[Dittus Boelter]

Hello,

I am encountering a persistent issue while running transient simulations of an Internal Combustion Engine (ICE) using Converge Studio CFD. The simulation halts prematurely with the following error message:

(Rank 148) FATAL_ERROR: [3500] An unknown error was encountered while cutting the geometry at time -5.783321e+02.

Upon repeating the simulation, I have observed that the error appears to occur at random crank angles, although the deviations from the reported time are generally small. Despite thorough diagnostics, including checks for intersections and problematic triangles, I have been unable to identify any obvious issues with the geometry.

Additionally, the simulation output includes several .dbg files:

• box148.dbg

• side_polygons148.dbg

• surface_polygons148.dbg

• surface_triangles148.dbg

which I am struggling to interpret. These files contain a series of coordinates along the liner, yet their location appears to be unspecific and does not align with known vertices.

To facilitate troubleshooting, I am attaching these .dbg files for further analysis:

https://drive.google.com/drive/folde...usp=drive_link

Any insights or suggestions on potential causes or troubleshooting steps would be greatly appreciated.

Thank you in advance for your help.

Sincerely,
Dan

[ksrivast]

Hello Dan,

Are there any CLOSE events occurring around -578 CAD? Intersection issues can arise if regions are not assigned correctly and poor disconnect triangles are created during CLOSE events that intersect with the rest of the geometry. Since disconnect triangles are virtual triangles created by CONVERGE during run-time, Studio will not be able to identify such issues.

Please confirm your regions are set correctly. You can also manually construct disconnect triangles to check whether they will intersect of not. Check for isolated triangles as well.

It would also be helpful to run the case in "no-hydro" mode. Simulation Parameters > Run parameters > Simulation mode > No hydrodynamic solver with triangle intersection checking through Studio or check_grid_motion_flag=2 in inputs.in , to see if the simulation still crashes. This can give us a better idea of what's happening.

Hope this helps.

Sincerely,

[Dittus Boelter]

Hello,

Thank you for your prompt response and valuable insights into troubleshooting the issue, I sincerely appreciate your support.

I can confirm that there are no CLOSE events occurring around the specific crank angle mentioned in the error message. Additionally, as I previously mentioned, the error occurred at various crank angles during repeated simulations, indicating a lack of consistency in its occurrence.

Regarding the suggestion to run the simulation in "no-hydro" mode, I can say that I did perform a run. Regrettably, the simulation still terminated with the same error message, albeit at a different, unspecified crank angle compared to previous attempts.

Although I thoroughly checked the regions' assignment for potential intersection issues and isolated triangles, no apparent mistakes were evident to me. However, I decided to redo the entire pre-processing and set-up procedure from scratch, which surprisingly resolved the problem. Despite this solution, the root cause of the error still remains a mystery to me.

I sincerely appreciate your efforts to assist me in troubleshooting this issue and thank you once again for your support.

Best regards,
Dan