CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Average passge method (http://www.cfd-online.com/Forums/cfx/23649-average-passge-method.html)

 sam February 16, 2007 06:55

Average passge method

hey freinds Is Average passage method is included in CFX for turbomachinery.

 Glenn Horrocks February 18, 2007 17:51

Re: Average passge method

Hi,

I am not familiar with this. Do you mean the average residence time? You can calculate this in CFX-Post by generating a sample of streamlines through the model and colouring by time. The time value as they leave the domain is the residence time of that streamline.

Glenn Horrocks

 sam February 20, 2007 00:12

Re: Average passge method

Hi Glenn, It is about the interface between rotating and non rotating interface. it is advance approach then mixing plane. Regards

 Glenn Horrocks February 20, 2007 17:03

Re: Average passge method

Hi,

Sorry, I don't understand what you mean. Do you mean the various methods of modelling rotor/stator interactions, such as frozen rotor, transient rotor/stator etc?

Glenn Horrocks

 Michele Cagna February 22, 2007 00:48

Re: Average passge method

Hi sam,

in case you mean the method of deterministic stresses: No it is not included.

Michele Cagna

 Bak_Flow February 22, 2007 11:42

Re: Average passge method

From a recent technical note:

http://scitation.aip.org/getabs/serv...cvips&gifs=yes

In layman's terms:

The correct way to solve a turbomachine would be to solve the full geometry from inlet to outlet for all scales of motion.

This would of course mean the all blade rows, all passages, and turbulent scales of motion. We know that for most engineering analysis this is too much effort and more info that we really need because we are going to average it anyhow to get average torque, total pressure rise, efficiency, etc.

In any flow we deal with the unsteadyness of turbulence by averaging the equations: called the RANS equations....these introduce unknown terms called the Reynolds Stresses and Reynolds Fluxes. Which really are the "effect" of the unsteady transport at scales that are not solved. So we need a model for these terms.

In a turbomachine there is a natural unsteadyness from blades passing each other (also looked at as passages passing each other). We could solve for this scale of motion and sometimes do if we are interested in something relevant to the transient...like flutter, etc. However, this is a lot of work and often more that we neet.....see this parallels the issues above in turbulence. So if we average over the scale of motion caused by the passages passing each other.....Average Passage...we get....

you guessed it....unknown terms from the "effect" of the neglected detail....Adamczyk called these Deterministic Stresses....since the passage motion is Deterministic! So you need a model for this....right!

How does mixing plane fit in well this is circumferentially averaging flow variables at the interface.

The technical note has some references. There also are a few handbooks now that cover this.

To my knowledge none of the big 3 commercial codes have this model.

Hope this helps.

Regards,

Bak_Flow

 sam February 27, 2007 17:23

Re: Average passge method

thanks every one I got the answer to my full satisfication. I thank to Glenn, Michele and Bac_flow for there technical help .

One more question, is this method difficult to program?

 Glenn Horrocks February 28, 2007 17:30

Re: Average passge method

Hi,

I haven't looked at Bak_Flow's reference because I don't have a log in, but from his description it looks like it results in modifications to the turbulence mathematical model. This should not be attempted by the inexperienced and is likely to be difficult to implement.

Glenn Horrocks

 All times are GMT -4. The time now is 01:16.