[combustion_lady]

I am using an open program called Multall for CFD/turbomachinery design. I have been able to run MEANGEN, STAGEN programs but not MULTALL (with Fortran 77). However, the output files (flow_out and grid_out) that come from running the last program, Multall, are somehow not in the right format. Therefore, I cannot use the last PLOTTING program that requires the two files as input.

The author provides example files for each MEANGEN, STAGEN and MULTALL, but not for the PLOTTING part.

I am wondering if anyone, would be willing to run a simple case on Multall (like a sample case such simple compressor contained in the "multall-test-cases" folder) and post the output files to test the last PLOTTING program and help me understand how to debug this problem.

Thanks,
This is the program I am referencing: https://sites.google.com/view/multal...tem?authuser=0
It can be freely downloaded from the dropbox.
Any help would be really tremendous! only takes a few minutes

These are the steps (Linux)
1. download multall from dropbox link (see above)
2. sudo apt-get install gfortran
3. in the "multall program" folder (located in the MULTALL folder), create a text file called "intype", it must have the single letter N (can also contain O)
4. gfortran -mcmodel=medium multall-open-17.5.f -o multall (a different version can also be picked, this must be run from the multall program folder)
5. ./multall < 3stg-compr+samods-17.5.dat (this file is located in the multall-test-cases folder and should be moved to the the multall program folder before running, there are other test cases, I was hoping to run a compressor one)
That's it! There should be a few output files, but the ones I'm looking for are the two unformatted output files "flow_out" and "grid_out"

[flotus1]

I don't quite understand your request.
You were able to run the program, and get the files you need. How would someone else running the program help you?

[combustion_lady]

Sorry, I should have been more clear. I can compile Multall, but running fails for me (segmentation). Based on the only thread that I could find on the internet, this is due to "nothing to do with the program" errors that some users get and others don't. This is why someone running the last program (Multall), would be really helpful, so I could at least have an idea of what a typical output looks like for post-processing purposes. I have tried running this with Ubuntu on a virtual machine.

[flotus1]

https://www.dropbox.com/s/r5vfwtstmi...st.tar.gz?dl=0
Let me know when you downloaded the data.

The case ran for 7201 iterations and then wrote the output files.
Last line in the shell output indicates some divide-by-zero error though.

 7201  0.0543  45  93  50  0.0005  0.0037   286  275.58  422.61  52  37  41     47.6278     47.7116      1.0018
Note: The following floating-point exceptions are signalling: IEEE_INVALID_FLAG IEEE_DIVIDE_BY_ZERO

The pre-compiled plotting tool "plotall-17.1.x" doesn't run, because some libraries are missing on my machine.

[combustion_lady]

Thank you very much for your help. I was actually able to solve the issue, and got the same error at line 7201. The plotting does work for me, but only on Windows with the .exe files.

[flotus1]

Btw, I ran a second test with all debug flags enabled that I know of. Set the max number of steps to 20 though.
The same warning occurs at the end, without any additional insight. So the error appears somewhere during post-processing the result, and would require actual debugging effort to get a hold of.