[Struggle_Achieve]

Hello Everyone,

I am considerable new to openFOAM and I have been trying to do RTD analysis using openFOAM from last few days.
I took the example case of scalarTransportFoam (pitzDaily). I wanted to plot the trademark RTD curve for the concentration at the outlet.
I read the post from Daniel: http://www.cfd-online.com/Forums/ope...tml#post413277

And accordingly I put the following code in the controlDict to get the velocity weighted average concentration at outlet using patchMassFlowAverged function object from simpleFunctionObjects.

functions
(
   massFlowAverage
   {
       type  patchMassFlowAverage; //patchIntegrate; //patchAverage;
       functionObjectLibs
         (
           "libsimpleFunctionObjects.so"
         );
       verbose true;
      fields (T);
       patches
         (
           OUTLET
         );
       factor 1.0;
       outputControl timeStep;
       outputInterval 1;
   }

);

But after running scalarTransportFoam I am getting such type of output

DILUPBiCG:  Solving for T, Initial residual = 9.99425e-07, Final residual = 9.99425e-07, No Iterations 0
 Averages of T :
Time = 8.6019

DILUPBiCG:  Solving for T, Initial residual = 9.99425e-07, Final residual = 9.99425e-07, No Iterations 0
 Averages of T :
Time = 8.602

DILUPBiCG:  Solving for T, Initial residual = 9.99425e-07, Final residual = 9.99425e-07, No Iterations 0
 Averages of T :
Time = 8.6021

There is no value for averages of T (T here is concentration):
In the folder postProcessing I am getting a file with only time values in it (no concentration value).

I also tried to run execFlowFunctionObjects and I got following error:

Time = 0.43
    Reading phi
    Reading U
    Reading p
--> FOAM Warning : 
--> FOAM FATAL IO ERROR: 
cannot find file

file: /home/1016914/Mangal_scalar/pitzDaily/0.43/p at line 0.

    From function regIOobject::readStream()
    in file db/regIOobject/regIOobjectRead.C at line 73.

After this i tried to run execFlowFunctionObjects -noFLow, for which the output is like following:

Time = 0.11
    Operating in no-flow mode; no models will be loaded. All vol, surface and point fields will be loaded.
Reading volScalarField T
Reading volVectorField U
Reading surfaceScalarField phi

But there is no data corresponding to this in the postProcessing folder.

I know that I am missing something very basic.
I would be very grateful if anyone can help me in this matter.

Thanks and regards,
Singh.

[Struggle_Achieve]

Hello Everyone,
Before anyone else makes fun of me, it will be right for me to point out my mistake..
It is all working fine. Sorry for the chaotic post, I just entered wrong patch name.
It should be 'outlet' instead of OUTLET. THATS IT!!

[HappyS5]

Hello,

I am doing tutorial 10 from the 3 week tutorial series. It is the RTD tutorial in the series but called tutorial 10.

The author used "ratio of that scalar mass inlet to the whole mass inlet" to calculate average T: http://cfd.at/downloads/FoamTutV4_10-ExampleTen.pdf on page with "Average value of T on the outlet for two inlets versus time" graph. I would like to know what the scalar mass inlet and whole mass inlet is. Is the scalar the "current" time and the whole mass the value placed in the documentation for "T"?

When calculating RTD, the author states: "Next, to obtain the RTD plots, simply calculate the gradient of change in average value of T on the outlet from time 0 to 120s, export the data to Excel, and plot the results." I would like to know how to calculate the gradient of change.

Thanks in advance.

[HappyS5]

Originally Posted by HappyS5

Hello,

I am doing tutorial 10 from the 3 week tutorial series. It is the RTD tutorial in the series but called tutorial 10.

Thanks in advance.

Hello,

In truth, I just needed to read the tutorial better. I was blind or something. Anyhow, I was able to Integrate Variables and then plot that data over time after I selected the row of integrated data that I wanted to graph over time.

Now, I need to read and understand the description of how to calculate the RTD data but the description is not as clear to me now. See above.

