
[Sponsors] 
October 23, 2011, 03:04 
Moving Mesh Problem

#1 
Member
omid
Join Date: Apr 2011
Posts: 98
Rep Power: 6 
Dear members
I'm trying to simulate flow over airfoil. this airfoil has a moving surface and the relation that specify mesh motion of this surface based on pressure distribution on surface. in this relation i need to access the arrays of pressure in this surface, such as pressure on a line on surface of airfoil. How can i do this? i use an expression like this: LogicalExp= if(x>0[m]&&x<10[m]&&z=1[m],Pressure,0) my purpose from this expression is the pressure of specified location( 0<x<10, z=1) store in it ( Logical expression). i tested it, but i think pressure of any node of location that i use in my relation be stored. (not in range 0<x<10 , z=1) !! Do you have any idea? Thanks, best regards. 

October 23, 2011, 05:36 

#2 
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 11,290
Rep Power: 88 
You have started many threads on this topic but never actually explained what you are doing. That is why you have not got any very useful comments.
Please show a picture of what you are analysing, how the motion of the moving section is controlled and what Re and Ma numbers you are operating at. 

October 29, 2011, 06:02 

#3 
Member
omid
Join Date: Apr 2011
Posts: 98
Rep Power: 6 
Hi
Thanks for your attention,sorry about this, i'm looking for answer. I'm trying to simulate flow on airfoil like figure(1). i use mass flow condition equal 4.9 kg/s of CH4 at inlet and average static pressure 41 bar at outlet. flow domain show in figure 2. Airfoil surfaces mesh have displacement along y direction. each node of grid of airfoil have a different displacement that have a relation like (Dy) by the pressure on that node. the expression that specify this displacement as follow: Dy = tts*Coefficient *((Pressure  Pin) tts and Coefficient are constant and Pin is : Pin = probe(Pressure )@PP PP is a point that have a known coordinate( on leading edge). When i use it, the mesh on leading edge move like figure(3), i think if i can access to pressure on each line of mesh on surface of airfoil and Probe pressure on each first node of line (first node on leading edge), this mesh folding be resolved. My questions: 1 how can i access to pressure at each line on the airfoil? i use an expression like this: LogicalExp=if(x>0[m]&&x<10[m]&&z=1[m],Pressure,0) but i think pressure of any node of location that i use in my relation be stored. (not in range 0<x<10 , z=1) !! Do you have any idea? Best regards 

October 30, 2011, 05:27 

#4  
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 11,290
Rep Power: 88 
Do not use a pressure of 41 bar at the outlet. Use a pressure of 0 and use a reference pressure of 41 bar.
You can apply your displacement equation as a simple CEL expression, Mesh Displacement y = constant *(pp_inlet). As simple as that. If you had explained what you were trying to do in the beginning we could have told you that weeks ago Sounds like a weird displacement equation to me, but who am I to comment? I still have no idea what you are modelling. Quote:
Quote:


October 30, 2011, 07:25 

#5 
Member
omid
Join Date: Apr 2011
Posts: 98
Rep Power: 6 
Thanks for reply dear ghorrocks.
I'm sorry about this, my problem is some complicated! My main idea optimize the shape of airfoil, and this problem is a first step of that. this optimization is based on pressure distribution on airfoil. There is no secrecy in displacement equation, the constant value is 3.07E7 and i use it for control and decrease displacement for avoiding mesh folding. the full description of displacement ccl is: Dy= 3.077E7[m^2 kg^1 s^1] * ((Pressure  Pin)) Pin is the pressure in a point at leading edge of airfoil that i use it for gauge my pressure than there. and i use probe function for measure pressure at this point during solution. i use this relation in airfoil boundary as "specified Displacement Y component" . but after solution, the grid be uneven like fig(1). i think if i can define this relation (Dy) for each line in surface of airfoil and have a gauge point (Pin) at the start of each line this mistake be resolved. and it is my reason for looking for way to access to pressure on the airfoil during solution .then i use LogicalExp=if(x>0[m]&&x<10[m]&&z=1[m],Pressure,0) 

October 30, 2011, 17:04 

#6 
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 11,290
Rep Power: 88 
Your displacement equation is clearly wrong. For instance it does not take into account the top and bottom surface of the wing  on one face you want to move the point down and one up.
With an incorrect displacement equation then you will get weird results. 

November 12, 2011, 05:23 

#7 
Member
omid
Join Date: Apr 2011
Posts: 98
Rep Power: 6 
Thanks Dear ghorrocks
i modify that equation and my result become better. but there is some problem i will fix it. There is a question in my mind, may haven't any relation by my problem. and i will be very grateful if you help me to find answer, that question is: How i can access to the pressure matrix? Best regards. 

November 13, 2011, 18:01 

#8 
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 11,290
Rep Power: 88 
Yes you can through user fortran. But from what I understand of your application this is not what you want. Your displacement function is wrong and will give rubbish until it is fixed, and that CEL expressions can access all the variables you need to get this to work.


November 19, 2011, 05:39 

#9 
Member
omid
Join Date: Apr 2011
Posts: 98
Rep Power: 6 
Thanks for reply Dear ghorrocks
i'm trying to work with user fortran, but its some complicated! there is an utility routine : USER_GETVAR and USER_GETGVAR but i can't set the argument true. Could you tell me if i want to store Pressure variable at a boundary with name "Moving" at each time step, how can i do this? 

November 19, 2011, 06:10 

#10 
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 11,290
Rep Power: 88 
Before you dive into fortran, can you explain what you are doing? Why does the y coordinate change with pressure? You have mentioned you are optimising the airfoil but I assure you your optimisation equation you described above will not work and will diverge. Can you explain how the equation you mentioned was derived? Or did you just make it up?
There are plenty of publications out in the literature on airfoil design and optimisation. Have you done a literature search for them? If you are doing airfoil optimisation with a seeing what has been done before you are wasting your time (and my time too, for that matter). 

November 23, 2011, 09:03 

#11 
Member
omid
Join Date: Apr 2011
Posts: 98
Rep Power: 6 
Thanks for reply ghorrocks
I'm trying to optimize airfoil. i read some studies about it, this is a preliminary simulation that i want to know is this simulation possible with CFX? 

November 23, 2011, 19:55 

#12 
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 11,290
Rep Power: 88 
This is certainly possible, but your optimisation equation is wrong and it will never work until you fix that. It is not CFX which is failing here but your optimisation equation.


Thread Tools  
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Moving mesh problem OpenFoam 141  kassiotis  OpenFOAM Running, Solving & CFD  30  April 14, 2015 23:10 
2D Mesh Generation Tutorial for GMSH  aeroslacker  Open Source Meshers: Gmsh, Netgen, CGNS, ...  12  January 19, 2012 04:52 
snappyHexMesh won't work  zeros everywhere!  sc298  OpenFOAM Native Meshers: snappyHexMesh and Others  2  March 27, 2011 21:11 
How to control Minximum mesh space?  hung  FLUENT  7  April 18, 2005 09:38 
Moving mesh problem  Jan Majer  CDadapco  2  June 10, 2003 04:19 