Fluid Turbulence - Science topic

I am running a coarse DNS case for pipe flow with 2.1 Million cells. My residuals are quite fluctuating as its a fully turbulent annular pipe flow case but its getting statistically converged to a mean value.

My doubt is, the residual values are quite high where its mean is getting converged for instant close to 0.1 or 0.01(refer attached .png), despite of giving tolerance of 1e-06. Due to this I think I have results of velocity profiles and shear stresses quite under predicted.

what can be the possible ways to reduce these residual values?? and what is the reason of having such high residuals??

NOTE: I am already using higher order schemes for solving Fluid flow equations in OpenFOAM

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 have a fully developed pipe flow in with Inner radius (r) and outer radius (R), using pressure driven flow condition due to buoyancy,

- (1/rho) dP/dx = g

and velocity scaling u* = sqrt ( (-R/ (2 rho)) * (dP/dx)) [ friction velocity ], if Reynolds number is fixed ( Re = 600 ), along with r and R

we can get y+ value based on y values we give for cell size at both the walls,

But the question is when y+ calculated from this formula is y+ at the outer wall ( general pipe flow condition ) but how to get a y= for the inner annulus? ( concentric annular pipe flow ) ?

is there any analytical method to find this y+ or the only solution is to get us after simulation run when we have calculated friction velocities wall shear stresses at wall cell centers.

basically two different y+ to get analytically, in order to set up minimum cell size for my LES grid.

Hello,

I am running a 3D density based transient problem with Spalart-Allmaras turbulence model. After few iterations I am getting Negative Nut in xxx cells. I tried to search the cause, but not successful in finding answers. Could anybody help me in understanding from where this problem arises?

In case of turbulent flow over an irregular shape, how can we find the first layer thickness near the walls of the irregular boundary.?

For flow over a flat plate, the skin friction co-efficient can be found by the direct formula, for irregular shapes how can we find the skin friction coefficient.

Hi all,

I'm planning to simulate flow past a floating body using CFD method with the main purpose of investigating its stability against hydrodynamic forces. A sketch of the problem is presented the the figure attached.

It seems that an accurate estimation of pressure field, and therefore hydrodynamic forces, is heavily dependent on correct prediction of flow topology, particularly separation and reattachment of the flow.

I'm wondering what turbulent models would best handle this problem. I would appreciate it if you provide details and specific reasoning.

Regards,

Armin

In 2010, Dr. Khmelnik has found the suitable method of resolving of the Navier-Stokes equations and published his results in a book. In 2021, already the sixth edition of his book was released that is attached to this question for downloading. Here it is worce to mention that the Clay Mathematics Institute has included this problem of resolving of the Navier-Stokes equations in the list of seven important millennium problems. Why the Navier-Stokes equations are very important?

The existence of turbulence in astrophysical fluids has been living as a well-known unsolved problem for a couple of decades. The range transitions among three distinct scales of the micro-fluidic kinetics are still lying obscure. In this context, is there any equation of state to describe turbulent fluid media in the fabric of the modified (by turbulence) macroscopic Navier-Stokes equation?

Hi,

I am trying to find about conditions related to relaminarization of fluid flow both in liquid and gases. In this regard, I would like to know if there are any theoretical or empirical conditions to be satisfied for relaminarization. If possible please attach the link to the references in your answers.

Thank You

Dear All,

We are using Autodesk cfd to calculate the air flow in a single sided naturally ventilated room, where air flow depends only on buoyancy due to temperature difference between inside and outside.

in the first case -as in attached pictures- the flow seems reasonable: hot air rises up, while cold air flows near to floor.

in the second case we tried to add a scalar representing Co2, the results shown in second picture can never be true, cold air rises up !!

I have been stuck in this for 2 months now (Hence: the air material in both cases is variable so that buoyancy takes place)

is Inertial Subrange can identify at any depth in the flow,

kindly suggest me.

Thank you.