[wasp0]

Originally Posted by HappyS5

Hello,

In truth, I just needed to read the tutorial better. I was blind or something. Anyhow, I was able to Integrate Variables and then plot that data over time after I selected the row of integrated data that I wanted to graph over time.

Now, I need to read and understand the description of how to calculate the RTD data but the description is not as clear to me now. See above.

HappyS5, did you manage to calculate the RTD data? Can you please share. This step is unclear for me too, and I haven't figured out how to do it.

Thanks.

[HappyS5]

Originally Posted by wasp0

HappyS5, did you manage to calculate the RTD data? Can you please share. This step is unclear for me too, and I haven't figured out how to do it.

Thanks.

Sorry, I moved on. I did not figure out how to normalize that data with ParaView. I do not understand the suggested process.

If you figure it out, please post it here.

[Struggle_Achieve]

Hello Everyone here,

Sorry I have very little experience with multiple inlets and outlets case. I have plotted the RTD graph for a single inlet and outlet case.
I used the 'function' utility of OpenFOAM to extract the required data at the outlet. That is:
I have defined a funcitonObject: 'massFlowAverage' of type 'patchMassFlowAverage' to calculate the concentration at the 'outlet' patch.

Then I did the required calculations in python to plot and analyze the RTD curve obtained like: Melf Flow Characterisation in Continuous Casting Tundishes, Sahai and EMI (1996)

Note: In order to normalize the Concentration to E_theta and Time to theta I used exit age function, Tmean. Basically I have multiplied my exit age function with Tmean to get the normalized concentration that is E_theta
and I have divided Time by Tmean to get the normalized time that is Theta.

Sorry for the confusing post. I did this thing a long time back.

Thanks and Regards,
PS

[italianjon]

Originally Posted by HappyS5

Sorry, I moved on. I did not figure out how to normalize that data with ParaView. I do not understand the suggested process.

If you figure it out, please post it here.

Hi All,



I'm an OpenFOAM newbie too, but found this thread while searching for a solution to this very problem. I was doing the 'Tutorial Ten' for RTD, but I could not work out how to calculate/export the gradient.


In the end, I managed it, but I'm not sure how. I thought I would write here what I think I did and perhaps someone can recreate my steps and confirm if this is correct.


When I integrated the variables, this worked as expected and I could create the plots for T against time, within Paraview. However, my spreadsheet window (within Paraview) would only display the single value for the selected timestep. If I exported the data only the single displayed value would be exported into a spreadsheet.


Within filters, I found the gradient filter, bit this was greyed out, requesting VTKImage data. I do not know what this is yet, I havent got that far through the tutorials. But, below 'Gradient' is 'Gradient of Unstructured Data' and I thought I would explore to see if this would work. I got myself in a bit of a mess and decided/had to delete from the Pipeline Browser, the outletVTK trees, and start again.



This time when I loaded the VTK files for the outlet, and then integrated the variables, the Paraview Spreadsheet window has a complete table with all 120 seconds of data. I could export this to (in my case) Libreoffice and produce the graphs at the end of the tutorial pdf.


I honestly have no idea what I changed, but now I can not return to the single data point with in the paraview spreadhsset view of integrated variables - the table is there all the time. Which is good, but I do not know what I did.

[RobertHB]

I've done several RTD calculations for single inlet/outlet cases using OpenFoam, Paraview and Matlab on basis of the mentioned tutorial. The workflow is as follows:


1. Conduct your simulation
2. Open your results in paraview and select the cells of your outlet patch. This is done using "select cells on". It's a bit easier if you only load the target patch and not the whole internal mesh.
3. With the cells selected use the "Integrate variables" filter. A spreadsheet window will open.
4. Select the spreadsheet window and use "Plot Selection Over Time". This might take some time, after which a plot window will open.
5. Select the plot window and "Save Data" as .csv .


The data can be further processed with Matlab/Python/whateverfloatsyourboat. Calculating the first derivative (e.g. with Matlabs diff() function) of the scalar results in the arrival rates at the selected patch.