Error in miu_fsi coefficient

I'm trying to simulate the deformation of a membrane to a fluid flow, but trying to solve it immediately gives me the error "error in miu_fsi coefficient".

Here's the settings:



May i ask if i'm doing something wrong?

Hi Timmy,

Thank you for reporting the issue. Could you possibly attach an actual model file saved just before you encounter the error, as it makes it much faster and easier to reproduce issues.

Sure. Here it is.
In this model i applied a small modification where i lowered the height under the hypothesis that the fluid may need to have a space to pass through. But it still give me the same exact error. NewTest.fea

Thanks for attaching the model.

It seems specifically for the FSI mode does not parse/substitute the coefficients as expected, so for now you will have to use the actual values instead of 'rho' or 'mi' in the Equation dialog coefficients. Moreover, the expression for 'E = ((270-250)/(0.298-0.258))e2' is invalid as the exponent 'e2' can only be affixed to numerical values, in this case you could do something like '((270-250)/(0.298-0.258))*1e2'.

thank you for the answer. However the system still keeps giving me the error message "matrix is singular to working precision" despite there being no reason to do so. TrialCorrected.fea
May i ask what i'm doing wrong (wether it's an input issue or a boundary issure etc.)?

Are you quite sure of your fluid and solid material parameters are correct? As changing them to for example as in the FSI tutorial does allow for the simulation to run. If these indeed are the parameters you want you probably need a much finer time steps and possibly grid too.

Thank you for the answer. Sadly those are the specifics of the material i'm supposed to work the simulation with. However i've still tried to use the tutorial settings to check if it works. Sadly it doesn't: it takes an extremely long amount of time (to the point that after leaving the pc on for 2 days has yet to complete half of the simulation) despite being much faster with other systems, finer meshes and higher time span with smaller sampling time. Also, during said solving, it keeps giving me NaNs I attached to this post a screenshot  

Also. i attached to this post a version of the file with a simplified mesh because the original file surpasses the file size limit of 5 Mb. Nevertheless, the times of computing, as well as the other issues, remain unchangedTest_tutorialSettings.fea

I'm not sure we can help further without going into consulting territory, as the benchmark and validation cases pass, as well as with your model with those material parameters, so it doesn't seem to us that is an issue with the toolbox.

As the FSI solver is developed externally, and free and open source, you could potentially try to use it directly and step by step check where the issues are (in matrix assembly or solving), and perhaps ask the FSI solver developers directly for specific advice regarding your case:

https://redbkit.github.io/redbKIT