Assuming two-dimensional flow, I'm trying to implement a code for a quasi DNS problem. There is no doubt that i should replace the velocities in Navier-Stokes equation (NSE) with its equivalent velocities in the form of u=u_base + uf, where u_base is the steady state solution for NSE for the channel flow and uf is the perturbed velocity which varies with time. By the new definition of the velocities, one can get new extra terms in NSE, as a function of u_base and uf. My point is to find the fluctuation behavior with time so that i can calculate the growth rate and start an instability study for the flow field.

There are two forms of the new NSE, maybe more but at least these two were the common equations which i found in most of the researches which i read. The first one is in the form of du/dt + div(uu) = - dp/dx+ iv(grad(u))/Re where u was defined earlier. The second form is: du/dt + div(uu) = - dp/dx+ div(grad(u))/Re - omega x u.

Which one should i use for the purpose of instability studies?

Note: You can check the attached files for the non-conservative form of the two previous models.

Hello

I hope some of you use Comsol.

So, I’m doing simulation of airflow through porous media. According to the characteristics of the flow and the geometry, the flow is turbulent, so after doing some simulations, the physics chosen was k-epsilon Low-Re, because of the wall treatment (Low Re). Geometry is a 2D rectangle representing a long tube with the porous bed in the middle of it.

I included a volume force in order to simulate porous media. At flow conditions, the equation that represent it is Ergun Equation (that could be expressed as Forchheimer eq). My implementation was as simple as :

Fx= -(spf.mu*u)/(ka )- b*spf.rho*u*sqrt((u^2)+(v^2))

Fy=-(spf.mu*v)/(ka)-b*spf.rho*v*sqrt((u^2)+(v^2))

So, when I defined a horizontal cut line in the middle of the bed in order to see velocity profile, I found a problem because a velocity peak is shown at each side, near to the wall (fig 2 and fig 3). I think profile should be just horizontal. Since properties are isotropic, there no reason for this, so, there is something wrong.

I think it is not a boundary layer problem because I simulated it when having a fully developed flow and with 12 boundary layers (as default), shown at fig 1. Minimum mesh quality is 0.45 (in fluent it would be 0.55 as maximum skewness), which I think is not so bad. I tried also using laminar flow conditions and Fluid and Matrix Properties as default which uses Darcy's Law, but the profile I found was similar.

You can see the images as attachments below.

Has anyone had this kind of problem? Am I missing something? How could I overcome this problem??

Any help is really appreciated, since I've been stuck there several days.

Thanks in advance,

Jhoan G.

As shown in the figure attached, the turbulent fluid flow around a floating (i.e. partially submerged) cylinder with a rectangular cross-section is to be modelled. Clearly, the forces and torques acting on the cylinder are oscillatory due to the vortex shedding phenomenon. The ultimate goal is to study the cylinder stability on the surface. The question is then how to characterize the fluid forces. Particularly, I have no idea if such a stability analysis could be based on the time-averaged forces or the instantaneous values may play a dominant role. How to characterize and quantify the significance of time-averaged values versus instantaneous values in this specific application?

I would appreciate any comment.

What causes turbulence in fluids. Why does a relatively high velocity cause turbulence? What causes the particles to not follow laminar flow at high Reynolds number? Thanks in advance for the answers.

Hi,

I was trying to run a simple duct flow case with laminar and inviscid flow. 

The objective was to check the outlet velocity if I have imposed a parabolic velocity profile at the inlet depicting a fully developed flow. 

For laminar flow, it works well with a minor difference in the outlet velocity compared to the inlet which is due to discretization errors. 

Now, when I am giving a fully developed profile and selecting inviscid model , I should get the same outlet profile, but continuity dilation keeps on oscillating and I get flow reversal errors at some faces of the outlet and the no of faces with reversal flow also keeps on oscillating. 

The outlet velocity profile does not look anything like parabolic. 

The grid is rectangular elements only so I am sure that the issue is not with grid. 

Boundary condition : Inlet: udf for parabolic developed profile. 

Outlet : Pressure outlet. 

I'm trying to find an empirical correlation for obtaining the critical Reynolds number in a 180-degree bend/curve/elbow with a circular cross-section (See the figure). I've searched for a lot of papers with no luck, but I know I’ve seen it somewhere before.

Can you recommend an article or book with a correlation for the critical Reynolds number as a function of tube diameter and curve radius?

