ANSYS Fluent NACA 4412 (or NACA 0012) 2D airfoil CFD Tutorial with Experimental Validation (2021)
ฝัง
- เผยแพร่เมื่อ 3 เม.ย. 2021
- Here's an example of the a CFD of a non-symmetric airfoil such as the NACA 4412. These exact techniques can be used on a symmetric airfoil such as NACA 0012 or NACA 0018. It uses nearly the same techniques I show in my NACA 0012 Video found on my Udemy - www.udemy.com/course/learning... . There's generally a lot of discussion to be had as non-symmetric airfoils have more complicated flow patterns compared to symmetric but I show aspects of
- ANSYS Design Modeler
- ANSYS Mesher
- ANSYS Fluent
- General Analysis
I do not provide free homework help or troubleshooting of any kind beyond TH-cam comments. However, if you are looking to get a bit of input or help on a project you can buy me a coffee and I'll respond: www.buymeacoffee.com/anthonytg
For business and research inquiries, I can be reached at anthonychbtl@gmail.com.
Here's a link to the Udemy: www.udemy.com/course/learning... - วิทยาศาสตร์และเทคโนโลยี
![[CFD] Inflation Layers / Prism Layers in CFD](http://i.ytimg.com/vi/1gSHN99I7L4/mqdefault.jpg)
I never comment on TH-cam videos but I just spent 12 hours trying to figure this out and this is the first video that actually helped
I wasted 5 days and then magically god guided me to this video
I just want to say that I have been using this and your other videos Anthony to do the CFD portion of my Masters Thesis for months now. I would be completely lost using Fluent without this video so thank you very much for being an amazing teacher.
Hi What is the wing length and wingspan for the naca 4412, how can I find it?
@@zln9384see it in airfoil database
Hands-down best ANSYS tutorial I've found on TH-cam
This is amazing !!
Please keep your channel active!
Thank you Anthony for this amazing tutorial and thank you for covering some small-small but really important topics.
this is one of the best tutorials in cfd i have ever seen!
Thanks Anthony, this helped me a lot!
Hey I tried meshing just using edge sizing however, my first layer thickness is too high. I want to use inflation but I cannot use it for more than 1 edge at a time. Any idea how to fix it? I have removed face meshing altogether to facilitate inflation
dude you just saved my endproject for school thanx man
Do you have any suggestions how to improve the mesh for higher angles of attack? At 12 degrees I get a quite large inaccuracy between 10-30%. I tried more Divisions and a higher bias but this just leads to fluent not converging the results and the cl and cd values just go up and down in the same pattern for 1000+ iterations. I know you mentioned papers where an inaccuracy of 20-30% is still considered ok, maybe you could cite them here because I wasn’t able to find them. Thank you very much.
I am facing a problem in meshing my own airfoil , the Bias does not seem to work and I am not getting a good enough mesh to validate the results. I am not able to understand where I am going wrong . Do you have any suggestions of how to diagnose this
can I use this to figure out if/when whirlpooling/stall occurs? I tried just changing the angle of inlet wind and looking at the path-lines but I cant exactly tell.
Hey i tried the meshing you showed on an airfoil with coordinates. But k-epsilon method is diverging where the turbulence viscosity ratio is going higher than 10^5. Whereas when I am using the spallart almaras there are no messages like that.
Hello, When I use boolean my smooth airfoil edges is changed to sharp edges, is this just a visual problem or is it something wrong with my geometry? I imported airfoil as a surface from catia in stp format( I had to do some modifications that I cannot imply in Ansys Design Moduler). How can I fix that? Thanks
Hey Anthony, thanks for the video, it has helped a lot although ive got some issues using it, there appears to be some faces on my mesh domain to be unstructured, not quite sure how to fix this any suggestions?, I have followed the steps in this video for the NACA 2412 aerofoil.
Thank you. Amazing video, clear, useful and professional.
