Learning Forum

[Usman]

Dear Team,

I hope this message finds you well. I generated the following 1/7 sector mesh using AMG. However, it shows the following error in FORTE. Could you please suggest me how to figure it out? Many thanks!

[Usman]

Can anyone please let me know how I can figure out this issue in AMG?

[Ren]

I'd suggest you refer to the advice I provided in other threads that you created.

 

[Usman]

Thank you, Ren, for the reply. I have gone through those tutorials. However, it the first tutorial, I face the same issue. When I create a sector mesh using AMG, it shows the same error as this thread.

[Ren]

Are you saying that the tutorial is producing the error? Could you tell me the tutorial name and the Forte release number that you use?

 

[Rob]

Moderator note: there are several threads and now issues in a tutorial. Whilst it's possible there's an issue in the Tutorial I'd be very surprised at this point in the year. 

So, I will leave this thread open and close https://innovationspace.ansys.com/forum/forums/topic/error-in-the-mid-of-the-simulations/  Any new threads that are opened and are related to this one will be removed.  

[Usman]

Can you please let me know how can the issue with AMG be figured out?

[Ren]

Is the piston initially at TDC or BDC? The screenshot suggests that the piston is at BDC. Please clarify.

 

[Usman]

It is at TDC. When we create a sector mesh in AMG, it creates the full stroke.

 

[Ren]

Please post the file "MONITOR".

 

[Usman]

Dear,

I am sorry. Actually, it is at BDC. When we use AMG, it creates a sector mesh like this. however, it does not run. Also, it does not calculate the CR as well. Even, if i follow the tutorial 1, and generate the sector mesh, it shows error when i start simulations.

[Usman]

Following is the MONITER file after using CR Utility:

FORTE Runtime Environment is 2023 R2 (15-May-2023 01:45:54 AM)
FORTE Runtime Environment is 2023 R2 (15-May-2023 01:45:54 AM)

LICENSE INFORMATION: 
 

LicNum: 373361.  License for SCEDWLW1123985_AD_HUD_AC_UK-u2061120-31576-001264 is obtained from the ANSYS license service running on SCEDWLW1123985.AD.HUD.AC.UK.
                                                                                                                                                                                                                                                                                                                                                              
Process #8 (rank 7) on host SCEDWLW1123985 will be used as the dedicated communication server.
Setting server memory parameter to 1000

     *****************************************************************
     *                                                               *
     *                      ANSYS Forte 2023 R2                      *
     *                    Build Date: May 15 2023                    *                   
     *                   Forte Simulation Package                    *
     *                                                               *
     *                Copyright (c) 2023 Ansys, Inc.                 *               
     * Unauthorized use, distribution or duplication is prohibited.  *
     *   This product is subject to U.S. laws governing export and   *  
     *     re-export.  For full Legal Notice, see documentation.     *    
     *                                                               *
     *****************************************************************

The following arguments were passed:
 -i CR_A.ftsim
 -l FORTE.log
 -rn Nominal_
 -o Nominal.ftres

Using MPI Version: Intel(R) MPI Library for Windows* OS, Version 2021.8 Build 20221129
This run is using 8 MPI processes.
1 MPI process(es) will be used as the dedicated communication server(s).
On process #1 (rank 0) on host SCEDWLW1123985 there are 3 core(s) available and 1 thread(s) will be used.
On process #7 (rank 6) on host SCEDWLW1123985 there are 3 core(s) available and 1 thread(s) will be used.
On process #5 (rank 4) on host SCEDWLW1123985 there are 3 core(s) available and 1 thread(s) will be used.
On process #3 (rank 2) on host SCEDWLW1123985 there are 3 core(s) available and 1 thread(s) will be used.
On process #2 (rank 1) on host SCEDWLW1123985 there are 3 core(s) available and 1 thread(s) will be used.
On process #4 (rank 3) on host SCEDWLW1123985 there are 3 core(s) available and 1 thread(s) will be used.
On process #6 (rank 5) on host SCEDWLW1123985 there are 3 core(s) available and 1 thread(s) will be used.
 Error reading variable: project_version

=============================================================
-------------------------------------------------------------
 Error reading variable: is_crank_based
MPI Finalized
Error: FORTE simulation failed to complete
Licenses Released

[Ren]

That does not provide much clue. 

Can you run the .ftsim file provided in tutorial 1 without any modification and see what happens?

[Usman]

Dear Ren,

However, when I followed tutorial 1 step by step, it showed the same error. 

In tutorial 2, a working model is available. I changed the dimensions of that model using the transform mesh option and used it for the analysis. It is a 1/6th model, however, I need a 1/7th model for the study. I compared the results of the 1/6th model with the experiments and they are in good agreement. I compared pressure, HRR and emissions. Is it possible to rely on these validations as most of the values are the same with the experiment except the periodicity? Or is there any option within Ansys forte to transform a 1/6th model to 1/7th?

 

Many thanks!

[Rob]

What is different between the tutorial model and your model other than the sector angle? That will answer the question as to whether you can use a random tutorial to validate your experiment. Have you spoken to your Prof about taking this approach? 

[Usman]

They are almost the same i.e., bore, stroke, mass of both fuels, injection timings. No, I have not discussed it with my Prof. If the results are reliable, I guess, he will have no issues with it.

[Rob]

You may want to discuss with him.  Is total mass the same, or sector mass? Note, I don't use Forte but Ren's out until Tuesday or so. 

[Usman]

what is meant by sector mass? I am providing the total mass. This is what we do in a sector mesh, I guess. We provide the total mass.

[Rob]

I'd check - guessing isn't something I'd rely on when trying to defend a modelling assumption. 

[Usman]

Thank you, Rob. Any suggestions will be highly appreciated.

 

[Ren]

Hi, Usman,

1. If you mean the tutorial "Simulating a Diesel Engine Using a Sector Mesh" by tutorial 1, I've run the CR Utility using the project file provided in the tutorial and the calculation completed successfully in a few minutes and without any issues.
2. I've also run a complete simulation using the same project file and the calculation also completed successfully in about 30 minutes without any issues.
3. It seems that you did not test the project file provided in the tutorial. Instead, you set up the project following the instructions in the tutorial and this project was having issues. In this case, please check the setup and make sure no mistakes are made.
4. It's not possible to convert a 1/6th model to a 1/7th model in Forte. You need to create a new geometry/mesh.
5. For a solid-cone injector the injected mass you input in the setup is for the full 360-degree domain. In light of this, there should not be any significant differences in global variables between a 1/6th model and a 1/7th model. However, you should expect local differences in the spatial distributions between these two models because the number of nozzles are different.

 

[Usman]

Thank you, Ren, for your response. 

Could you please also let me know if we can change the piston movement?  

[Usman]

also, the issue with tutorial 1 is with the AMG. If you run the simulations on a model, imported from AMG, it shows an error.   

[Ren]

1. I'm not sure what you mean by "piston movement". Please explain.
2. Regarding tutorial 1, it uses the sector mesh generator (SMG). Are you saying that you created a geometry, set up the model using the data in the tutorial? I don't see why it should not work.

 

[Usman]

Regarding tutorial 1, I will give it another try in newer version of ANSYS.

Regarding piston movement, it is considered as a moving wall in Forte which moves TDC to BDC and vice versa. Is it possible to change the speed of the movement in Forte?

 

[Ren]

That's engine speed. You input this parameter (RPM) in the "Simulation Controls" task page.