Convergent Divergent Rocket Nozzle Simulation Problem!

KruX · July 22, 2020, 07:57

Hi @a7medalsalmi,

I am facing similar problems simulating an aerospike nozzle. Could you maybe elaborate a little bit more about your solution? Do you use Pressure Field as the domain boundary? Did you use the FMG-Initialization? Maybe you could share some picture of your mesh and its quality. I would really appreciate it, because I am running out of time for my thesis and nothing seems to work.

Kind regards,
Roman

a7medalsalmi · July 22, 2020, 09:04

I used pressure farfield for the atmospheric domain. I ended up initializing from Inlet. I would share some of the images maybe later on as I currently do not have direct access to my project folder. You should really understand the flow problem you're tackling. For instance, some of the pressure ratios that I have inputted caused an unsteady flow through the nozzle which I assume is the cause of the errors I encountered due to the axis condition that I applied. Creating a planar nozzle rather an axisymmetric one could give you some ideas on what issues you might be encountering. You could decrease the atmospheric domain size or lower the number of elements to test it and get some intial values at least. Hope this helps for now.

KruX · July 22, 2020, 09:20

Thank you for your quick response! I am designing an aerospike nozzle for a sounding rocket with a comercial solid-propellant motor, so my Inlet conditions are defined through the combustion chamber. But I have read that the axissymmetric solver can cause problems, so I am going do try a planar model of my nozzle. Did you end up using the planar model or did you just use it for initial values?
One more question regarding your advice for the Hydraulic Diameter: Did you just used the height of your inlet in the 2D model or should it be 2 times the height in the axissymmetric case?

Many thanks for your efforts!

KruX · July 24, 2020, 10:17

Hi a7medalsalmi,

I really don't want to bother you, but your advice would be very helpfull for me. If you find the time, it would be really nice if you could send me some pictures of your mesh/settings.
I have one more quation regarding the Pressure Farfield: Did you apply the same Mach number to all Pressure-Farfields? This was suggested to my by another member of cfd-online but I am not sure if its correct since the axial free stream Mach number is not equal to the radial free stream Mach number.

Thank you for your time!

a7medalsalmi · July 27, 2020, 07:15

Quote:

Originally Posted by KruX

Thank you for your quick response! I am designing an aerospike nozzle for a sounding rocket with a comercial solid-propellant motor, so my Inlet conditions are defined through the combustion chamber. But I have read that the axissymmetric solver can cause problems, so I am going do try a planar model of my nozzle. Did you end up using the planar model or did you just use it for initial values?
One more question regarding your advice for the Hydraulic Diameter: Did you just used the height of your inlet in the 2D model or should it be 2 times the height in the axissymmetric case?

Many thanks for your efforts!

I tried a planar model of the nozzle to see if I would face similar issues in which then I realized that some of the pressure values I have been inputting have caused the flow to be assymetrical. You can learn more about that when researching "Free shock separation". The planar model took a lot of time for the solution to converge thus I prefered using the axisymmetric model in which the solutions above fixed my problem. Only used the planar model for some initial results. For the hydraulic diameter I used the specified diameter from the experiment for the axiysmmetric model.

a7medalsalmi · July 27, 2020, 07:23

Quote:

Originally Posted by KruX

Hi a7medalsalmi,

I really don't want to bother you, but your advice would be very helpfull for me. If you find the time, it would be really nice if you could send me some pictures of your mesh/settings.
I have one more quation regarding the Pressure Farfield: Did you apply the same Mach number to all Pressure-Farfields? This was suggested to my by another member of cfd-online but I am not sure if its correct since the axial free stream Mach number is not equal to the radial free stream Mach number.

Thank you for your time!

Separated Flow (Vel Mag): https://ibb.co/4Rhc5qf

Mesh (all): https://ibb.co/DGH09QH

Mesh (zoomed): https://ibb.co/g7dC0yz

Boundary Conditions: Pinlet, walls (for nozzle), axis, and pressure farfield for the rest.

Yes I did apply the same Mach Number for all pressure farfields.

KruX · July 27, 2020, 07:35

Thank you very much! This helps me a lot. What is your max. AR? If I refine the mesh at the boundary layer for a y+=1 my AR is quiet high in this regions.