Thank you in advance.

Guys,

I'm reading this Article:

You can download it from here:

Please take a look at Fig. 8. I'm trying to export the exprimental data of the velocity profiles in this graph. Both X and Y axises are normalized(non-dimensionalized). Y is normalized by H and there is no problem to me(just need to multiply it by H) but X axis is strange to me. Anybody could explain it to me that how is this graph drawed? I just want to export the exprimental data but don't know how to dimensionalize the X axis.

I am attempting to implement an atmospheric boundary layer in a 3D fluent simulation using a UDF. The UDF is proven to work in a 2D simulation, however approximately 10 hours into a transient 3D solution using 16 cores, fluent crashes with the only log give being a segmentation fault. I am suspicious that something is going wrong with the parallel nature of the solver, however I can't see any issues with my code - nor can I see any issues that would result in a crash 10 hours in, where the code is not time dependent.

I am in search of latest topics in the field of Open Channel Flow and Hydraulics for my research work. Your suggestions in this regard shall be highly appreciated.

Now I am reading your paper. That is "Sedimentation Scour Model" Gengsheng Wei, James Brethour, Markus Grünzner and Jeff Burnham.2014

and in Scour at cylindrical piers part like attached Figure. Please help me to explain about boundary condition such as: Xmin. Xmax, Ymin Ymax, Zmin, Zmax in FLow 3D Program

Thanks you so much

The farther backward we can look, the farther forward we can see. On 10th death anniversary of John Maddox, we retrospect his four-combo interrogations upon quantitative dimension of molecular biology^1-4, which invoked intensive arguments^5-12. Early 1990s, British biologist John Maddox (1925 – 2009), an editor of Nature from 1966 to 1973 and 1980 to 1995, published the four-combo editorials to scold molecular biologists for their naïve reliance on qualitative observations.

In January 1992, John Maddox doubted that “Is molecular biology yet a science?”:

“The great successes of the past few years suggest that living processes consist of well-ordered events executed under strict control, but a few numbers would give a different and more fuzzy impression.”¹

In May 1993, John Maddox chastised that:

“The general enthusiasm for the fortieth anniversary of the structure of DNA should not blind us to a few blemishes on the otherwise fair face of molecular biology.”²

In August 1993, he iterated that:

“Complaints that molecular biology is indifferent to quantitative considerations are belied by brave efforts to understand the hydration of glucose.”³

In March 1994, he punctuated that:

“Even the best known systems in biology may not be as well-understood as is generally believed, which means that understanding is incomplete and may even be misplaced.”⁴

In retrospect, Lee Segel and John J. Tyson defended those Maddox’s assertions by rapid correspondence.⁵ In 1993, M. F. Shlesinger and his colleagues reponsed to Maddox’s assertions and proposed that:

“The once abstract notions of fractal space and time now appear naturally and inevitably in chaotic dynamical systems and lead to ‘strange kinetics’ and anomalous transport properties. An understanding of this kind of dynamical behaviour should provide insights into, for example, turbulent fluid dynamics and particle random-walk processes.”⁶

This critical review invoked intensive arguments upon the philosophy of molecular biology among scientists (see Figure). Douglas Braaten argued against John Maddox’s reductionism – broadly understood as a belief in explanation of biological phenomena wholly in physical and chemical terms.⁷ Robert Lortle valued the nonlinear quantification effort in the open dynamic systems – molecular biology field, but he also argued against the reductionist approach in molecular biology.⁸ By coincidence, Philip R. Steinmetz endorsed that molecular biologists should look as biological systems as complex, nonlinear processes rather than complex, linear processes. He enlisted “Three Sins” of molecular biology – “the shortcomings of molecular biology referred to may be side-effects of rapid progress”:⁹

However, Denys N. Wheatley dismissed Maddox’s assertions and Segel and Tyson’s echo.¹⁰ In 1999, Douglas B. Kell agreed with Maddox’s assertions, but refuted his argument ground.¹¹