Hello Pro,
I have some questions related to the ansys workbench, it is multiple choices and take few minutes, so could you help me please?
Hi Anthony. I am really grateful for this video, but I really need your help concerning something
so i ran this test using a different airfoil but using the same method, but my pressure coefficient and my wall Y plus value are way above 1. Please how do I rectify this?
23:38 for inlet velocity! How can we do it through UDF(user defined function) ? What formula/expression/equation for velocity has to be used? Could you direct me towards - how can this be done in UDF!
In Messing , I did the same as in your video , but I get the mesh of the two rectangles is very bad , it's not like in your video , I guess it has a problem at edge sizing 1 . can you help me fix that error?
doesn't the experimental paper use reynold's number of 1500000, and the lift and drag values at the end of the video from NASA website, aren't they for 1.52e6 reynold's? please clarify if I am missing something sir.
Excellent tutorial man.
does this kind of mesh work if you try to change angle of atack?
I´m trying to simulate an airfoil for different Angles of Attack, but I´m having trouble with it´s mesh.
Thank you so much for the great explanation, I would like to ask, how can I export a plot as x or y data not XY?
Can you help me complete this? Something is going wrong with my mesh on the underside of the airfoil and the fluid doesnt seem to be sticking to the airfoil in that section
What is the procedure of testing surface roughness of airfoil in ansys?
great video! A question: I can use these steps for different airfoils at different velocities and angle of attacks right? Also how would you adjust the Yplus to be have the first layer thin enough?
This video was informative,but the project I am currently working on has a different shape which is creating some issues for me while meshing,it is having a flat upper surface and a naca 0012 profile as the lower surface,that is it is like the naca 0012 profile exactly splitted along the chord,so will you please suggest me a domain and a how to mesh for this profile..
Hi i have one question. Since we are doing a 2d simulation are we actually calculating cl or Cl?
⚠️Please, i have some questions... in fact, i made the fluid simulation around NACA, i followed all the steps but the question i have is: - - - where i can change the velocity (Inlet) the pressure (outlet) and also the angle of attack, i can't find their boxes?
- do I continue to work on Fluent or should I switch to CFX?
-If I did not modify anything, that is to say I worked with default parameters, is it so?
Thanks in advance
I followed the same steps as in the video but my mesh failed, showing the message : “An error occurred on the meshing”, due to computing power or something like that. I have tried to truce the number of divisions quite a lot (maximum 50) and is still not working. Any advice?
I repeated the geometry and it allows me to mesh the structure, but it results in an asymmetrical meshing. I have changed the bias factor, created multiple different size Edges and still get similar results, all of them asymmetrical.
When generating the mesh, I get the projection lines showing. Why is that, and how do I get rid of them? thanks
Is the recent iris xe graphics with the 11th gen i5 16gb ram compatible and sufficient with ansys for low medium 3d simulations?
İf i want to make a graph of cl/angle of attack, can i do it by using ansys and Excel ? Or is using xflr5 more logical?
Hello bro , how to take pictures of pressure contour with varying angle of attack?
I did everything as follows but my txt file is not seen after generating in ansys. It is there somewhere but no matter how much i zoom in or out it is not visible( i opened space claim and it is visible dont know why it is invisible in design modeler)
Hi Anthony, your video tutorial has been really useful. However, could you please explain to me what is that "wall-surface_body" we see underneath the airfoil in wall boundary conditions? why is it present? does it affect the simulation results?
I am getting failed mesh errors, it's saying at least one lunch control has a tolerance index below one or several scoped size control.
i wanna validate a paper which is about studying flow over an airfoil considering surface roughness .. my lift co is much lower that what it should be :(. i don't know why..
can this simulator calculate static pressures at certain locations?
hi bro. when building geometry, you say that there's studies that show this height V1 and V2 should be at least 10 times. Can i have link to this studies?
