udf boundary condition (vorticity)

yassine.bt · May 18, 2020, 08:37

Hi
I don't know how to write udf boundary condition for vorticity .I bring to your attention that my work has been done on the nanofluid and more precisely I am working on the configuration of Rayleigh Bénard (natural convection).

Fig. 1 shows a schematic diagram of the Rayleigh-Bénard (RB) problem. For the RB problem, the distance between the upper cold and lower hot plates is defined by H and the width of the top and bottom plates is defined by W. The plates’ width W is considered infinite and treated by a periodicity boundary condition. The bottom plate is maintained at a hot temperature TH=50 C whereas the top plate is maintained at a cold temperature TC=22 C. The fluid enclosed between the plates is water based nanofluid containing CuO nanoparticles. The nanofluid is assumed incompressible and the flow is assumed as laminar and twodimensional.
thanks
Inkedgh_LI.jpg

yassine.bt · May 18, 2020, 09:08

Can anyone tell me how to apply dv/dx=0 and omega=du/dy at a boundary using UDF.

[v=delta(psi)/delta(x) u=-delta(psi)/delta(y)]
Inkedgh_LI.jpg

vinerm · May 18, 2020, 09:15

Fluent uses momentum conservation and not vorticity or Poisson's equation. The equation you have is written in terms of streamfunction and implies no-slip condition. That is the default condition in Fluent and you don't need to do anything

yassine.bt · May 18, 2020, 09:52

Quote:

Originally Posted by vinerm

Fluent uses momentum conservation and not vorticity or Poisson's equation. The equation you have is written in terms of streamfunction and implies no-slip condition. That is the default condition in Fluent and you don't need to do anything

But I fond in the results that dv/dx is not equal 0

vinerm · May 18, 2020, 09:55

That won't be 0 and should not be. Look at the document you shared. The condition is NOT

$\frac{d^2\psi}{dy^2} = 0$

but $\psi = 0$ , implying velocity is 0.

yassine.bt · May 18, 2020, 10:05

Quote:

Originally Posted by vinerm

That won't be 0 and should not be. Look at the document you shared. The condition is NOT

$\frac{d^2\psi}{dy^2} = 0$

but $\psi = 0$ , implying velocity is 0.

I think psi=0 that mean (noslip) and omega(rotational flow)??

vinerm · May 18, 2020, 10:15

That's correct and no-slip implies velocity is 0 at the wall, i.e., the default condition in Fluent. As far as rotation is concerned, in a 2D case, rotation is possible only around z-axis, so, $\Omega = (0~0~\omega_z)$ where $\omega_z$ is a function of velocity gradient.

yassine.bt · May 18, 2020, 10:21

Quote:

Originally Posted by vinerm

That's correct and no-slip implies velocity is 0 at the wall, i.e., the default condition in Fluent. As far as rotation is concerned, in a 2D case, rotation is possible only around z-axis, so, $\Omega = (0~0~\omega_z)$ where $\omega_z$ is a function of velocity gradient.

so I don't need to do anything
okay thank you so much

alneama · October 5, 2022, 17:27

Hi everyone,
How to make a UDF to have a fully developed velocity profile in a 3D rectangular microchannel having a cross sectional area of 2mm (H) * 1mm (W)? I want to interpreted it to ANSYS (Fluent).
Regards.
Dr. Ahmed F. Al-Neama

AlexanderZ · October 7, 2022, 00:38

Quote:

Originally Posted by alneama

Hi everyone,
How to make a UDF to have a fully developed velocity profile in a 3D rectangular microchannel having a cross sectional area of 2mm (H) * 1mm (W)? I want to interpreted it to ANSYS (Fluent).
Regards.
Dr. Ahmed F. Al-Neama

look for DEFINE_PROFILE in Ansys Fluent Customization manual

alneama · October 7, 2022, 15:00

Hi Alexander,
I did it but unfortunately I could not found what I want (which is fully developed laminar flow in the entrance of a 3D rectangular microchannel). I have found UDF fo 2D duct with turbulent flow regime as shown in the figure attached. I would be very grateful if any one could help me.
The cross sectional area of the rectangular microchannel is 1mm width and 2mm height, while the length of the microchannel is 20mm. Assume the inlet velocity is 0.2 m/sec (laminar flow)
Regards.