1. Maddox, J. Is molecular biology yet a science? Nature 355, 201–201 (1992).

2. Maddox, J. The dark side of molecular biology. Nature 363, 13–13 (1993).

3. Maddox, J. Putting molecular biology into water. Nature 364, 669–669 (1993).

4. Maddox, J. Towards more measurement in biology. Nature 368, 95–95 (1994).

5. Segel, L. & Tyson, J. J. Law of mass action. Nature 357, 106–106 (1992).

6. Shlesinger, M. F., Zaslavsky, G. M. & Klafter, J. Strange kinetics. Nature 363, 31–37 (1993).

7. Braaten, D. Physics of life. Nature 364, 477–477 (1993).

8. Steinmetz, P. R. A strange discipline. Nature 364, 753–753 (1993).

9. Lortie, R. Molecular biology. Nature 364, 753–753 (1993).

10. Wheatley, D. N. Water in life. Nature 366, 308–308 (1993).

11. Kell, D. B. Revolutionary ideas come round again. Nature 397, 644 (1999).

12. Spiller, D. G., Wood, C. D., Rand, D. A. & White, M. R. H. Measurement of single-cell dynamics. Nature 465, 736–745 (2010).

why lateral-force fluctuations frequency is half longitudinal-force fluctuations in turbulence flow or wind simulation?

I need to compare results for a simulation with nanofluids in a coiled tube using the single-phase and two-phase methods in FLUENT. The single-phase method works perfectly well. However, I am confused with the development of the two-phase Mixture model. The pertinent literature states that the fluid properties need to be defined through the use of the effective properties (as in the single-phase simulations). Hence, I am not sure on how to define the model in FLUENT. 

Therefore, if we define the mixture with two phases, my understanding is that the base fluid properties (water) will be defined with the UDFs for the effective nanofluid properties. Then how can we define the properties of the nanoparticles i.e. the second phase? 

Am I right to assume that for the second phase, the granular option needs to be activated?  

Is there any population distribution law (particle concentration) derived in the presence of fluid turbulence effects in the literature?

Your active participation alongside references is anticipated.

I need your expert suggestions. What is the simplest way to incorporate fluid turbulence in the fluid momentum equation?

Common vortex identification criteria like lambda -2 and Q-criteria identify vortex by considering a vortex core as a region of local pressure minima. I understand that the reason a vortex core is considered to be a region of local pressure minima is because the pressure gradient is able to provide the required centripetal force for the fluid to rotate. Is it necessary that vortex core have a local pressure minima? If yes, are anticyclones not vortices even though they have a rotating mass of air?

best (achievable) x-reg for singular solutions of nd incompressible euler equations by Székelyhidi jr.+de lellis and remarks/links with pressureless (regular) sols ?

--on solutions "convex optimized" by l. Székelyhidi jr. and c de lellis (and descendants) of nd (or 2d, 3d if there is a difference for LS-CD frames and if so, then how ?) incompressible euler equations, could anyone tell what is the best x-reg (so better than C ^ s, s about 1/3) that LSjr+CD or their descendants, 1 / have already, 2 / could be expected and what are the obstacles getting (or not) a better x-reg ? (all with the same question in t-reg and mixed (x, t) -regs), this regardless of their first motivations, that is, breaking the uniqueness (more generally realistic up to C ^ s for all s <1) and onsager (related to s=1/3) or the final answer is close to s=1/3 (and then why)?

--Can anyone confirm that the solutions coming from the classical and regular theory (that is to say not LS&CD theory), which are the particular solutions that are  pressureless (see in 2 / 3d, majda diperna pl lions but also in Rn, all n, in my jmpa95 and thesis92-ch3/90 etc, quoted by yudovich) are out of reach by LSjr+CD theory and their singular frameworks (because for LSjr+CD, the pressure is, at first, basically + - equal to cst lul ^ 2 and then p = 0 gives u = 0 or all of this can be (partly) overcome and how, for example, include these sols in LS&CD setting?) = What could be said about these sols regarding the LS&CD theory ?