In the naca file I transferred to ansys, error reading coordinates file.. is coming, I saved the file like you, but the error could not be resolved, can you please help me?
off topic, anyone know how to fix this error? "Unable to spawn node: license not available.
ANSYS LICENSE MANAGER ERROR:Not enough HPC Parallel licenses 6/4." i am using ansys 2022 r1 student version
I generated the same mesh and I tried to run model for different velocities but I got the same drag and lift forces. They should be some problem with domain geometry
Great vid, but how you apply a different angle of attack for moment coefficient since its on the XYZ axis
Hi! In what video have you explained the Y+? :)
First of all, thank you for sharing your work. But I have a question. Even though I followed the steps exactly in the video, my Cl coefficient was calculated as 1.56. There is no problem with my CD coefficient. What do you think could be the reason for this? I would be very happy if you answer.
I am currently working on a project but after a number of analysis i am not getting the value as reference can you please help???
Thanks a lot you saved hours of my time!
Hello sir.
Do you have any course on phase change flow through evaporator and condenser sir?
I've been trying this on 3D NACA0012 aerofoil with SERRATIONS for the past week keeps failing
my mesh fails to generate, im using the student version 2023 R1. Could someone please help?
Please i want an answer, how to calculate the mass averaged total pressure loss, using fluent or tecplot, please anyone answers me ?!
Amazing tutorial helped me simulate my first airfoil NACA 2312!
One thing I found useful trying to learn Fluent is it is good to re-initialize before calculation because it will reset your iterations back to 0.
I'm new to CFD and Fluent, what's the mean of re-initialize before calculation? Isn't it only needed one initialize? thanks
@@bmw009a3 re-initializing gives the initial solution / guess needed to do any iterative method to solve any differential equation. It should be irrelevant to any final solution, however if you have a bad initial solution you will run into convergence issue. You need to do it before a new simulation but if you do one simulation after another with just slight changes you can use the solution from the previous solution as your initial condition, assuming that previous simulation is not very different than your new one.
I couldn’t get a straight line of a dark mesh starting from trailing edge. My dark mesh straight line spreads towards the top right corner of the rectangular domain starting from trailing edge. I followed every steps shown on your video. It could be better if I get to contact you since I am doing my project on effect of vortex generator on wing using PIV technique. May be you can help me computationally on my project. Your help is much appreciated.🙂
can anyone gudie me why my Y-axis values are so large while running calculations
I've found that when creating edge sizing #2 using a bias factor of 50 creates a vetter aspect ratio transition between the 2 regions
I find your video highly educative. I followed steps (simulation for 23012) but the mesh in bottom C doesn't want to work. Shows triangles... Please somebody help ::(((
hey hi, i am getting floating point execption for this case, can anyone help
i've just get the right curve and i'm so happy
ah, no, it's not right
oh, right finally. Tnx for the vid, i mean
can you please share the excel file linl to genrate airfoil
bro can you send the link of the website from where you got the cp data
thanks a lot!! is there any way to ask questions about flunet ?? if you dont mind
Do you have any suggestions for how to split the fluid domain and do the sizing for an aerofoil with a Fowler flap?
it depends on the geometry you can do a structured mesh with the right considerations or you can always do unstructured with inflation layers
Thank you for such an in-depth video on the topic!
I have a similar setup with a leading-edge rotating cylinder.
Using K-Omega SST model, I'm getting acceptable results for zero angle of attack.
But in case of 13 Degree AOA, the solution doesn't converge. There's a periodic pattern on the lift and drag plots. Any suggestion?
I think it's already converged if the pattern is same....go for transient case then
can this be used to model the Eppler 473 airfoil
What are the flow and boundary conditions
may i know the paper you are using ? thank
What about minimum orthogonality quality that is required to be greater than 0.1? This method produces minimum orthongality quality elements on the scale of 10e-4.
