2~3 and the inflation layer around the face of interest contains 10 prism layers. The Re number based on the body's width (perpendicular to the flow) is 1.7e+4.

The problem is that my model overpredicts the reattachment length, which in turn leads to delayed pressure recovery.

I have a suspicion that longitudinal decay of turbulence values specified at the inlet might be to blame. Consulting the Ansys CFX-solver Modeling Guide, I learnt that one solution is to prescribe appropriate turbulence values at the inlet based on the desired values at the body. An alternative approach also suggests some additional source terms for k and w transport equations in order to preserve the inlet values up to some distance upstream the body, from where decay is allowed.

Here are my questions:

1- Is my suspicion valid in the case of my problem?

2- Is the decay of turbulence of physical basis or a numerical artifact?

3- which of the two methods works better? Are there any attempts in the literature?

I appreciate your comments.

I'm currently using fluent simulating the cold flow of a supersonic combustion chamber with using hydrogen as fuel. But I can't make it to start the first iteration. Can someone give me some advice?

These are the error respond from fluent

WARNING: Invalid cp (0.000000e+00 J/kgK) for oxygen at temperature -1.#IND00 K

reversed flow in 500 faces on pressure-outlet 8.

Stabilizing nut to enhance linear solver robustness.

Divergence detected in AMG solver: nut

WARNING: Invalid cp (0.000000e+00 J/kgK) for hydrogen-air at temperature -1.#IND00 K

Divergence detected in AMG solver: nut

WARNING: Invalid cp (0.000000e+00 J/kgK) for hydrogen-air at temperature -1.#IND00 K

Divergence detected in AMG solver: nut

WARNING: Invalid cp (0.000000e+00 J/kgK) for hydrogen-air at temperature -1.#IND00 K

Divergence detected in AMG solver: nut

WARNING: Invalid cp (0.000000e+00 J/kgK) for hydrogen-air at temperature -1.#IND00 K

Divergence detected in AMG solver: nut

WARNING: Invalid cp (0.000000e+00 J/kgK) for hydrogen-air at temperature -1.#IND00 K

Divergence detected in AMG solver: nut

WARNING: Invalid cp (0.000000e+00 J/kgK) for hydrogen-air at temperature -1.#IND00 K

Divergence detected in AMG solver: nut

WARNING: Invalid cp (0.000000e+00 J/kgK) for hydrogen-air at temperature -1.#IND00 K

Divergence detected in AMG solver: nut

WARNING: Invalid cp (0.000000e+00 J/kgK) for hydrogen-air at temperature -1.#IND00 K

Divergence detected in AMG solver: nut

WARNING: Invalid cp (0.000000e+00 J/kgK) for hydrogen-air at temperature -1.#IND00 K

Divergence detected in AMG solver: nut

Error at host: floating point exception

Error at Node 0: floating point exception

Error at Node 1: floating point exception

Error at Node 2: floating point exception

Error at Node 3: floating point exception

Error at Node 4: floating point exception

Error at Node 5: floating point exception

Error at Node 6: floating point exception

Error at Node 7: floating point exception

Error at Node 8: floating point exception

Error at Node 9: floating point exception

Consider a pipe in a cylindrical coordinate. The flow is a fully developed laminar flow and the temperature field is T(x, r, θ). At a certain cross section x_0, the mean temperature is T_m and the mean velocity is u_m (To calculate T_m or u_m, integrate the temperature or velocity over the whole cross section and divide the result by the area). A conjecture is that u*partial T/partial x=u_m*dT/dx, where partial denotes the partial derivative. This seems reasonable since the heat transfer should be easier where fluid flows faster, which leads to a smaller gradient. But how can I prove it strictly?

I am aware that several definitions of stream-function in 3D flow do exist in the literature. But I am at a loss on what stream-function physically means in 3D.

Hi all,

I am going to use either Ansys CFX or Ansys Fluent solver to simulate flow past a semi-submerged rectangular cylinder, as shown in Fig. 1. The main goal here is to achieve an accurate prediction of pressure distribution over the cylinder, which seems to be particularly reliant on the capability of the CFD model to predict the separation and the reattachment of the flow correctly. I would appreciate any tips regarding the following questions.

Q1. How important is the choice of the turbulence model? Which models are superior? Could the flow be modelled as being laminar at all?

Q2. How important is the approach to the free surface? Could the free surface be modelled as a free-slip wall to reduce computational costs? Is it necessary to precisely track the free surface using the VOF model for this study?

Q3. What are the most appropriate boundary conditions for the simulation?

Q4. Which is more suitable for this problem? CFX or Fluent?

The number of papers published every year on medical CFD has increased by 10 times in the last 20 years (see attached figure).

We all know that there are many papers are just out there for the sake of publication or to get points for a funding program. However, CFD has indeed become quite a mature technology for medical simulation applications in many areas such as vascular and respiratory medicine.

What are the genuine drivers power this increase in the number of medical CFD papers?

Would love to read your thoughts.

I am working on modeling of Direct Contact Membrane Distillation system with COMSOL Multiphysics software. I have some problems with the mass transfer part of the simulation because I don't know which boundary condition is appropriate for boundaries between membrane-cold and membrane-hot channels. Is anybody knows how I can solve this problem?