--regular theory =eg, for a (short) interval of non nul times, +- Du or rotu (:Du-tDu) are in x, C^s (s in ]0,1[), C^o, or L°° or bmo or (eg in 2d:) Du or rotu in Lp(loc), for one p in [1,oo] etc

--LS&CD singular theory at the state (?)= eg , for a (short) interval of non nul times, u itself (and not its gradient Du) is only in C^s, s in ]0,1[ or s<1/3 or s= 1/3 etc and eg nothing (yet?) on (atleast L1(loc)/loc-measures) DERIVATIVES of u, eg Du or rot(u) obtained (and conserved) for non nul times (even if it is supposed at t=0) etc

-- Are there (zones of) junctions or intersections (and where and what) between regular theory and LS&CD singular theory, existing same common sols (even particular) for these 2 theories ?

--eg 2d, or 3daxi, rotu (and/or its moments 0,1,2 and/or the axi structure) could be? (even "half" or "under" or partly in some way) conserved in SDth as in regTh etc ?.

--questions extended of course to all flu mech models already treated (by extensions from IncEE) by SDth. 6/7/18

Ribs are used on rectangular channel surfaces to promote turbulence and enhance thermal performance. Different shapes has been performed such as rectangular rib, triangular rib, V- ribs , inverted V-ribs, broken V- ribs, semicircular V- ribs. These ribs increase the level of mixing of the cooler core air with the warmer air close to the channel sidewalls and restart the boundary layer after flow reattachment between ribs resulting in enhanced convective heat transfer coefficients. What is the permissible range of the ratio of rib height to channel hydraulic diameter (e/Dh)?

Dear all researchers,

Can we create a turbulent flow with the following conditions?

1- isotrope & non-homogeneous

or

2- an-isotrope & homogeneous

Regards,

Dear Sir,

I am doing Haemodynamic analysis study  in idealised curved artery model. I carried numerical simulation in ANSYS FLUENT using Newtonian fluid. In CFD-POST am not able to get contours and results for Time-Averaged Wall Shear Stress (TAWSS) and Oscillatory Shear Index (OSI). Kindly explain how to get the results and contours for TAWSS and OSI in CFD-POST? Provide some tutorials, procedures, materials on it.

I am studying external aerodynamics on Ground vehicles. And I am using Star CCM+ for my simulation. To achieve better drag and lift results, I got to know that Y+ plays an important role. So how can I specify the first wall distance in Star CCM+? 

I tried to change the number of prism layers, prism layer stretching, Prism layer total thickness. But I didn't find any improvement. 

I am trying to model the viscosity of polymer using Carreau model. However, the options of Carreau model is not available for turbulent flow. So I am trying to write UDF in order to model the viscosity of polymer using Carreau model in turbulent flow. Anyone can help?

Hi,

Let us consider a large ensemble of vortices immersed in a flow domain. The Helmholtz free energy is given by F=E-Tk

F=E-Tk ln N,

where k is the Boltzman constant, and W is the number of the possible positions of a vortex.

However, the scale of turbulent vortices is much larger than the scale of molecular motion on which the measure of entropy and therefore the Helmholtz free energy is based. In particular, in incompressible Navier-Stokes turbulence, the temperature T is decoupled from the nonlinear dynamics underlying turbulence.

My question is whether it is possible to replace the temperature T by Reynolds number Re in defining the free energy for macroscopic vortices interactions in turbulence.

Thank you.

Wang Zhe

Coherent structures identification.

I am using cross -Hot wire anemometer for characterizing the flow field of a free round jet. Is there any way to identify the coherent structures from the radial and axial velocity variation data?

I have a question about turbulent energy spectra. I am now doing some analysis on PIV results. I do not know how to get this energy spectra out of 2-D PIV Data.

I have let's say 1000 instantaneous velocity fluctuation (u',v') maps. The sampling frequency was not so high to resolve in time. So I decided to find spatial spectra. Here I have power of 2 numbers of points (8,16,32,etc) in one horizontal line in my PIV domain. I have 1000 u' and v' samples for each point. location of these points (x,y) is known. I have to do FFT on these u' and v' values. but the thing is I do not have the frequency or wave number here. How can I relate the locations of these points to wavenumber? can anyone help me with that?

I have some ambiguities on pressure distribution around a turbine blade.

