
[Sponsors] 
March 8, 2014, 12:00 
rotating and pitching UDF help/check

#1 
New Member
sam daysley
Join Date: Feb 2013
Posts: 28
Rep Power: 3 
Hi everyone,
I'm looking to simulate a turbine (see attached pic) with airfoils that rotate about an axis and change angle of attack at the same time (in a sinusoidal sense). Currently i have three zones, the large stationary zone, the rotating turbine zone and then the airfoil zones. i thought if i set the turbine zone rotating at x rad/s and then set up a udf motion for the airfoil zone relative to the turbine zone it would allow me to change the angle of attack whilst rotating? however this doenst happen, so i wanted to know whether im using the correct udf macro? should i be using DEFINE_CG_MOTION() i.e. dynamic zone? or is my code wrong? here is what i have as my code: Code:
#include "udf.h" DEFINE_ZONE_MOTION(sine_motion,omega,axis,origin,velocity,time,dtime) { real posx, posy, rad, angle, angvel, pi; rad = 0.4572; pi = 3.141592654; angvel = 10; angle = angvel * time * (180 / pi ); *omega = cos(10 * time); posx = rad * cos(angle); /*where posx and posy are the position of the airfoils rotational origin as the turbine rotates*/ posy = rad * sin(angle); origin = (posx, posy, 0); return; } thanks in advance for the help 

March 17, 2014, 09:43 
Re: Rotating and Pitching UDF

#2 
New Member
Mahdi Torabi Asr
Join Date: Dec 2013
Posts: 10
Rep Power: 3 
Here It is a UDF for simultaneous Rotating and Pitching.
be noticed that the UDF should be set for the blades (considering initial angular its position) and the "Releative specification" option should be set to the rotor's rotating frame. #include "udf.h" DEFINE_ZONE_MOTION(pitching_blade,omega,axis,origi n,velocity,time,dtime) { real theta0, thetap1, omegac, omegar, radius, pi; pi = 3.141592654; omegar = 10; /* pitching velocity of blade [radians/s] */ omegac = 5; /* angular velocity of rotor [radians/s] */ theta0 = pi/2.; /* initial angular position of blade origin [radians] */ radius = .5; /* radius of rotor [m] */ thetap1 = omegac*(time+dtime); /* angular change from initial position at t+dt */ *omega = omegar; /* pitching velocity of blade */ /* timevarying origin of the local rotor zone coordinates in meters */ origin[0] = radius*cos(theta0+thetap1); origin[1] = radius*sin(theta0+thetap1); origin[2] = 0.0; } In your case *omega variable should represent the sinusoidal time function. 

April 7, 2014, 18:16 

#3 
New Member
sam daysley
Join Date: Feb 2013
Posts: 28
Rep Power: 3 
Thank you mahditorabiasr
I will try this method now so thanks in advance 

April 8, 2014, 01:06 

#4 
New Member
Mahdi Torabi Asr
Join Date: Dec 2013
Posts: 10
Rep Power: 3 
Welcome daysley,
please let me know if I can be of any further help. 

April 12, 2014, 14:18 

#5  
New Member
sam daysley
Join Date: Feb 2013
Posts: 28
Rep Power: 3 
Quote:
I've been playing around with the udf you wrote but i'm having some problems when running it in Ansys. as the turbine rotor rotates, the blades rotate away from their position in the mesh as though the changing origin is incorrectly changing? I don't suppose you have any idea why it would do this do you? everything seems right in the code and the motion is set relative to the rotating zone? does it have anything to do with the line: Code:
thetap1 = omegac * (time + dtime) thanks in advance i've attached my altered code all i have changed is the Pi function and radius etc. Code:
#include "udf.h" DEFINE_ZONE_MOTION(pitching_blade,omega,axis,origi n,velocity,time,dtime) { real theta0, thetap1, omegac, omegar, radius, pi; omegar = sin(time) /* arbitrary sine function*/ omegac = 10.94 ; /* pitching velocity of blade [radians/s] */ theta0 = M_PI/2.; /* initial angular position of blade origin [radians] */ radius = .4572; /* radius of rotor */ thetap1 = omegac*(time+dtime); /* angular change from initial position at t+dt */ *omega = omegar; /* pitching velocity of blade */ /* timevarying origin of the local rotor zone coordinates in meters */ origin[0] = radius*cos(theta0+thetap1); origin[1] = radius*sin(theta0+thetap1); origin[2] = 0.0; } 

April 15, 2014, 12:58 

#6  
New Member
Mahdi Torabi Asr
Join Date: Dec 2013
Posts: 10
Rep Power: 3 
Quote:
Sorry for late reply. let me ask whether you have defined separated UDFs or not? Be sure that you have a UDF for each blade including initial angular position for each ("theta0"). In addition any little difference from the real values in your initial mesh can lead to incorrect position. Answering your question, "thetap1" returns the angular position in next time step("dtime" is equal to time step which you have set in Fluent). Code:
thetap1 = omegac*(time+dtime) 

Tags 
fluent  udf, udf, vawt 
Thread Tools  
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Rotating mesh in a moving reference frame  mafz1g11  FLUENT  0  March 5, 2014 13:50 
2D Pitching Airfoil  samcfd  Fluent UDF and Scheme Programming  11  February 25, 2014 07:08 
How to simulate pitching oscillation?  zkdkeen  CFX  1  December 11, 2013 17:34 
Simulation with a rotating frame within another?  siw  CFX  0  April 18, 2013 08:13 
Blade pitching, butterfly valves, moving mesh  Andy Fiedler  CFX  6  May 28, 2010 09:55 