I have just simulated a steady state 2D methane-air combustion case with species transport model and eddy-dissipation turbulence-reaction model with the use of Fluent code. The solution is converged and the results match very well with the experimental data. For some purpose, I need to show manually that the transport equation for N2 equals zero since it is an inert gas and does not participate in the reactions. I got all the necessary flow field variables from the CFD solution such as density, axial and radial velocity, and N2 mass fraction. For some points in the flow field, I am trying to solve manually the same turbulent transport equation for N2, which Fluent code solves, with the central finite difference method using the flow field variables from CFD. My step size in the flow field is 0.0002m for the finite difference method. However, my finite difference method results are far from 0, which means as if N2 gas participates in the reactions. I find the reaction rate term (Rn2) for N2 around 5 kg/m3.s, as it must be very close to 0 for the numerical solution.

Where do you think I am wrong or missing something? Thank you very much.

I'm now studying the transportiveness of discretisation schemes using An introduction to Computational Fluid Dynamics: The finite Volume Method; HK Versteeg & W Malalasekera; 1995. In Page 144 it says "Since there is no diffusion φ_P is equal to φ_W. If the flow is in the negative x-direction we would find that φ_P is equal to φ_E" (The x-direction is W -> P -> E, and there is only convection and no diffusion, so Peclet number = inf. See the appended figure for details.).

This really confuses me, does it mean if the flow is in the x-direction, then φ_P is unequal to φ_E and they are equal only if the flow is in the negative x-direction?? I cannot understand it because if the flow is in the x-direction, E will be downstream of P and will only be influenced by P, so φ_E is equal to φ_P, which follows that φ_W=φ_P=φ_E.

Do I make a mistake? Please help me, thanks a lot!

Dear all,

Does anyone know how to edit the geometry of the dambreak.msh file provided by ANSYS Fluent?

Or does anybody have a project file for the dambreak case without the statement that the mesh file is imported and therefore not editable.

We want to add a geometry in the downstream region as an obstruction.

Hi there,

I'm working 2D axisymmetrical pipe groove (CFD) model with constant wall temperature. During the hybrid initialization, I have a non-positive element error along with a floating point exception. I believe I'm missing something. Any suggestions would be really helpful?

Thank you

Hi, I have been trying to import the thermophysical properties of CO2 (for the supercritical region) to Fluent by using UDF (only for 80 bar). End of the analysis, the result is not correct. I think Fluent is pulling a Tref value which causes my enthalpy to go negative; therefore, the energy equation does not give accurate temperature results, so the result of thermophysical properties, which are functions of temperature, are wrong. I want to be grateful if you could share your recommendations, especially on the methods of correct calculation of enthalpy. My specific heat code is as follows:

#include "udf.h"

DEFINE_SPECIFIC_HEAT(supercritical_cp, T, Tref, h, yi)

{

real cp;

if (300. >= T)

{

cp = 2.256999118844760E-02 * pow(T, 4.) - 2.606618818487370E+01 * pow(T, 3.) + 1.129129898467780E+04 * pow(T, 2.) - 2.174235523392760E+06 * T + 1.570278261279480E+08;

}

if (306. >= T > 300.)

{

cp = 1.147018354719330E+01 * pow(T, 4.) - 1.385519872072390E+04 * pow(T, 3.) + 6.276090044658930E+06 * pow(T, 2.) - 1.263530241490370E+09 * T + 9.539285116131690E+10;

}

if (307.8 >= T > 306.)

{

cp = -1.611612838918710E+04 * pow(T, 4.) + 1.978085636985280E+07 * pow(T, 3.) - 9.104578256081260E+09 * pow(T, 2.) + 1.862478067464320E+12 * T - 1.428738938943020E+14;

}

if (310. >= T > 307.8)

{

cp = -1.576135476848110E+03 * pow(T, 4.) + 1.945866130859760E+06 * pow(T, 3.) - 9.008641337104630E+08 * pow(T, 2.) + 1.853611584329810E+11 * T - 1.430227164258690E+13;

}

if (320. >= T > 310.)

{

cp = 1.770529718817220E+00 * pow(T, 4.) - 2.242833202341920E+03 * pow(T, 3.) + 1.065432793165650E+06 * pow(T, 2.) - 2.249454519629040E+08 * T + 1.781003821884030E+10;

}

if (T > 320.)

{

cp = 2.086961941877520E-03 * pow(T, 4.) - 2.851496372954040E+00 * pow(T, 3.) + 1.461548345109060E+03 * pow(T, 2.) - 3.330816449664890E+05 * T + 2.848088978782360E+07;

}

*h = cp*(T-Tref);

return cp;

}

How can I add FLOTRAN CFD in ANSYS together with the elements 2D FLOTRAN 141 and 3D FLOTRAN 142?

I want to develop sediment transport 2D model and flow 2D model on floodplain river. What is te best software package for it? How can I get them? What is process of acquiring CCHE2D. FLOW 2D, SRH 2D, Delft 3D, Ansys or CFD kind software with full version?