I know that value of pressure at pressure surface of the blade should be greater than that at suction surface of the blade. Positive (static)pressure(i.e. High pressure) is at pressure surface of the blade and negative pressure(i.e. low pressure) is at suction surface of the blade. Why does  negative pressure happen at the suction surface? Why doesn't  positive pressure happen at the suction surface? What is meaning of "negative pressure" here?

I think that if negative(low) pressure at suction surface decrease more, this means a delayed separation because according Bernoulli’s equation, if pressure decreases, velocity increases on the blade and this means that flow can have higher power transfer to the blade and hence has higher torque.

But regarding pressure surface of the blade, I studied some papers that authors say that the higher pressure on the pressure surface shows the flow is attached, stall has been delayed and the blade loading is high. I agree that blade loading is high in this time because the confined area of Cp plot increases but high pressure on the pressure surface of the blade show that there is  low velocity of fluid particles on the blade and lower power transfer to the blade and hence has lower torque. Therefore, How does separation delay? There is a problem and How can I interpret this cases?

Thanks.

I have the records of hydrophone deployed in a river I would like to find such a method to separate the water turbulence from other signals. 

My understanding is that the signals of sound intensity Sound to Noise Ratio (SNR) comprises both water turbulence and bedload noises in addition to some cases the noise which comes from human activities like (transportation, etc.).

I am interested to find  a way to decompose those signals. Is there any direction!

Simulation is transient in a bubble column reactor. I've tried information of some articles with different reactor dimension, gas velocity, orifice type and ...Physic is Euler-Euler. Both laminar and turbulent conditions do not converge no way. Please help me end this.

I am working on solving fluid flow problems. Mainly using vorticity stream function formulation. In that getting problem to write a code. I am using c++ code for programming. I ve boundary conditions for stream function on entrance, top and bottom. For vorticity i ll get from stream function vorticity transport equation. Can anyone one help me to write a logic for solving vorticity streamfunction formulation?

Hello,

I was investigating the incompressible Navier-Stokes equations analytically and hopefully being able to extract its bifurcation characteristics. During which I obtained a set of periodic solutions on the level surfaces of the helicity density: h=u*w. I realized that these solutions may be interpreted as the periodic exchange between the kinetic and vortical energy while preserving the local helicity density.

1.) I was wondered is this solution sort of related to the direct and reverse energy cascade?

2.) I am particularly interested in the physical interpretation of limit cycles on the surface of h=0. Will the limit cycles shrink down to a point, the origin?

3.) I am equivalently intrigued by the possibility of solution trajectories flip across that particular surface h=0.

Thank you.

Wang Zhe

Reynolds Stress Turbulence Mathematical Model

Modeling a turbulent confined swirling annular flow using a compressible RANS solver. Have to deal with the axial and circumferential velocities at the inlet. Velocity inlet seems not workable due to BC inconsistency. Any suggestions will be highly appreciated. 

Paul

I am working on aerodynamic shape optimization of an axisymmetric body (at zero angle of attack) in hypersonic flows (Mach 5-10). The standard k-epsilon turbulence model involves a significant numerical error due to it not being able to capture the turbulent shockwave-boundary layer interaction (as mentioned in earlier works) and hence cannot be used.

Dear LES / DNS experts;

We know that the dynamic procedure introduced by Germano 1991 has an asymptotic behavior towards walls and in laminar and transitional flows. The model has been widely studied with very interesting results and several variants.

However, my questions is more related to the physics behind the operation of the tensors.

Although, the model is highly unstable (huge jumps in the Cs in time and space) the averaging proposed by germano seems to alleviate the problem.

But here is the kicker; what does really make the value of Cs tends toward zero in laminar flows?

Thinking about the numerator and denominator of the famous equation to compute Cs; I 'll follow Lilly's definition . The denominator shows L_ij * M_ij and denominator M_ij * M_ij. Although it is based on an exact identity (Germano identity) I still see that one constant satisfying 5 independent equation is a little bit hard from a mathematical view point to fully accept.

Why do we get Cs = 0 from the previous expression?. Is this due to the fact that the difference in the S*S_ij at two different filter level is very similar that yields to zero.? Or is the fact that L_ij is the quantity that tends towards zero faster than M_ij?