May 18, 2020, 08:37	udf boundary condition (vorticity)	#1
yassine.bt New Member yassine Join Date: Apr 2020 Posts: 8 Rep Power: 5	Hi I don't know how to write udf boundary condition for vorticity .I bring to your attention that my work has been done on the nanofluid and more precisely I am working on the configuration of Rayleigh Bénard (natural convection). Fig. 1 shows a schematic diagram of the Rayleigh-Bénard (RB) problem. For the RB problem, the distance between the upper cold and lower hot plates is defined by H and the width of the top and bottom plates is defined by W. The plates’ width W is considered infinite and treated by a periodicity boundary condition. The bottom plate is maintained at a hot temperature TH=50 C whereas the top plate is maintained at a cold temperature TC=22 C. The fluid enclosed between the plates is water based nanofluid containing CuO nanoparticles. The nanofluid is assumed incompressible and the flow is assumed as laminar and twodimensional. thanks Inkedgh_LI.jpg

May 18, 2020, 09:08	boundary using UDF -vorticity-	#2
yassine.bt New Member yassine Join Date: Apr 2020 Posts: 8 Rep Power: 5	Can anyone tell me how to apply dv/dx=0 and omega=du/dy at a boundary using UDF. [v=delta(psi)/delta(x) u=-delta(psi)/delta(y)] Inkedgh_LI.jpg

May 18, 2020, 09:15	Wall boundary	#3
vinerm Senior Member Vinerm Join Date: Jun 2009 Location: Nederland Posts: 2,946 Blog Entries: 1 Rep Power: 35	Fluent uses momentum conservation and not vorticity or Poisson's equation. The equation you have is written in terms of streamfunction and implies no-slip condition. That is the default condition in Fluent and you don't need to do anything yassine.bt likes this. __________________ Regards, Vinerm PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.

May 18, 2020, 09:55	Gradient	#5
vinerm Senior Member Vinerm Join Date: Jun 2009 Location: Nederland Posts: 2,946 Blog Entries: 1 Rep Power: 35	That won't be 0 and should not be. Look at the document you shared. The condition is NOT $\frac{d^2\psi}{dy^2} = 0$ but $\psi = 0$ , implying velocity is 0. __________________ Regards, Vinerm PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.

May 18, 2020, 10:15	Noslip	#7
vinerm Senior Member Vinerm Join Date: Jun 2009 Location: Nederland Posts: 2,946 Blog Entries: 1 Rep Power: 35	That's correct and no-slip implies velocity is 0 at the wall, i.e., the default condition in Fluent. As far as rotation is concerned, in a 2D case, rotation is possible only around z-axis, so, $\Omega = (0~0~\omega_z)$ where $\omega_z$ is a function of velocity gradient. __________________ Regards, Vinerm PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.

October 5, 2022, 17:27	How to make a UDF to have a fully developed velocity profile in a 3D rectangular micr	#9
alneama New Member Dr. Ahmed F. Al-Neama Join Date: Oct 2021 Location: Mosul - Iraq Posts: 3 Rep Power: 4	Hi everyone, How to make a UDF to have a fully developed velocity profile in a 3D rectangular microchannel having a cross sectional area of 2mm (H) * 1mm (W)? I want to interpreted it to ANSYS (Fluent). Regards. Dr. Ahmed F. Al-Neama

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
How do you use heat flux as a boundary condition in a non-wall area in UDF?	JuBong	Fluent UDF and Scheme Programming	2	July 2, 2018 21:15
several fields modified by single boundary condition	schröder	OpenFOAM Programming & Development	3	April 21, 2015 05:09
vorticity boundary condition	bearcharge	Main CFD Forum	0	May 14, 2010 11:32
RPM in Wind Turbine	Pankaj	CFX	9	November 23, 2009 04:05
UDF Boundary Condition	Victor	FLUENT	2	November 6, 2003 11:37