[infrared]

Hello, I performed a diesel combustion case with the following configuration:

- Cylindrical domain with impingement on the wall, where the inlet nozzle is placed vertically at the center of the top wall.
- The boundary conditions are initialized as top_wall, bottom_wall, inlet, and outlet.
- The temperatures were set as follows: wall temperature at 500K, vessel temperature at 1000K, and diesel temperature at 338K.
- Vessel pressure is 4 MPa.
- Oxygen is 21% of the air.
- The domain has a diameter of 40 mm and a height of 20 mm, with the inlet diameter being 1 mm.
- The mesh was set to 200x12x64 cells, with a denser configuration near the bottom wall.
- The combustion model is PaSR, using the LES turbulence model and the reactingFoam solver.
- Diesel is continuously injected into the domain.
- The file structure for this case is as follows:
/0 directory: contains alphat, mass fractions, p, T, U, Ydefault.
/constant directory: contains polyMesh, chemistryProperties, combustionProperties, reactions, thermo.compressibleGas, thermophysicalProperties, and turbulenceProperties.
/system directory: contains blockMeshDict, controlDict, FOBilgerMixtureFraction, fvSchemes, and fvSolution.

The problem is that I cannot see any flame in the temperature distribution from this combustion case, as shown in the attachment.

T70.jpg

Is there any issue with the setup I’ve provided above? Any suggestions and comments would be greatly appreciated. Thank you for your kindness.

Best regards,
Ir

[GeekCFD]

Hi Rohmawati,
Can you please share the case setup, particularly the constant and system directories?
It is otherwise difficult to comment on the problem if one doesn't have access to the files.

-------------------------
Thanks,
GeekCFD

[infrared]

Originally Posted by GeekCFD

Hi Rohmawati,
Can you please share the case setup, particularly the constant and system directories?
It is otherwise difficult to comment on the problem if one doesn't have access to the files.

-------------------------
Thanks,
GeekCFD

Hi @GeekCFD,

Thank you for your response. After carefully checking the case, I found that the problem was that I forgot to define the O2 amount (0.21) in the 0 directory. I successfully ran the case using biodiesel (C17H34O2) as the fuel, but I couldn't observe any flame for the case using diesel (C7H16 or C12H26) as the inlet fuel.

The setups were similar, as I ran them in a constant volume with the fluid impinging on the bottom wall. Here is the file structure of the case:

*0 directory*: alphat, C7H16 or C12H26 or C17H34O2 (depending on the fuel variation), CO2, H2O, k, N2, nut, O2, p, rho, T, U, Ydefault.

*constant directory*: polyMesh, chemistryProperties, combustionProperties (PaSR), reactions (complete and incomplete combustion), thermo.compressibleGas, thermophysicalProperties, transportProperties, turbulenceProperties (LES).

*system directory*: blockMeshDict, controlDict, FOBilgerMixtureFraction, fvSchemes, fvSolution.

Here is the temperature comparison of diesel and biodiesel at the same time, but no flame was visible for the diesel case.

Your response and suggestions are greatly appreciated. Thank you in advance.

[GeekCFD]

Hi Rohmawati,
Sorry for the late reply. I was quite busy last month.

I had a look at your case, and I think there are a few things that could be causing the issue:

1. Reaction Mechanism and Species Properties:
Make sure to update the 'reactions' file and the species properties in the 'thermophysicalProperties' file when switching from biodiesel to diesel. These files might still have values which were originally set for biodiesel. Plus, I’d recommend using a more detailed mechanism like GRI or a similar setup to ensure an accurate representation of diesel combustion.

2. Initial and Boundary Conditions (Possible Major Issue):
Double-check that the initial and boundary conditions are realistic for diesel combustion. For instance, the initial temperature near the injection zone might need to be higher due to diesel's lower volatility compared to biodiesel. Sometimes, the initial conditions might be such that no significant heat generation or fuel spread occurs, which would prevent combustion from initiating.


Please take a look at these points and see if they help resolve the issue. If possible, please share the relevant files from your case setup (like the constant and system directories) so I can better debug the problem.

Let me know if you need any further assistance.


---------------------------
Best regards,
GeekCFD

[infrared]

Originally Posted by GeekCFD

Hi Rohmawati,
Sorry for the late reply. I was quite busy last month.

I had a look at your case, and I think there are a few things that could be causing the issue:

1. Reaction Mechanism and Species Properties:
Make sure to update the 'reactions' file and the species properties in the 'thermophysicalProperties' file when switching from biodiesel to diesel. These files might still have values which were originally set for biodiesel. Plus, I’d recommend using a more detailed mechanism like GRI or a similar setup to ensure an accurate representation of diesel combustion.

2. Initial and Boundary Conditions (Possible Major Issue):
Double-check that the initial and boundary conditions are realistic for diesel combustion. For instance, the initial temperature near the injection zone might need to be higher due to diesel's lower volatility compared to biodiesel. Sometimes, the initial conditions might be such that no significant heat generation or fuel spread occurs, which would prevent combustion from initiating.


Please take a look at these points and see if they help resolve the issue. If possible, please share the relevant files from your case setup (like the constant and system directories) so I can better debug the problem.

Let me know if you need any further assistance.


---------------------------
Best regards,
GeekCFD

Dear @GeekCFD,

Thank you for your responses. I apologize for the delayed reply. Let me address the points you mentioned:

1. I have already investigated both cases (biodiesel and diesel). There is nothing wrong with the `reactions` and `thermophysicalProperties` files for either biodiesel or diesel.

2. I am actually confused about the initial condition for `T`. I am using a cylindrical domain where the inlet diameter is D, the height is 20D, and the diameter is 40D. The fuel is injected through the inlet, which is located at the center of the top_wall boundary. The boundary conditions are set as follows: inlet, top_wall, bottom_wall, and outlet. Here are the initial conditions for `T`:

dimensions [0 0 0 1 0 0 0];

internalField uniform 900;

boundaryField
{
bottom_wall
{
type fixedValue;
value uniform 500;
}
inlet
{
type fixedValue;
value uniform 338;
}
top_wall
{
type fixedValue;
value uniform 500;
}
outlet
{
type zeroGradient;
}
}


For the inlet, I set `T` to 338K. Is this acceptable, or should I set it to match the internalField (900K)?

3. Recently, I used `setFieldsDict`, thinking it would help ignite the diesel more quickly. Here is the content of my `setFieldsDict` file:


defaultFieldValues
(
volScalarFieldValue T 900
);

regions
(
cylinderToCell
{
p1 (0 0 0);
p2 (0 0 0.01);
radius 0.0005;

fieldValues
(
volScalarFieldValue T 1100
);
}
);
```

Do you have any thoughts on this approach?

4. In the `constant` directory, I have the following files: `chemistryProperties`, `combustionProperties` (PaSR), `reactions`, `thermo.compressibleGas`, `thermophysicalProperties`, `transportProperties`, and `turbulenceProperties` (LES). I initially thought that in real engine conditions, the ratio of fuel to O2 is about 1:18, but I used a mass fraction of 1:1 in the species definitions in the `0` directory (rich mixture). In the `reactions` file, the ratio of diesel to O2 for complete combustion is already set to 1:18. Could this be one of the issues? For biodiesel, I also use a rich mixture scheme.

5. In the `system` directory, I have the following files: `blockMeshDict`, `controlDict`, `FOBilgerMixtureFraction`, `fvSchemes`, `fvSolution`, and `setFieldsDict`.

Based on your suggestions, I added the GRI mechanism as well. The case is currently running. I will provide you with an update once it is completed.

Your responses and suggestions are greatly appreciated. Thank you in advance.

Best regards,
Ir