What is the advantage of a cut-cell mesh for fluid CFD calculations with ANSYS/FLUENT?

Does the cut-cell mesh obtained with ANSYS/Meshing cut "better" than that with ICEM/CFD Hexa?

Mainly for stratified calculations of gases under turbulent conditions.

In any kind of fluid, rotation of fluid element is result of the non-zero curl of the non-conservative forces, which are acting on the fluid element. If the rotation of fluid element is along a circular path then it indicates that the curl of net forces is non zero at every point of closed path. But in the case of elliptical where curl of forces is not same at every point of a closed path. It seems that at some points of closed path curl of forces is zero and fluid element try to move in a straight line. Is there any condition which makes the rotation of fluid element along the elliptical path instead of circular path? Is there any relation which distinguishes both rotating motions? Please share your thoughts or any link.

Dear

I am trying to simulate VIV of CYLINDER IN CFX rigid body solver. I have made a sub domain around cylinder which follows the rigid body motion. Which type of mesh motion should I set on both sides of fluid fluid INTERFACE between the sub domain and outer fluid domain. If I use rigid body motion on both sides then mismatch between the mesh occurs on both sides of INTERFACE. SHOULD I USE conservative INTERFACE FLUX on both sides. What are effects of both. Moreover when I use conservative INTERFACE FLUX on both sides I get negative volume mesh but when I use rigid body solution on both sides I don't get negative volumes . The stiffness is set to 1[m5 /s]*VOLUME OF FINITE VOLUMES. Please guide.

Homogeneous and isotropic turbulence is known to be a good approximation of fully developed turbulence away from solid boundaries.

If I want to match the flow conditions in homogeneous isotropic turbulence to the bulk of a channel flow, which is the matching parameter that I should use (Reynolds number based on taylor microscale, Reynolds number based on integral scales)?

I recently made some measurements of channel flow using LDV to obtain the turbulence intensity in the channel. 

The turbulence intensity was calculated but I had trouble to quantify the uncertainty of turbulence intensity.

Does anyone know how to quantify the uncertainty of turbulence intensity?

As a flow enters or leaves a control volume it takes in or takes out some amount of energy as work. This concept mathematically leads enthalpy to appear in the 1st law of thermodynamic equation for control volume; But what is the physical sensible concept of this flow work? 

There are n number of references describing energy cascading hypothesis put forth by  Kolmogorov. According to this theory (hypothesis), the disturbance in the system is propagated though large eddies to small scale eddies till the viscous forces take over inertial forces and subsequently it is dissipated by the viscosity of the medium. Now i have problem visualizing this process.

For example if we consider a water column (sea system) using layered approach ( topmost layer close to the sea surface and bottomost layer close to seabottom) and lets say surface is perturbed vertically by a huge stirrer e.g wave breaking which is removed after some time:

how should the distribution of eddies be visualized in such a system ? Will it be like smaller eddies enclosed within larger eddies for each layer  or  larger eddies formed close to surface and smaller eddies in the subsequent bottom layers? 

Any explanation in this respect is appreciated.

I need to find out the residing time of a droplet,and that too in a turbulent background flow.Let the droplet introduced in the flow at some point Z₁ traveled along the turbulent co-flow and reached at a point Z₂ .I am having the point data of velocity on these points.How can I find the residence time.Should I go for a direct substitution or should I go with integration.Please help me to resolve the problem.Provided I am having more point data in between Z₁ and Z₂ 

Hello,

I have done LE simulation for a flow over flat plate to study about transition and saved the results for every 100 time steps. I ran simulation for the time period of 2.5s and I have sampled the data for the last 1.5s. Now I would like to generate a plot for time averaged skin friction or shear stress distribution along the inclined plate using tecplot or CFD post. I am having hard time generating the result. If someone could spare some time and suggest me something it would be of great help.

Thank you

Kind Regards,

Kumar

For example one solution that I came across in the literature is by introducing a term in the NS equations for compensating the pressure fluctuations present in the flow (because of not having a divergence free synthetic eddy).

