CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > ANSYS > CFX

Simulation of a Venturi Nozzle in ANSYS CFX

Register Blogs Members List Search Today's Posts Mark Forums Read

Like Tree1Likes
  • 1 Post By ghorrocks

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   January 13, 2025, 02:47
Question Simulation of a Venturi Nozzle in ANSYS CFX
  #1
New Member
 
Lukas
Join Date: Jan 2025
Posts: 6
Rep Power: 2
Lugga is on a distinguished road
Hello everyone,
I am currently working on a project that I would like to simulate using ANSYS CFX 2024 R2 Student (Autodesk CFD was not detailed enough for my purposes). The project involves a circular Venturi nozzle (see cross-sectional view in the attachment), which I modeled in Inventor.
To the right of the nozzle, there is a wall that needs to be cooled.
So far, I have successfully simulated the airflow within the nozzle. However, I am struggling to simulate the impinging jet on the wall, as I keep encountering issues.
Challenges:

  1. Defining the Air Region (Flow Domain):
    How can I define the air region (flow volume outside the nozzle, i.e., the ambient air) in ANSYS if I haven’t done so in Inventor? Is there an easy way to do this, for example, in SpaceClaim or DesignModeler? Do I need to subtract the nozzle and the gap from the ambient air?
  2. Simulation Settings:
    What are the appropriate simulation settings for this type of problem (e.g., transient vs. steady-state, turbulence model, boundary conditions, mesh, inlets)?
  3. Inlet Definition:
    Should I define the nozzle outlet as the inlet for the simulation, or should I use the nozzle’s air supply line as the inlet?
  4. CFX vs Fluent:
    Is CFX the right software for this simulation, or would Fluent be more suitable?
  5. Ensuring Convergence and Realistic Results:
    How can I ensure that the simulation converges and provides realistic results?
  6. Wall Cooling Simulation:
    In the future, I would like to simulate the cooling of the wall. Is it possible to include this in CFX, and if so, how?
I am relatively new to CFX simulations, so I would greatly appreciate specific tips and guidance.
Thank you in advance for your support!
Best regards,
P.S.: This is my first post 😊
Lugga is offline   Reply With Quote

Old   January 13, 2025, 04:45
Default
  #2
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,901
Rep Power: 144
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
Nice post, and extra kudos for the animation

On your challenges:
1) Yes, you will need to extract the air volume. You can do this in most solid modelling packages, or you can do it in all the ANSYS solid modelling packages as well (Spaceclaim, DesignModeller etc). They all have a different way of doing it. DesignModeller is the one I know best and you can do it several ways, the easiest is probably using boolean operations (subtract the solids from a block to give the air domain).

2) That is not a simple question, and it depends on what you are trying model. Do some tutorials to get a feel for it, read the documentation on the various options and talk to your CFD guru at school/uni.

3) That depends on whether you want to include a model of the nozzle, or just assume the output of the nozzle. This links back into what you want to do and challenge 2.

4) They both can do it fine. Use the one you prefer. If you choose Fluent you should post on the Fluent forum, however.

5) Good question, and again there is not a simple answer. See https://www.cfd-online.com/Wiki/Ansy..._inaccurate.3F and https://www.cfd-online.com/Wiki/Ansy...publishable.3F

6) Yes. The model then includes a heat model. This is a reasonably simple addition to the fluids model - but note it does make accurate results more challenging as accurate heat transfer results are often difficult (just like accurate heat transfer measurements are difficult).
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.
ghorrocks is offline   Reply With Quote

Old   January 13, 2025, 09:25
Default
  #3
New Member
 
Lukas
Join Date: Jan 2025
Posts: 6
Rep Power: 2
Lugga is on a distinguished road
Thank you for the quick response!
Regarding your answers:
  1. I tried that, but unfortunately, in the meshing step, I can no longer define the inlet and outlet as surfaces because the ambient air is no longer a single body—it’s split into multiple parts (likely due to the subtraction). However, I’m not sure which body is which, as some appear to be invisible.
  2. I’ve tried looking into it, but unfortunately, no professor could help me.
  3. As mentioned, I’ve already simulated the flow inside the nozzle, so I would only need to simulate the external flow. However, if it’s easier to simulate both the internal and external flow together, I would prefer that.
  4. Then I’ll choose CFX, as the Fluent interface is quite difficult to get used to.
  5. Got it, thank you.
  6. Okay, but first I’ll focus on the air simulation.
Additional Questions:

- Does the inlet (and outlet) need to be inside the outer “box,” i.e., the ambient air domain?

