[MFGT]

Hello,

i have a couple of questions, i hope its fine to collect them here in one thread:

1.) Does the Solver get efficient with newer releases? E.g. i am using the 2.3.10 version now, and update to the newest as soon as possible, because there are always some nice added features. Are there changes in results? I remember a comparison I found for major updates (2.2 vs 2.3 for example).

2.) Regarding Injection, i sometimes have issues using the spray rate calculator. In general, i set it up with injection time (CA or sec), injection mass and injection pressure. Then i copy the resulting discharge coefficent and enter it in the corresponding box. I do this, because Studio often loses the entered values in the additional settings (fuel pressure).

Which discharge coeff. model is recommended?
Sometimes i receive error message: Warning, area correction factor > 1.0, setting = 1.0 (velocity_coeff = 0.76914) when using 1 (using correlation for cv).

3.) For Gasoline DI and PFI, i am using the Kuhnke wall film model with default inputs. How are the experiences there or what is best practice?

4.) The optical representation of the fuel in EnSight has a lot of options. Are there recommended settings to use? E.g. i use sphere with size by dp_radius with a value of 5. And reduced polygons of 0-2. Otherwise i have too many dots and chaos on screen.

edit:
5.) In my recent simulations i experienced a strange behaviour with the reaction ratio. If combustion starts, it should decreasae, because O2 is consumed. However, it does exactly the opposite and i can verify in EnSight, that its rising at the location, where my fuel and O2 is consumed (actually both go to zero). The reaction ratio is the same as equivalence ratio, only without considering CO2 and H2O, right?

6.) Can i ignore the following warning which i receive with default inputs? WARNING: in inputs.in mult_dt_coll_mesh is greater than 1.0. this may lead to large errors in spray calculations


Thanks a lot.
Tobias

[MFGT]

Any news, feedback?

[alemoine]

Originally Posted by MFGT

1.) Does the Solver get efficient with newer releases? E.g. i am using the 2.3.10 version now, and update to the newest as soon as possible, because there are always some nice added features. Are there changes in results? I remember a comparison I found for major updates (2.2 vs 2.3 for example).

Answer:
With each new CONVERGE release, the solver is optimized in terms of features and/or performance. With these optimizations, the behavior of the code can be different in certain cases when run in high fidelity. However, these variations should be minimal if the simulated system is in a stable mode of operation. It is a good practice to assess the variability brought in by the version upgrade before completely switching to the new version.

2.) Regarding Injection, i sometimes have issues using the spray rate calculator. In general, i set it up with injection time (CA or sec), injection mass and injection pressure. Then i copy the resulting discharge coefficent and enter it in the corresponding box. I do this, because Studio often loses the entered values in the additional settings (fuel pressure).

Which discharge coeff. model is recommended?
Sometimes i receive error message: Warning, area correction factor > 1.0, setting = 1.0 (velocity_coeff = 0.76914) when using 1 (using correlation for cv).

Answer:
After clicking "Calculate rate", the value for discharge coefficient will display in the log. If you go up to the discharge coefficient dialog box, enter in the new value (based on what is visible in the log), and click on the "Update sources" button, there will be a prompt asking you if you would like to change the settings in your Injector setup. By clicking yes, you will find that the discharge coefficient is updated in the Injector settings.

When you receive the Warning message, this means that one of the inputs for the calculator may be incorrect. Double check the Max. target pressure (units of bar), the injection duration, and mass to verify they are the values you are expecting. There is no recommendation for discharge coefficient model. The "cv=1" is simpler in that it sets the contraction coefficient to be the same as the discharge coefficient whereas the "use correlation for cv" required dynamic updating during run time.

3.) For Gasoline DI and PFI, i am using the Kuhnke wall film model with default inputs. How are the experiences there or what is best practice?

Answer:
Kunhke model is not our first choice for typical GDI and Gasoline PFI cases. Please refer to our tutorials for recommended settings for these type of engines. This does not mean that Kunhke model is not recommended for use in Gasoline engine simulations, but rather we see more usage of this model in Urea/SCR system modeling for engine after treatment.

4.) The optical representation of the fuel in EnSight has a lot of options. Are there recommended settings to use? E.g. i use sphere with size by dp_radius with a value of 5. And reduced polygons of 0-2. Otherwise i have too many dots and chaos on screen.

Post-processing in EnSight is really case-dependent. As such, we do not have any recommended settings for post processing as it usually comes down to personal preference of visualization.

5.) In my recent simulations i experienced a strange behaviour with the reaction ratio. If combustion starts, it should decreasae, because O2 is consumed. However, it does exactly the opposite and i can verify in EnSight, that its rising at the location, where my fuel and O2 is consumed (actually both go to zero). The reaction ratio is the same as equivalence ratio, only without considering CO2 and H2O, right?

Answer:
Your understanding of Reaction ratio definition as used in CONVERGE is correct. We will have to take a look at your case to figure out why you do not get the expected behavior with reaction ratio output. Please send your case file (*.in and *.dat) to support@convergecfd.com and we will get in touch with you.

6.) Can i ignore the following warning which i receive with default inputs? WARNING: in inputs.in mult_dt_coll_mesh is greater than 1.0. this may lead to large errors in spray calculations

If you do not want to run with the collision mesh for spray, then you can ignore this warning. Our typically recommended settings always set the Collision grid time-step multiple to a very high number so as to deactivate the model.

Thank you,