# Moving reference frame ?

 User Name Remember Me Password
 Register Blogs Members List Search Today's Posts Mark Forums Read

 October 28, 2015, 06:40 Moving reference frame ? #1 New Member   Join Date: Oct 2015 Posts: 8 Rep Power: 3 Hello everyone, I have a fan composed by an intake, a rotor, a stator and an exhaust. I have therefore two volumes : one composed of the rotationnal air around the rotor ("fluid_MRF") and another composed of the other parts ("fluid"). To simulate the rotation of the blade I have placed the fluid_MRF in a moving reference frame. The rotor is consequently stationnary in regard to this reference frame. The other volume "fluid" is placed in a stationnary frame and the walls in contact with itare also stationnary. The problem I obtain when I plot the vectors of absolute and relative velocities is that the fluid is purely tangentiel around the rotor and that the velocities at the intake or at the exhaust are not in the right direction (the fluid goes from the outlet to the inlet) and almost null. Therefore my question is : how do you make the fluid goes properly from the inlet to the outlet (without having a fluid purely tangentiel around the rotor) ? For information I have a pressure-inlet and a pressure-outlet with a direction of the velocity normal to the boundary. I heard that it might be due to the interface but I don't know more precisely. Thanks a lot !

 October 31, 2015, 17:29 #2 New Member   Join Date: Mar 2015 Posts: 21 Rep Power: 4 HELLO, first thing to consider is that the mrf follows right hand law,so when you are defining the axis of rotation you should apply this.

 November 2, 2015, 05:54 #3 New Member   Join Date: Oct 2015 Posts: 8 Rep Power: 3 Yep I used the right hand law to define the rotation axis. I progressed through some calculations but now I have another problem : with a higher rotationnal velocity (100 rad/s) I have the fluid that comes from the inlet to the outlet. But when I plot the relative velocity I dont get the triangle "absolute velocity equals to the relative velocity plus the rotationnal velocity" (the absolute velocity is more quals to the relative velocity minus the rotationnal velocity). Do you know where can that comes from ? Also another question : when you don't really know what you have at the outlet of your geometry, is it better to use a pressure-outlet or a outlet-vent ? And what are the criterions to choose either one or another ?

 November 2, 2015, 07:25 #4 New Member   Join Date: Mar 2015 Posts: 21 Rep Power: 4 i did a wind turbine recently but i did it srf, because mrf is a little tricky,i had reverse flow with outlet boundary condition,but when i did it with pressure outlet it disappeared.my problem was a little like yours and it became cleared that i have mistaken the rotational axis direction. about the interface problem,from what i know if you do it wrong,fluent would diverge at the beginning. another thing that may not be your case but did you use a small enough time step?

 November 2, 2015, 08:25 #5 New Member   Join Date: Oct 2015 Posts: 8 Rep Power: 3 I only did steady simulations, with a constant rotationnal speed. At the beginning the rotationnal speed was set to 13rad/s, which corresponded with my geometry to (only) 1.2m/s with a radius at mid-span of 100mm and a span of 30mm. At that time I had the problems I described in my first post (purely tangentiel fluid, abnormal behaviour at the inlet and at the outlet). But with a rotationnal speed set to 100 rad/s (10m/s) I have a "normal" behaviour of the rotor : the air comes from the inlet to the outlet and I have no reverse flow. Therefore I don't think that the pb comes from the fact that I work in steady flow, but I'm not sure maybe it takes some time for the flow to stabilize... Do you think I should use unsteady mode ? Thx for your answer !

 November 2, 2015, 09:13 #6 New Member   Join Date: Mar 2015 Posts: 21 Rep Power: 4 this kind of problems are naturally unsteady and if you plot cl in fluent you see even in steady state it is periodic and it shows the flow characteristics,the big difference is the convergence that is better in unsteady,if you post a picture of your mesh which the directions and rotational direction is clear and a picture of your mrf setting may be i could help.

November 3, 2015, 06:19
#7
New Member

Join Date: Oct 2015
Posts: 8
Rep Power: 3
Here they are. Since I am working in 3D, the mesh I present you corresponds to a plane situated at the tip of a rotor blade, with a normal x. The fluid is moving from the Y+ to the Y- and you can see 3 stator blades after the rotor blade, aiming to stabilize the fluid. The red arrow corresponds to the rotation of the rotor blade.
Attached Images
 mesh_coupe_z-_fleche.png (148.0 KB, 48 views) params_fluide_MRF.PNG (16.3 KB, 40 views)

 November 3, 2015, 13:00 #8 New Member   Join Date: Mar 2015 Posts: 21 Rep Power: 4 hello,may be im not seeing correctly but from what im seeing your axis of rotation should be y=1 not y=-1.i hope i have assisted you.

 November 4, 2015, 11:02 #9 New Member   Join Date: Oct 2015 Posts: 8 Rep Power: 3 Hi ! Yes the blades are rotating with the direction +y but since the fluide is supposed to turn on the other side (toward the pressure side) I put it in the direction y-. It seems legit for me but maybe I am mistaken !

 November 16, 2015, 08:22 #10 New Member   Join Date: Oct 2015 Posts: 8 Rep Power: 3 [Repost] Hi guys ! Sorry to ask again but if the blades are moving with the direction of the axis of rotation equals to y+, sould I place the rotation of the rotating frame toward y+ or y- ?

 Thread Tools Display Modes Linear Mode

 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 OffTrackbacks are On Pingbacks are On Refbacks are On Forum Rules

 Similar Threads Thread Thread Starter Forum Replies Last Post r_gordon OpenFOAM Running, Solving & CFD 92 August 16, 2017 03:24 Tobard FLUENT 0 April 14, 2011 15:18 radix FLUENT 0 December 30, 2009 07:02 skabilan OpenFOAM Installation 3 July 28, 2009 00:35 aiqch Main CFD Forum 0 April 3, 2009 00:23

All times are GMT -4. The time now is 19:16.

 Contact Us - CFD Online - Privacy Statement - Top