Acoustics – Basic Postprocessing With ElmerSolver & Paraview

[Adev]

Ok, I edited the sif copying and pasting from the CWENG sif file to mine, yielding the attached sif below, ran it and got the error message also attached, stating that "AddEquationBasics: Variable > pressure wave < exists but it is not associated to any equation". The statement "Variable = -dofs 2 Pressure Wave" appears in the Solver 1 routine, as far as I can tell, no different from the CWENG sif file ??

PipeOnly.sif

(4.72 KiB) Downloaded 237 times

error msg 1 of 12-16-19.txt

(964 Bytes) Downloaded 247 times

[kevinarden]

the body entries call out the equation, the solvers specify the available equations, the equation entery specifies the equation

cweng sig has this

Equation 1
Name = "Helmholtz"
Frequency = Variable time; Real MATC "freqVec(tx - 1)"
Active Solvers(3) = 1 2 3
End

yours has this

Equation 1
Name = "Helmholtz"
Frequency = Variable time
Real MATC "freqVec(tx - 1)"
Active Solvers(3) = 1 2 3
Active Solvers(1) = 3
End

[Adev]

Thank you again Kevin for your generous time.
I have a question: One obvious item from your review is that I had an extra statement "Active Solvers(1) = 3" in my sif file. My question regards the other statements "Frequency = Variable time; Real MATC "freqVec(tx - 1)"; and Active Solvers(3) = 1 2 3" which CWENG had in 1 line and I thought Icould separate them into 1 line each … is that allowed or should I keep 1 line as by CWENG?

[kevinarden]

Frequency = Variable time; Real MATC "freqVec(tx - 1)"

the semicolon means it is really two lines so it can be entered as

Frequency = Variable time
Real MATC "freqVec(tx - 1)"

[Adev]

I fixed the previous sif file error and ran it. The run progressed a little further but then crashed on the following error msg:
MAIN: Reading Model: PipeOnly.sif
LoadInputFile: Scanning input file: PipeOnly.sif
LoadInputFile: Loading input file: PipeOnly.sif
Load: FATAL: Can't find procedure [2_]

I suspect it has to do with a 2nd Flux Solver I loaded in the Definitions file, for which I followed the process described elsewhere on the forum (by mzenker at viewtopic.php?f=4&t=5001#p17872).
Here is my updated sif file and a screen capture of my definitions

PipeOnly.sif

(4.7 KiB) Downloaded 242 times

Definitions Capture 12-17-19.JPG (43.06 KiB) Viewed 3943 times

[kevinarden]

In the referenced topic they replaced fluxsolver with a fluxsolver2 however they did not use both, and when they replaced it they replaced it using the same name fluxsolver. In your case you appear to be trying to use both of them.

first on in Solver 2 and the second one in Solver 3. You renamed the second one in the sif file but not in the PDE that was added. in your list of PDE both are called fluxsolver. So when you reference "FluxSolver 2" in your sif file it can't be found because it is called fluxsolver.

If you intend to use both then you have to use a unique name for the added PDE and use that name in the sif file.

[Adev]

Thank you Kevin,
When I added the 2nd flux solver from the File-Definitions menu, I did indeed add the renamed flux solver 2 by editing and renaming the flux solver.xml file. The confirmation I have is that modified xml file was added is in the opened definitions tree where you can see the flux solver name is Flux and Gradient 2 (vs the first instance without the "2"). The way I added the 2nd Flux solver was by following mzenker's instructions. I attach the xml file I edited per mzenker and then added.

fluxsolver2.xml

(1.82 KiB) Downloaded 235 times

[kevinarden]

<PDE Name="FluxSolver" >
<Name>Flux and Gradient 2</Name>

<Solver>
<Parameter Widget="Edit" >
<Name > Procedure </Name>
<DefaultValue> "FluxSolver" "FluxSolver" </DefaultValue>

PDE name FluxSolver conflicts with the one that is already there so if you need both then this one needs renamed. In your sif you refer to it as "FluxSolver 2"

if you want it to be "FluxSolver 2" then you have to name it "FluxSolver 2" in the PDE

<PDE Name="FluxSolver" 2 >
<Name>Flux and Gradient 2</Name>

<Solver>
<Parameter Widget="Edit" >
<Name > Procedure </Name>
<DefaultValue> "FluxSolver 2" "FluxSolver 2" </DefaultValue>

[Adev]

Thank you Kevin,
I figured out that, to edit the definitions file (PDE), I simply double-clicked on the name, changed it as you suggested, and then as there is no "Save" utility (Only Save As", I just closed it, then checked to see if it took it … and it worked!? This may be worthwhile to know for other beginners like me.
Anyway, I re-ran the Solver and it made a lot more progress. It looks like with your help, I'll get there.
It crashed again, this time related to Save Scalars, and there were a number of warnings. If you could find the time, here is the Solver Log file

Solver log 12-18-19 A.txt

(1.41 KiB) Downloaded 222 times

[kevinarden]

You are getting beyond my experience now. However the literal interpretation of the message is that the variable does not exists so it can't be output.
I went through the other topic you posted

"I suspect it has to do with a 2nd Flux Solver I loaded in the Definitions file, for which I followed the process described elsewhere on the forum (by mzenker at viewtopic.php?f=4&t=5001#p17872)."

but could find no reference to that variable there.