- What boundary condition would you recommend for the walls? No-slip wall?


Thanks again!
Lugga is offline   Reply With Quote

Old   January 13, 2025, 18:09
Default
  #4
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,901
Rep Power: 144
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
1. This is basic solid modelling and Meshing. Have a look at the CFX tutorials on pre-processing for examples of how this is done. Note these are the tutorials from CFX-Mesh and other pre-processing software, not the CFX tutorials (as they do not cover pre-processing).

2. In that case do some tutorials to get a feel for it, read the documentation on the various options. If you have specific questions you can ask the forum, but the forum is not intended for broad questions like "how should I model a gizmo". A better forum question is "Should I use a mass flow or a pressure boundary at the inlet when I model a gizmo".

3. This really depends on the coupling between the two. If the downstream things do not affect what happens in the venturi then they can be separated and modelled as two separate models with simple boundary conditions. If the downstream things do affect the venturi then it is best to model them together in a single model.

"Does the inlet (and outlet) need to be inside the outer “box,” i.e., the ambient air domain?"

The inlet and outlet needs to be faces on the air domain so that those faces can be appropriate boundary conditions.

"What boundary condition would you recommend for the walls?"

The default wall (no slip) is appropriate for most cases.
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.
ghorrocks is offline   Reply With Quote

Old   January 13, 2025, 18:09
Default
  #5
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,901
Rep Power: 144
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
1. This is basic solid modelling and Meshing. Have a look at the CFX tutorials on pre-processing for examples of how this is done. Note these are the tutorials from CFX-Mesh and other pre-processing software, not the CFX tutorials (as they do not cover pre-processing).

2. In that case do some tutorials to get a feel for it, read the documentation on the various options. If you have specific questions you can ask the forum, but the forum is not intended for broad questions like "how should I model a gizmo". A better forum question is "Should I use a mass flow or a pressure boundary at the inlet when I model a gizmo".

3. This really depends on the coupling between the two. If the downstream things do not affect what happens in the venturi then they can be separated and modelled as two separate models with simple boundary conditions. If the downstream things do affect the venturi then it is best to model them together in a single model.

"Does the inlet (and outlet) need to be inside the outer “box,” i.e., the ambient air domain?"

The inlet and outlet needs to be faces on the air domain so that those faces can be appropriate boundary conditions.

"What boundary condition would you recommend for the walls?"

The default wall (no slip) is appropriate for most cases.
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.
ghorrocks is offline   Reply With Quote

Old   January 14, 2025, 03:10
Default
  #6
New Member
 
Lukas
Join Date: Jan 2025
Posts: 6
Rep Power: 2
Lugga is on a distinguished road
I’ve already gone through many tutorials but couldn’t make progress, which is why I posted in the forum. I’ll try looking into it again—thank you!


3. Okay, thank you. To account for any potential stagnation pressure effects, I’ll simulate both the internal and external flow together.


I understand that the inlet and outlet need to be faces of the air domain. However, in DesignModeler, you can use the "Enclosure" function, which needs to encompass the entire geometry. My question is whether this is a good approach, or if it would be better to leave the inlet of the nozzle outside the ambient air domain (see attachment).


I’ll give the simulation another try later this week and will update you. Thanks again!
Attached Images
File Type: jpg Inlet.jpg (24.3 KB, 7 views)
Lugga is offline   Reply With Quote

Old   January 14, 2025, 03:55
Default
  #7
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,901
Rep Power: 144
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
It is hard to recommend anything because I do not know what you are trying to do.

Is the main flow through the venturi driven by something? Is the only thing driving the flow through the venturi the entrained flow from the compressed air?

