CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Pitch Change & interface connections (http://www.cfd-online.com/Forums/cfx/24351-pitch-change-interface-connections.html)

Kumar August 3, 2007 05:18

Pitch Change & interface connections
 
There was already one thread in this index about pitch change but still i need clarification.

For an impeller(rot domain)& for a vaned diffuser (stationary domain) no of vanes are half the no of blades,I have two diff meshes from turbogrid.So I need frozen rotor interface in between.Also as I am going with a 360 degree case i need a ggi for periodic connections for them.Finally rot domain is connected to Inlet nozzle/pipe(Icem hex)with frozen rotor &Diffuser is connected to Volute(ICEM tet mesh) with a GGI.

hence I have four diff meshes with two frozen rotor interfaces and 3 GGI.

Does these no of interfaces affect the solution results? (resolutions are not differing too much)

Stiil I need any pitch change in between rot impeller domain and a stationary vaned diffuser?

My solutions are converging and there are no imbalances untill now even with no pitch change.Should I believe them?(Still validation of results is to be done.)

waiting for comments.

Thanks and regards,

Kumar.


Robin August 3, 2007 08:52

Re: Pitch Change & interface connections
 
Hi Kumar,

The number of interfaces won't affect your solution. There can be some loss of the profile at an interface if the mesh resolution changes significantly, but the conserved variables, such as mass, momentum and energy, will all be conserved across the GGI.

Since you have a full 360 degree model there is no pitch change and you should therefore set your pitch change to "None". Also note that at the frozen rotor interface you can specify a pitch offset. This allows you to clock your blades to see the influence on the solution. You will find that the solution will vary with the relative position of the blades due to the wake profile and should try 10 to 20 different positions.

On the last point, trying different clocking positions, you might want to consider running a transient instead and calculating the time-averaged velocity, pressure, temperature, etc. Statistical fields can easily be set up in the Output Control in Pre when you have a transient analysis. In the end, it is less work and more accurate than running the series of clocked position.

Regards, Robin

Kumar August 5, 2007 02:57

Re: Pitch Change & interface connections
 
Thanks Robin.

That means I was on the right path.

Still not thought of any transient runs. But what will be the changes i should do to run it as a transient case? And can i see a moving impeller in post with a steady run? or what should i do to see it?



All times are GMT -4. The time now is 10:03.