Dear colleagues, I am performing DNS with the Burger equation (1D) without force and with periodic boundary conditions. So, the initial condition (the energy content) will decay towards zero due to the viscous forces. In other words, the viscous force will consume all the energy introduced by the initial conditions. Thus, I will never get a steady state solution where statistics can be computed. Specifically the mean of the velocity. Since I need to compute the turbulent kinetic energy and its dissipation (kappa and epsilon) I need to compute the fluctuating velocity but to do so I need the mean of the velocity. How can I compute the mean of the velocity from a data set that never reaches a stationary condition

I am confused with the database on the web and do not know how to use it

Hello all,

I am trying to run multiple simulations of a supercritical aerofoil with a high velocity jet at trailing edge using the SA turbulence model. So far I have managed to obtain mesh independence even if I maintained a y+ value of between 15 - 35. I know the y+ is important for SA and it should be kept as close as possible to 1 but when I try to do so, my simulations diverge and crash.

I am using ICEM to generate a C grid structured mesh with boundaries at 25 chords away. All simulations are done at low Mach (0.1), chord of 1m and jet velocities between Mach 0.2 -0.4.

Anyone who has encountered this issue before or knows how to solve it?

A water turbine is a rotary engine that converts kinetic and potential energy of water into mechanical work. What are the factors that will affect the functionality of the system, and how can I improve turbine system  effectiveness, rotating speed, and power output?

I would be obliged if someone knows good lecture to this topic. In my case, I'have to focus on a compression shock in shock tube from round to square geometry.

I'm glad for every source of information.

Sincerely yours

Which is the best turbulence model I should use for the simulation of a water stream through a valve with Re > 10000? Simulation will be done in OpenFoam.

Use which contour function and variable?

How to calculate turbulence within a trap?(expression?)

Software: ANSYS CFX AND FLUENT

Dear all

I am working on Flow controls in wind turbine blade.As a part of that,vortex generator should be installed on the blade and analysed. CFD analysis would be a better to startup the optimization.I was stuck in modelling of vortex generator in the wind turbine blade profile.To describe in detail,the wind turbine blades are twisted .In CATIA or SOLID WORKS,the procedure is disintegrate the blade to considerable parts and generate planes.In each plane,airfoil with corresponding twist and chord length is modeled and finally joined to design a wind turbine blade.Now ,modelling of wind turbine with vortex generators will be the next step.Here I have to place the VG in span wise manner.How can I model the generator on the blade surface.If the blade is untwisted,it could be done while multi-section option.But the blade is twisted and the designer isn't sure the modeled vortex generator is on the surface or not.Hope you get my problem clearly.Waiting for your valuable reply !!!!! Thanks all

Try several times and my answer was 3

When we report the fluid salinity of a particular inclusion, based on Fluid Inclusion studies, can we relate the results to salinity of primary fluid before trapping or not?

I am simulating an axial turbine.

Moreover, I tried different mesh topologies and all give me almost the same number. In fact, BL is not captured well. RE is equal to 4.4* 10^5 and according to mentioned formula in the following link, I computed y_1(first cell height) by using Y+=30 and 60:

but in theses two cases(Y+=30 and Y+=60), BL is not captured well. Why? In the other words, there is not maximum turbulent viscosity ratio(logarithmic region) adjacent wall. And also, I studied the following link:

For this reason, performance parameters are not validated well. In fact, there is difference with the experimental value. This difference is not big but it is not good! I think that it is difficult for capturing BL by using low Re number turbulence model( for example, K-e model)!!!

I am grateful that guide me about these problems.

Thanks.

Best.

Ali

The sensor are use in experimental set up for Air Flow Bench in fluid mechanics laboratory. The data collected will be directly transfer to the National Instrument and DaisyLab Software. 

Which sensor should be used?

need to compute velocity of moving body using camera.i found KLT tracker can be used for this.is there any other trackers for tracking up.corner...and estimating the velocity of object ??suggest some best tracker for this application??

 I was solving simple 3D problem in which water is flowing around heated cylinder with laminar flow and energy equation in fluent. Problem is in steady state condition. I simulate it around 2000 iteration but I am not getting converged solution. How can I simulate my simulation to get converged solution?