If yes, then I would model this entirely inside the domain, with a boundary condition for the compressed air and an external boundary for the atmosphere (opening, pressure=0), Note that you are going to need a fine mesh to resolve the Coanda effect on the compressed air. Also note that this model appears to be 2D axisymmetric - which would dramatically simplify the model.
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.
ghorrocks is offline   Reply With Quote

Old   January 14, 2025, 04:43
Default
  #8
New Member
 
Lukas
Join Date: Jan 2025
Posts: 6
Rep Power: 2
Lugga is on a distinguished road
I want to simulate the impinging jet on the wall and, in a second step, analyze the heat transfer. My goal is to evaluate the nozzle's efficiency by determining the ratio of entrained air to injected air.


The main flow is driven solely by compressed air. 5 m³/h of air are injected into the nozzle (via an inlet pipe), and the air is expelled through the nozzle. Due to the Venturi effect, additional air should be entrained.


What do you mean by "inside the domain"? Are you referring to the right-hand "drawing" in my image?


If I were to design three inlet pipes and cut through one of them, would it still be considered axisymmetric?


Thank you so much for taking the time to help me!
Lugga is offline   Reply With Quote

Old   January 14, 2025, 05:14
Default
  #9
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,901
Rep Power: 144
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
Your left image looks definitely wrong, and the right one is sort-of wrong as well. The issue is that you do not need to have an inlet boundary for the venturi. It should just be connected to the surrounding air domain so the air can be drawn in as it is entrained. The only active boundary condition I am aware of is the compressed air which can be modelled as a inlet with a defined mass/volume flow rate. There is also the atmospheric boundary around the whole domain, it should be an opening at zero pressure.

A 2D axisymmetric model means you can take the 2D slice you are modelling and it represents the 3D volume generated by revolving that slice around the axis. So a single venturi looks 2D axisymmetric from what I have seen, but an array of 3 venturis does not sound 2D axisymmetric. 2D models are MUCH easier, so definitely start with that one.
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.
ghorrocks is offline   Reply With Quote

Old   January 14, 2025, 05:46
Default
  #10
New Member
 
Lukas
Join Date: Jan 2025
Posts: 6
Rep Power: 2
Lugga is on a distinguished road
I think we might be misunderstanding each other a bit.
The nozzle is supplied with compressed air, just like a standard nozzle. For this, I do need an inlet. Inside the nozzle, the injected air flows along the wall of the nozzle and is supposed to entrain ambient air. The pipe supplying the compressed air needs to be far enough away from the nozzle outlet to avoid influencing the simulation.
However, if I model the inlet as shown on the right in the sketch, won’t ambient air also get entrained into the compressed air?
Lugga is offline   Reply With Quote

Old   January 14, 2025, 05:51
Default
  #11
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,901
Rep Power: 144
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
Yes, I do not think we are talking about the same thing. How about you model up what you want to use and we discuss that. Even if you are not sure how to model it, just have a guess and we can go from there.
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.
ghorrocks is offline   Reply With Quote

Old   January 14, 2025, 05:56
Default
  #12
New Member
 
Lukas
Join Date: Jan 2025
Posts: 6
Rep Power: 2
Lugga is on a distinguished road
Okay, I'll get back to you here then, right?
Lugga is offline   Reply With Quote

Old   January 14, 2025, 06:00
Default
  #13
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,901
Rep Power: 144
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
Sure. Whatever you think will show what you are trying to do best.
Opaque likes this.
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.
ghorrocks is offline   Reply With Quote

Reply

Tags
ansys, cfd, cfx, nozzle analysis, venturi

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
ANSYS CFX Setup for Aerospike Nozzle LCash CFX 3 January 9, 2019 18:11
How is ANSYS CFX is related to RANS based Simulation Ramcad ANSYS 0 July 17, 2014 12:40
Supersonic Nozzle Exhaust Simulation mikeh FLUENT 0 May 1, 2014 22:28
simulation of a radial hydrodynamic bearing in ANSYS CFX Jalen ANSYS 1 October 4, 2013 11:40
Verifying results for a wind turbine blade simulation in ANSYS CFX Joystix CFX 3 April 27, 2012 18:52


All times are GMT -4. The time now is 02:09.