
[Sponsors] 
May 19, 2019, 10:51 
First CFD Analysis  Results Question

#1 
Member
Join Date: May 2019
Posts: 30
Rep Power: 3 
Hello,
I did my first CFD Analysis with Ansys CFX and it worked pretty well. I have different shaped Bodys (different simulations) and I want to compare how they affect the flow (Mach 0,3  Mach 0,7) behind the body. I want to find out, which one has the at least bad effect on a rotor, which will be placed 200 mm behind the body. (the rotor is not part of the CFD analysis) What would be a good approach to analyze the results I've got from the CFD Analysis? I hope you can help me, I'm not experienced (still studying) and if you need some information which I forgot, please ask me. Best Regards 

May 23, 2019, 11:55 

#2 
Senior Member
Matt
Join Date: Aug 2014
Posts: 847
Rep Power: 14 
Look at the total pressure deficit and uniformity in the wake. Loss of total pressure can be correlated back to drag. It is a direct indication of the loss of energy of the flow. The more energy that is lost, the harder the rotor will have to work to accomplish the same task. The uniformity will give you some idea as to the flow conditions the rotor would see. Assuming this is an axial (rather than centrifugal) rotor, they generally prefer nice uniform flow fields... or are at least designed under the assumption of one.


May 29, 2019, 09:53 

#3 
Member
Join Date: May 2019
Posts: 30
Rep Power: 3 
Hi,
thank you very much for your answer. So I plot the total pressure with a contour like on the picture below (total pressure) and then I compare it between the different shapes? Would maybe calculating the drag coefficient with CFX be another good approach? I have a question regarding my Mesh. Today I checked Y+ on the body, the used turbulence model is kEpsilon. The Normal Speed on the Inlet is Mach 0,3 in one simulation and Mach 0,7 in the other one. Outlet Option for Mass and Momentum is an average static pressure with a relative pressure of 100000 Pa. I have 5 prism layers, first layer has a thickness of 0.1 mm. I have 5 layers with a grow rate of 1.2. I read that a y+ beyond 100 should be fine with this turbulence model.Does the Y+Plot looks okay? (check the picture y+ plot) Thank you for your help. 

June 14, 2019, 13:27 

#4 
Member
Join Date: May 2019
Posts: 30
Rep Power: 3 
So, I'm now at the point, where i want to compare the different profile shapes and it's effect on a rotor which would be downstream the profile. As fluid23 said, I want to check the total pressure loss and uniformity in the wake at Mach 0.7.
I have 3 different Profile shapes, which you can see on the Picture Profile shapes. (please ignore the blue Points on the leading edges of the profiles) I ploted the total pressure contour for all 3 shapes with a direct flow and for an angle at 20°, as you can see at the two other Pictures. My questions: It seems like there is no difference for the 3 Profile shapes if the flow comes directly with no angle Variation. Does this seem legit? Why is the total pressure loss for both rounded profiles weaker for the case where the flow comes with an angle of 20°? This doesn't make any sense to me. What could be an explanation? Thank you guys very much. 

June 14, 2019, 13:40 

#5  
Senior Member
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 5,620
Rep Power: 60 
Quote:
Actually, I see differences at zero AOA ...the lengt of penetration of the recirculation zone appear different (see the yellow/orange region). I suggest to do a 1D cut at several locations behind the body and superimpose the 3 pressure plot in the same figure for each station 

June 14, 2019, 13:58 

#6 
Member
Join Date: May 2019
Posts: 30
Rep Power: 3 
Okay, thank you very much. Than I think it would be a good idea to measure the total pressure on various points on different horizontal lines behind th Body and plot it into a diagram.
Is the total pressure loss a good indicator to find out how the flow downstream the body is affected? As i said in my case there will be an axial rotor in a at the Moment not known distance behind the body. Or would compare the velocity would be better? How can it be explained, that the total pressure loss for the rounded bodys seems to be smaller at an angle of 20°? 

June 14, 2019, 14:02 

#7  
Senior Member
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 5,620
Rep Power: 60 
Quote:
Smaller than the case of the same body at zero AOA? After you get the 1D plot you can quantify the difference in a better way 

June 14, 2019, 14:37 

#8 
Member
Join Date: May 2019
Posts: 30
Rep Power: 3 
Yes, as you can see in the picture. On one picture you can see the total pressure plot at 0 AOA for all 3 profile shapes and on the other pictures you can see the total pressure plot for the same shapes at an angle of 20°. The chosen range for the total pressure and the number of contours is the same for every plot.


June 14, 2019, 14:40 

#9  
Senior Member
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 5,620
Rep Power: 60 
Quote:
I see a different setting of the max value .... 

June 14, 2019, 14:48 

#10 
Member
Join Date: May 2019
Posts: 30
Rep Power: 3 

June 15, 2019, 16:19 

#11 
Member
Join Date: May 2019
Posts: 30
Rep Power: 3 
I made 2D Plots today and measured the specific total pressure values for the different shapes at different distances to the body at horizontal lines (through the wakes).
I got the results as you guys can see in the plots I posted yesterday. The total pressure loss for the U and "NACA"Profile is much more lower for an flow angle of 20° than for the flow with an AOA of 0. How is this possible? Is there any explanation? 

June 16, 2019, 13:59 

#12 
Member
Join Date: May 2019
Posts: 30
Rep Power: 3 
The volicity on one side of the ushaped profile goes up to 500 m/s. Could this be a problem?


June 16, 2019, 14:40 

#13 
Senior Member
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 5,620
Rep Power: 60 

June 16, 2019, 16:32 

#14 
Member
Join Date: May 2019
Posts: 30
Rep Power: 3 
I set the Inlet speed to Mach 0.7.
You can find the plot of the streamlines below. What could be the mistake? 

June 16, 2019, 16:50 

#15 
Senior Member
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 5,620
Rep Power: 60 
The acceleration in that region is right but such a high value of the averaged velocity seems excessive. Try to refine the mesh and check also the density field


June 17, 2019, 10:03 

#16 
Member
Join Date: May 2019
Posts: 30
Rep Power: 3 
Okay, thank you for your answer.
My Default Domain Turbulence model is kEpsilon. I have a question regarding the Inlet Boundary conditions. For Turbulance I Chose "Medium Intensity 5%". Is this a mistake? I can also chose k and Epsilon there, but than I have to define values for "Turb. Kinetic Energy" and "Turb. Eddy Dissipation". Is this what I should use or is "Medium Intensity 5%" good enough? If not, what should I chose for values for "Turb. Kinetic Energy" and "Turb. Eddy Dissipation"? THank you very much in advance. Best Regards 

Thread Tools  Search this Thread 
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
General question about CFD applied to moving parts  Voxel  Main CFD Forum  0  May 11, 2019 13:35 
CFD Analysis Over Three Dimensional CD Nozzle  vickyroczzz  Main CFD Forum  0  March 25, 2018 03:19 
Validating CFD results with Experimental results for Indoor climate  Mikki  FLUENT  0  January 4, 2016 17:34 
Is it usual in CFD community to guarantee the results?  oj.bulmer  Main CFD Forum  8  November 12, 2014 08:05 
Is it usual in CFD community to guarantee the results?  oj.bulmer  CFX  0  November 5, 2014 06:31 