There is no output to .vtu generated in the fsi-results folder

[LeoXXVII]

Currently using Gridap v0.16.5, GridapGmsh v0.4.2, and GridapFSI v0.1.0

The terminal gives a result, but there is no output created in the corresponding folder.
The only file is "fields.pvd" that has nothing inside

[oriolcg]

Hi @LeoXXVII , which driver are you running? Are you adding the argument is_vtk=true when calling main?

By the way, the branch refactoring_to_Gridap_0_17 is already working with Gridap@0.17 and GridapODEs#TransientMultiFieldCellField. I will merge it to master once GridapODEs is updated.

[LeoXXVII]

Sorry for not replying before, I am using the ElasticFlag_FSI1.jl drive.
Also, no I didn't add the is_vtk. I tried, and yes, it gives differents .vtu results for each step (sadly in the name of that file it's called fields_x.0.vtu and the ".0" part interfere to visualize on paraview as a group, so yeah, i just change the file name)

Also, I was in the middle of writing the result it wasn't in the description, but right now it works it.
But I still didn't update to Gridap 0.17, because of the DiscreteModel() thing. I write about that in this thread gridap/Tutorials#125 (comment)

[LeoXXVII]

another example about the names of the files.
It can't visualize in only one model on Paraview.

[oriolcg]

Hi @LeoXXVII , you should be able to open the file "fields" in paraview and this should contain all time steps of the simulation

[LeoXXVII]

I mean, with that "dot", paraview separates the files into 9 groups.
I recently edit all the files names to delete that dot and work well.

but if you refer to the field.pvd..... em...that contain nothing.

[LeoXXVII]

Aside, em... The second test in the readme.md of GridapFSI.jl

The output is different compare with the first example, I mean, is the same model but different time-lapses, and no parameters like Re, rho_s & Um. Causing this result. (I change the final time for 1 second instead of 10 sec, tf=1.0 for less space usage)

This is an image of 0.04s passed

Aside, which program do you use for the FSI2.gif? I mean, if this didn't get by the output in paraview.

[oriolcg]

This is the result that I get at t=0.04 with the latest version merged to master.

calling with this set of parameters

output = main(
         strategy="biharmonic",
         alpha_m=1.0e-6,
         alpha_m_weight="volume",
         model="../models/elasticFlag.json",
         dt=0.01,
         tf=5.0,
         theta=0.6, is_vtk=true, coupling="weak"
         )

Note that this is with a coarser mesh and I use the weak coupling approach.

Regarding the .gif file, I just use a web application to convert a set of png files to a gif.