[pd.iiest]

Hi
I am working on turbulent pipe flow problems, three dimensional. I am trying to find out the values of fluctuating velocity of all three components, u',v',w'. Mean velocities inform of velocity magnitude, velocities in x,y,z directions are available but not the fluctuation parts. Please help me. Thanks in advance....

[nasir.iitd]

In 'General' window tick on transient option. Then for getting u', v' and w' you need to switch ON 'Data sampling for time statistics' option in 'Run Calculation' window for transient flow before starting simulation.

[flotus1]

You have several options.
You can run the case in steady-state using a Reynolds stress turbulence model to get the individual components of the Reynolds stress tensor.
Or you can perform a LES or DNS to gather information about the fluctuations.

[pd.iiest]

Thank you all...

[pd.iiest]

Dear nasir i have done the simulation as per your suggestions..and now i am facing problem to collect data. from where i can get the u',v',w' etc? i am a beginner in post processing. please help me.. thanks

[nasir.iitd]

Hi Prasun,
Follow the sequence:
1. Open xy plot window
2. Mention plot direction for your case.
3 On the right of xy plot window, under Y axis function, open the first dropdown list and select unsteady statistics.
4. There after, in the 2nd dropdown list, you will find rms quantities.
5. Before opening xy plot window,
you have to define the line or surface in your domain where you want the plot using following sequence:surface>line/rake or plane>define your line/ plane with specific name.
6. This new surface will be seen in xy plot window under surfaces. select the newly defined surface and click on PLOT.
I think this may help you.

[pd.iiest]

Thank you nasir... its help me a lot.. thanks again

[pd.iiest]

From fluent the RMS values of velocity fluctuation can be obtained but these are mean values(<u'^2>) how can I get fluctuation velocity (u') with time?
Thanks in adv............

[nasir.iitd]

Hi Prasun,
If you want to calculate temporal velocity fluctuations, follow the below steps:
1. You need to define the point/points in the domain where you want to calculate the fluctuating components.
2. Then go to Monitors> Surface monitors.
3. In surface monitor panel:Name(file name)>Write(click and write the location where you want to save file)>X axis(choose flow time)> Get data every (choose time step)>Integral (you can choose vertex avg or any suitable)>Field variables(choose velocity) and then your required component>Surface(choose the point you have defined> OK.
4. Then in the directory you specified, under the file name at every time step data (instantaneous value)will be stored.
5. In FFT, there is option for subtracting mean from instantaneous quantity of temporal data.
I hope this may be helpful to you.

[pd.iiest]

thank you nasir.. it works .. in this way i got temporal velocity fluctuations for different points.. but in my problem i have to calculate velocity fluctuations for a lots of points, so is there any option from where i can easily get values of all the points or i have to make separate surface monitor for different points? Like if I create a line or rake or something. Actually my problem is a simple pipe flow simulation, i want to get values of u',v',w' at different locations.

[nasir.iitd]

Hi Prasun,
Can you tell me, by getting fluctuating components of the velocity field,1. what do you want plot? 2. Plotting along axial location or diametrical location? 3. Plotting of temporal or spatial data?

[pd.iiest]

Actually I want to plot total turbulent kinetic energy budget in different axial and redial direction in a pipe bend flow. Reynolds stress transport equation can be defined as:
∂k/∂t= Pk + εk + πk + Dk + Tk − Ck
where
(i) Production (Pk) = −<uiuj> ∂Ui/∂xj
(ii) Dissipation (εk) = −ν <∂ui/∂xj∂ui/∂xj>
(iii) Pressure diffusion (πk) = −1/ρ ∂(<pui>)/∂xi
(iv) Viscous diffusion (Dk) = (ν/2) ∂^2(<uiui>)/∂xj^2
(v) Turbulent diffusion (Tk) = −(1/2) ∂<uiuiuj>/∂xj
(vi) Convection (Ck) = <uj>/2 ∂(<uiui>)/∂xj
That's why i need u',v',w'

plz find the attachment for better understanding of positions where i want to obtain the values of u',v',w'.

Thanking you nasir for your help...

[p.1]

Originally Posted by pd.iiest

Actually I want to plot total turbulent kinetic energy budget in different axial and redial direction in a pipe bend flow. Reynolds stress transport equation can be defined as:
∂k/∂t= Pk + εk + πk + Dk + Tk − Ck
where
(i) Production (Pk) = −<uiuj> ∂Ui/∂xj
(ii) Dissipation (εk) = −ν <∂ui/∂xj∂ui/∂xj>
(iii) Pressure diffusion (πk) = −1/ρ ∂(<pui>)/∂xi
(iv) Viscous diffusion (Dk) = (ν/2) ∂^2(<uiui>)/∂xj^2
(v) Turbulent diffusion (Tk) = −(1/2) ∂<uiuiuj>/∂xj
(vi) Convection (Ck) = <uj>/2 ∂(<uiui>)/∂xj
That's why i need u',v',w'

plz find the attachment for better understanding of positions where i want to obtain the values of u',v',w'.

Thanking you nasir for your help...

Hi, I have to do the same thing. Could you please tell me how to do?

[Jinjun Ye]

Originally Posted by flotus1

You have several options.
You can run the case in steady-state using a Reynolds stress turbulence model to get the individual components of the Reynolds stress tensor.
Or you can perform a LES or DNS to gather information about the fluctuations.

Hi, I would like to know how to obtain the individual components of the fluctuating velocity, under the condition of steady state calculation with the RSM model.
Thanks for your reply！

[Jinjun Ye]

Originally Posted by pd.iiest

From fluent the RMS values of velocity fluctuation can be obtained but these are mean values(<u'^2>) how can I get fluctuation velocity (u') with time?
Thanks in adv............

Hi, I would like to know how to obtain the RMS values of velocity fluctuation, in the steady state using the RSM model.
thanks