CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   relative velocity (https://www.cfd-online.com/Forums/cfx/77022-relative-velocity.html)

Zaktatir June 18, 2010 03:05

because Water.Velocity U-Air.Velocity U on the inter-phase is almost small because of the shear layer near the interface..


<---------------------------------
<---------------------------
<-----------------------AIR
<---------------------
<-----------------
<----------
<--
-->

--------------->
-------------------->
----------------------->WATER
--------------------------------->
---------->
-->

The assumption is accordingly to KH-theory: two Fluids flowing over each other with constant velocity:


<--------------------------Air--------------
************************************
---------------------------WATER-------->

ghorrocks June 18, 2010 07:08

Yes, and what is your point?

If the interphase drag is large then the velocity profile looks like your top diagram. If the interphase drag is small then the velocity profile looks like the bottom diagram. Either way the relative velocity between phases is Water.Velocity U-Air.Velocity U.

Zaktatir June 18, 2010 08:22

i have the first case.. and i am trying to model water separation at the crest of the established waves on the interphase.. and its correct.. i only need the average velocity of the air flow and the average velocity of the water flow and then build the difference..

ghorrocks June 20, 2010 07:29

Maybe I am dumb but I have no idea what you are talking about. Can you explain what you are trying to do again?

Zaktatir June 20, 2010 09:38

No, you are not dumb i think i could not explain my problem.

I have very big drag force between the both phases nevertheless i want to estimate or in other words to create a relative velocity between them as it is the secand case, that means i want theoretically simplify my problem assume that the drag is not too big (assume that there is no velocity boundary layer in the vicinity of the interphase) and get a value that alows me to calculate the wavelength and growth rate of the waves due to the KH. (e.g.: getting the max Air Flow Velocity - the max. Water. Velocity for x= const, y is variable (height of chanel 90mm))

ghorrocks June 20, 2010 19:45

What you describe sounds easy - draw a line where you want to calculate it and use the function calculator on that line to get the maximum Air.Velocity and maximum Water.Velocity. Doesn't that do what you want?

Zaktatir June 21, 2010 03:59

yes that is it but i need this not in post i need it in pre for the run

ghorrocks June 21, 2010 08:54

Firstly - you have not mentioned any of this before. If this is what you are looking for then why didn't you say so at the start?

You cannot generate line objects in the solver to my knowledge. The max and min functions should work fine in the solver. You could make little surfaces with a narrow width, or you could make a series of monitor points and take the maximum of them. If you put one monitor point at each node you will have the same resolution as the post-processing approach anyway.

Zaktatir June 21, 2010 12:58

that means defining a variable location in pre is not possible.

ghorrocks June 21, 2010 23:13

What do you mean by a variable location?

Zaktatir June 22, 2010 03:03

Something like this:
in CFX-Pre i set an Expression which will be calculated for every ZY-plane at X ( e.g. X between 20cm and 40 cm)

ghorrocks June 22, 2010 20:52

These functions need to be applied to either a physics region or a meshed region. You would have to have these regions present in the mesh but they don't need to be boundaries, just mesh regions is OK - see the documentation on CEL.

Zaktatir June 23, 2010 03:16

it means that i should define a lot of these planee to solve my problem. how many planes are suitable, and if mesh this planes in ICEM i will face the problem multiple edge e.g...

ghorrocks June 23, 2010 09:21

Yes, that's right. It is not a nice solution but you don't have much other choice. THe only other option I can think of is to define monitor points and manually get the max/min values of them. This does not require the mesh to be changed. But if you have lots of monitor points then best set it up using perl or some other text editor.

Zaktatir June 23, 2010 10:28

i´ve already started dealing with the mesh solution meanwhile
i am trying to build an entity like the relative velocity based on the Velocity u. Gradient Y or the shear strain rate in the vicinity of the inter phase and multiply this with a hydraulic radius of the channel..


All times are GMT -4. The time now is 04:31.