Simulation of flow in Ellipta Inhaler

MFiX
1

ellipta-model_2025-12-03T143217.241030.zip (206.7 MB)

Sulaiman-et-al-2022-Ellipta-DPI.pdf (6.0 MB)

VTK-error-message
VTK-error-message886×690 89.1 KB

Error-message
Error-message975×440 28.8 KB

Warning-message
Warning-message969×424 37.1 KB

Hello,

I have set up a model to simulate single phase flow of air through the Ellipta Inhaler. I am following the attached paper. The model starts running but after sometime blows up. I have attached the Zipped file, the warning message, the error message and the stl file. I also get a vtk error message when starting. The mesh I am using is 80x120x80.

Kindly have a look at the files in the zipped folder and let me know how I can make it work. I feel it is probably due to mesh but needs some help and advice from you.

Thanks and Best Regards,

Sincerely,

Debashis

Ellipta-Inhaler-Truncated
Ellipta-Inhaler-Truncated692×807 26.2 KB

Ellipta_Mod0_30Nov_Ready_v1.stl (618.1 KB)

2

The inlet is Zmin and the pressure is 0 gauge (101325 Pa total pressure). The outlet gauge pressure is -4500 Pa (96,825 Pa total pressure).

Thanks and Regards,

Debashis

3

Mesh-80x200x100
Mesh-80x200x1002496×914 197 KB

New-Mesh-Parameters
New-Mesh-Parameters1523×613 86.7 KB

Hello,

The file and the mesh I uploaded yesterday had some discrepancy in the mesh. The mesh was generated with the default parameter setting in Meshing.

Today, I went ahead and increased the snap tolerance, decreased the intersection and dot product tolerances, increased the wall distance tolerance to get a better mesh. I increased the mesh resolution to 80x200x100. The new mesh did not show the deficiencies from the earlier mesh. This is the link with the Zipped files

The zipped file is located at

It is running. But the time step is always decreasing as shown in the figure attached. Also it is taking very small time step.

Please have a look at this file and let me know if the simulation parameters, boundary conditions look ok. I think it will again blow up with NaN.

I need help and advice on how to make this work.

Thanks and Best Regards,

Sincerely,

Debashis

Time-step
Time-step864×907 133 KB

4

Time-step-setting
Time-step-setting1511×498 73.5 KB

I have reduced the time step. But still it is blowing up after some time. Kindly let me know if the model looks all right with initial conditions and boundary conditions and other settings. I look forward to hearing from you.

Thanks and Best Regards,

Sincerely,

Debashis

5

  1. Please use xmin, xmax etc when you define regions instead of hard-coded numbers when referring to the edges of the box. I have mentioned this several times. Using hard coded numbers makes it tedious to review your settings, and your posts will likely not get any response.

  2. I don’t understand where are your inlet and outlet. The top-inflow-pressure-inletis located along the zmin plane and right -outflow-pressure-outlet is located at the ymax plane (gravity is along the negative y direction). Their type is also miss-assigned (XZ-plane for top-inflow-pressure-inlet and YZ-plane for right -outflow-pressure-outlet). You can use the predefined regions at the top of the table for easier setups:

image, and again, keep xmin, xmax etc whenever possible.

  1. Using a mass inlet may be more stable that using a pressure inlet.

  2. Always start with FOUP discretization. You can switch to Superbee once you are a running simulation.

  3. You can delete the mesh file before creating the zip file, this will make it smaller and easier to upload.

  4. You shouldn’t use the top-inflow-pressure-inletand right -outflow-pressure-outlet regions for vtk files. These are planes and won’t have any cell data in them. vtk regions should be volume regions.

6

Dr. Dietiker,

Thank you very much for your reply. I will rectify the mistakes, implement your suggestions and let you know. I will keep you posted.

Thanks and Best Regards,

Sincerely,

Debashis

7

ellipta-single-phase_2025-12-08T085929.117220.zip (108.1 MB)

Dr. Dietiker,

I was able to set up a mass inflow case, and it is running. So far so good. I am also monitoring to velocity and pressure. I am attaching the Zipped file and some figures. Kindly have a look at the set up and please let me know if everything looks all right. It is using 0.4 million grid points.Also suggest if you feel I need to change any settings or parameters. I am using FOUP discretization.

Thanks, and Best Regards,

Sincerely,

Debashis

BACKGROUND.csv (35.8 KB)

.

Pressure
Pressure884×909 77.3 KB

Gas-Velocity
Gas-Velocity884×911 82.5 KB

8

Hello,

Any feedback on the model setting and values used ? The single-phase simulation is running and I will incorporate particles. I am attaching two movie files. Kindly let me know if the settings look correct or if I need to change anything.

Thanks and Best Regards,

Debashis