Like many cases, a orthogonality really depends on the type of flow and in airfoils with the flow almost purely in one direction and isn't sensitive to cell orthogonality similar to how it is insensitive to the huge aspect ratios but since the flow is in the direction of the length of the cells, it works out fine. In general, any rule of thumb requirements are really just to help if there are any convergence issues and you need to trouble shoot. If you check out any of NASA's published meshes, they have pretty much similar or worse aspect ratios, growth rates, and orthogonality quality because they sought and published the lease computationally expensive meshes (I believe as low as ~3000 cells) that yielded accurate airfoil analysis .
Hi Anthony, I need help with morphing of naca 4412 for my thesis using ansys. If you have time, please let me know if we can get in contact. I am waiting for your response
Hi, I've applied them on NACA0012, but I got negative drag. I've checkced the direction of drag, it's x=1 y=0. I could not find a solution for this. drag is not normal value even it would be positive such as -5e-01! Could you help for this stiuation? Thank you.
did you success about negative drag? I also got near zero drag. If you did that, could you help?
I prefer to use Catia, to import airfoil CAD file to see the effect of dimple, flap etc. It is a bit difficult to draw in design modular. However, i was rotating the airfoil to investigate it at different angle of attacks. It is easier to change x-y component of course. I have tried, but didnt notice i also have to change the components from the report definitions. Thank you!
I think the way forward approach is modelling in ANSYS SpaceClaim. I made the switch from ANSYS Designmodeler more than a year back and I fell good about it.
I am not able to mesh it shows error for defeature
The upper right face fails during meshing every time. I've repeated this process about 15 times now with different variations
What value of drag you are getting...i am getting a negative value
what champ do you main on league
I think you have a mistake in lift coefficient coordinates vector. If your angle of attack is positive (which is in this case like that).. The lift vector should be : cos(13) in y direction and -sin(13). Lift must be in upward-front direction (your front airfoil in negative x). So think like that. Please correct me if im wrong
Why do we want a Re number of 3 million? Is it just because we want it to be fully turbulent everywhere?
why the velocity minimum in velocity contour is zero
I literally tried doing the same thing as you during the geometry part but I can't even get the lines from sketches (12:20) to work. it kept giving me "Boolean operation failed"
Are you doing the Boolean or the projection? For the Boolean you just need to make sure both bodies are defined and closed
can u do a video for an airfoil with dimples
Hi sir, I have a unique problem
I simulate a 2D axisymmetric steam ejector
the flow is a very high speed, 1300m/s at the outlet of converging-diverging nozzle
the case is a Density-based problem, with water vapor(density: ideal gas) as a working fluid
my mesh is coarse (2000element) the simulation goes well
but when my mesh is fine (40000-60000 element) the floating due to excessive temperature occur
actually, my orthogonal quality, skewness, and aspect ratio is very good
I don't really know what the problem is
Without looking at it, I immediately know that high of a speed you will need a much much more refined mesh for proper results and probably a different solver. Its high compressible with thermal effects and there's no way you can do it. At low elements it have a lot of generalization and simply isn't accurate, doesn't matter if it converses. At higher refinement it starts to properly solve the physics, but that mesh doesn't sound fine enough. You need Y+ < 1 and at that speed you likely are not.
you keep the inlet velocity tho, you should make the inlet velocity lower to much NASA Reynold number
Hi Antony, Thanks for this video.
I can see in BC section under the walls section, you have two wall zones first one is the airfoil profile but the second one is undefined which I guess should be the projection lines.
I also get this zone but I defined them as an interior which I am not sure this is a good approach. is there any way to fix these edges? I tried to de-activate those but it would remove all boundaries.
Thanks in advance for your answer.
Generally you shouldn't touch any of the interior boundaries.
Thanks for the video! Please Help I'm at 9:50 and want to create surface from sketch but it keeps saying: Invalid Profile Selection. Looked around on the internet but I can't find a solution
Same here
@@gayathrigunasekaran8550 My guess is that the sketch is not fully enclosed. Double check to make sure all lines are coincident with each other.