Best regards,
Roman

KruX · July 29, 2020, 05:09

Hi,

so my simulation is now converging when I assume the upper boundary as a wall but the results are somewhat strange. When I assume a pressure farfield for all the boundary of the domain (as you successfully did) my simulation is diverging really fast. Did you specify the flow direction of the pressure farfield, for example with an axial component of -1 for the outlet? Do I have to apply a radial component for horizontal boundarys?

Kind regards,
Roman

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
shocks in convergent divergent nozzle, convergence issue	yogeshghadge314@gmail.com	Main CFD Forum	19	January 2, 2020 17:54
Inverse Design Optimization	khavart	SU2 Shape Design	0	June 20, 2019 04:37
compressible, rhoSimpleFoam, multi-species, steady state, rocket nozzle	David_C	OpenFOAM Running, Solving & CFD	1	April 18, 2017 12:01
mass flow rate issue in supersonic nozzle simulation	xkang	FLUENT	0	July 31, 2014 17:06
Supersonic Nozzle Exhaust Simulation	mikeh	FLUENT	0	May 1, 2014 22:28

July 22, 2020, 07:57		#21
KruX New Member Roman Krückel Join Date: Jul 2020 Posts: 17 Rep Power: 6	Hi @a7medalsalmi, I am facing similar problems simulating an aerospike nozzle. Could you maybe elaborate a little bit more about your solution? Do you use Pressure Field as the domain boundary? Did you use the FMG-Initialization? Maybe you could share some picture of your mesh and its quality. I would really appreciate it, because I am running out of time for my thesis and nothing seems to work. Kind regards, Roman

July 22, 2020, 09:04		#22
a7medalsalmi New Member Join Date: Feb 2020 Posts: 13 Rep Power: 6	I used pressure farfield for the atmospheric domain. I ended up initializing from Inlet. I would share some of the images maybe later on as I currently do not have direct access to my project folder. You should really understand the flow problem you're tackling. For instance, some of the pressure ratios that I have inputted caused an unsteady flow through the nozzle which I assume is the cause of the errors I encountered due to the axis condition that I applied. Creating a planar nozzle rather an axisymmetric one could give you some ideas on what issues you might be encountering. You could decrease the atmospheric domain size or lower the number of elements to test it and get some intial values at least. Hope this helps for now.

July 22, 2020, 09:20		#23
KruX New Member Roman Krückel Join Date: Jul 2020 Posts: 17 Rep Power: 6	Thank you for your quick response! I am designing an aerospike nozzle for a sounding rocket with a comercial solid-propellant motor, so my Inlet conditions are defined through the combustion chamber. But I have read that the axissymmetric solver can cause problems, so I am going do try a planar model of my nozzle. Did you end up using the planar model or did you just use it for initial values? One more question regarding your advice for the Hydraulic Diameter: Did you just used the height of your inlet in the 2D model or should it be 2 times the height in the axissymmetric case? Many thanks for your efforts!

July 24, 2020, 10:17		#24
KruX New Member Roman Krückel Join Date: Jul 2020 Posts: 17 Rep Power: 6	Hi a7medalsalmi, I really don't want to bother you, but your advice would be very helpfull for me. If you find the time, it would be really nice if you could send me some pictures of your mesh/settings. I have one more quation regarding the Pressure Farfield: Did you apply the same Mach number to all Pressure-Farfields? This was suggested to my by another member of cfd-online but I am not sure if its correct since the axial free stream Mach number is not equal to the radial free stream Mach number. Thank you for your time!

July 27, 2020, 07:35		#27
KruX New Member Roman Krückel Join Date: Jul 2020 Posts: 17 Rep Power: 6	Thank you very much! This helps me a lot. What is your max. AR? If I refine the mesh at the boundary layer for a y+=1 my AR is quiet high in this regions. Best regards, Roman

July 29, 2020, 05:09		#28
KruX New Member Roman Krückel Join Date: Jul 2020 Posts: 17 Rep Power: 6	Hi, so my simulation is now converging when I assume the upper boundary as a wall but the results are somewhat strange. When I assume a pressure farfield for all the boundary of the domain (as you successfully did) my simulation is diverging really fast. Did you specify the flow direction of the pressure farfield, for example with an axial component of -1 for the outlet? Do I have to apply a radial component for horizontal boundarys? Kind regards, Roman