@@anthonyt6564 It works when the dimensions of the c-mesh are small (like around 200) and doesn't work when I change dimensions to be larger, or just make the sketch larger
@@gayathrigunasekaran8550 you can dimensionally scale it down if needed but the geometric limit of ANSYS isn't that small. I've done larger sections with domain height and length around 100m before with no issue. There is an upper limit but its on the scale of millions of meters.
@@anthonyt6564 Oh, that makes sense. Thank you!!
Hi, i was thinking on nasa's website that M=0.09 is mach number and ıt means 30 m/s. So ı am confused about reynolds number and velocity. I am beginner in fluid mechanical. Can you explain please?
And the same thing for NACA 0012 in your Udemy Video. for M=0.15, your inlet condition is 88 m/s ?
A simple question. If i have to study any profile for different angle of attacks how would i change the angle of attacks eache time in DEsign Modeler?
Thanks i got the answer from ur video. we have to vectorize the inlet velocity and drag and left
I appreciate the level of depth you went into in this tutorial.
There are two problems I ran into on the setup: For the wall section (24:34), I have airfoil but I don't have "wall-surface_body". Second problem is, I have continuity, x & y velocities, nut and time values (like yours at 31:33) but drag is missing - which is strange because I created reports for both lift and drag like you did in this video.
usually you don't need to define the wall-surface body so as long as it isn't effecting results I wouldn't worry. It's likely some settings in the mesh you may have done differently. As for the drag output, there's a button to have it outputted in the console like the rest or make sure it is in a drag report and that is being plotted. Depending on what version, you may have different default settings.
@@anthonyt6564 Hey airfoil is not turning up, but wall_surface body is. How can I resolve this issue?
Thank you Anthony, your explanation has saved many of us. I have a question: If I did this same exercise but with transient flow, what time step should I use?
For any transient fluid simulation, you should generally follow the rule that your the time time step must be small enough for your Courant (or CFL) number to be less than 1.
I am studying the performance of the airfoils in ground effect, What will you give the boundary conditions? FOR THAT..
A ground effect will still be a no slip and its caused by the wall being close to the actuation so it's more so based on your geometry. It's the same logic to a wind tunnel wall effects
Sir can you make video on dimple airfoil
How to create dimple on airfoil
I did the same as you but to the undivided half of the circle. How may I contact to you for more details
I don't know what you mean by the undivided half circle. Do you mean the leading edge of the airfoil?
thank you, can you send me the wbpj file please?
I could not generate txt file :(
If I’m doing a laminar flow instead of turbulence, can I follow every steps you do except changing spalart allmaras to laminar? Is there anything to look out for besides changing certain parameters like velocity / viscosity those?
Yes, it should all be the same. You would want to set your velocity so that it's laminar, but you'll still get decent results. The slight discrepancies would be seen in your results.
@@anthonyt6564 Hmmm, i tried it out few days ago using ur Udemy tutorial. but i get 0.2 and 0.1 for lift and drag respectively. Not sure whats wrong with my simulation. Will try again later
@@anthonyt6564 i ran my simulation with velocity of 4.075m/s with reynold number of 260941.05
I tried increasing mesh size and trying different viscous solver but my simulation just doesn’t converge. Why? 😢
@@user-du2pt4zv5i is this at 0 degrees AoA? If youre doing laminar solver the lift would generally be somewhat correct and the drag is almost always under-estimated. How are you gauging convergence?
The mesh i generated looks all skewed for some reason
How do I generate the mesh for 0012, should the first size be "hard"?😵💫
When I mesh in soft, it comes out error: One or more faces with map mesh controls failed to be mapped mesh. How to solve it🥲
What is the wing length and wingspan for the naca 4412, how can I find it?
he used the length of 1 meter
Hey what's your summoner name :D ?
thanks bro!