XANSYS Message: 38577 [Go back to message list]
[bookmark on del.icio.us]
No rating yet
Rate item:

Subject: Re: FLOTRAN - Reynolds Equation - pressure
Author: Roberto Porto
Date: 2002-06-06 18:22:00

You can use two methods:

1 - use the same node numbers to both the model and the SE masters

2 - couple coincident nodes; as they will have exactly the same position, no
moments will be created in the model.

Regards,

Roberto Porto
DSC Engenharia

----- Original Message -----
To: 'xansys@yahoogroups.com'
Sent: Thursday, June 06, 2002 5:38 PM

Barman,

Here's my knowledge about using ANSYS for your interest.
I've done Reynolds flow analysis with ANSYS before, and ANSYS can handle
general thin
channel flow problems. ANSYS, however, doesn't have the Reynolds equation
solver. So, you need to use thermal conduction analysis capability since
the governing equation is the same (Poisson's equation).
Once you compare heat conduction equation and Reynolds equation, it will not
be hard to figure out
how to use ANSYS for Reynolds analysis (e.g., thermal conductivity ~
viscosity).

An incompressible Newtonian flow analysis is kind of straight forward (still
quite complex
because you have to write several macros or, external FORTRAN or C routine
to be used with ANSYS).
However, Non-Newtonian fluid analysis would not be straight forward.
You can use FLOTRAN for viscoplastic or power law model.
However, you will be highly limited to do complex analysis.
As you know, the ideal Bingham model has an inherent singularity at zero
shear strain rate, so it's not practical and I've never seen commercial
software that has the ideal Bingham model implemented. ANSYS/FLOTRAN has a
bi-viscous model instead to simulate Bingham-like fluid.
FLOTRAN also is capable of simulating Power Law fluids.
However, I am not sure if viscoelastic fluid model (e.g., Carreau) is
implemented.
You may want to refer to ANSYS manual for this.

Here's my answers for your specific questions.
>1. Is it possible to solve Reynolds Equation for analyzing the load
>carrying capacity, pressure distribution where both surfaces are
>moving with different velocity?
For Reynolds analysis, you only need to specify the relative velocity of the
two moving surfaces.
So, ANSYS can handle this. It should be specified as heat generation (or
heat source) term in the heat conduction equation (Poisson's equation).
Once you find fluid pressure, then you can use INTSRF to integrate pressure
to get the load carrying capacity. Note that INTSRF command only works with
fluid element, so you need to convert thermal element to fluid element.

>2. Is there any possibility to apply oscillation movement on one of
>the surface which has a small amplitude or in other word one surface
>has constant velocity and other surface is moving with positive and
>negative velocity (changing direction) with different temperature.

I've never done this with ANSYS, but I think it's doable without including
temperature effect.
You can use transient heat conduction analysis (equivalent to transient
Reynolds equation) to do this.
Is there a very high temperature gradient between the two surfaces?
If not, I would not think that there is large variation of viscosity in
filmwise direction.
If this is the case, then using the viscosity at average temperature would
work for most applications.
It would be pretty complicated to include viscosity variation with
temperature, even for simple Newtonian flow problem.

>3. How could I apply macro for using the Bingham Fluid, Carreau or
>Power Law for my viscosity.
As mentioned above, the simplest way to do this is to use FLOTRAN for
Bi-viscous and Power Law fluids.
For analysis with such fluid, you need to evaluate the shear strain rate to
evaluate viscosity.
Reynolds equation doesn't explicitly gives velocity field, so using ANSYS'
heat conduction solver would not help at all in this case.

I hope this helps.

Regards,
Andrew Kim
TCAD
972-995-3355
Silicon Technology Development (SiTD)
Texas Instruments, Inc.
Dallas, Texas.

-----Original Message-----
Sent: Tuesday, June 04, 2002 9:15 PM
To: xansys@yahoogroups.com

Hi all Xansys group members,

I have developed my own code written in SEPRAN (CFD-TUDELFT-HOLLAND)
for analysing the lubrication problem by solving the Reynolds
Equation, compressible and incompressible problem for 1D and 2D. Now I
need to include the temperature effect on both moving layers which
will effect the viscosity of the lubricant and I have decided to use
ANSYS/FLOTRAN version SOLARIS64 release 6.0 for this Reynols Equation.

My question are:
1. Is it possible to solve Reynolds Equation for analysing the load
carrying capacity, pressure distribution where both surfaces are
moving with different velocity?
2. Is there any possibility to apply oscillation movement on one of
the surface which has a small amplitudo or in other word one surface
has constant velocity and other surface is moving with positive and
negative velocity (changing direction) with different tempearture.
3. How could I apply macro for using the Bingham FLuid, Carreau or
Power Law for my viscosity.

Many-many thanks for all your help and in giving my some information.

Barman Tambunan
Dept of Mechanical and Manufacturing Engineering
Bundoora East Campus
RMIT University - AUSTRALIA
Email: anies@a...
barman.tambunan@m...

Posts possibly associated with message #38577AuthorDateScore
38497FLOTRAN - Reynolds Equation - pressurebarmantam anies@2002/06/04 
38574Re: FLOTRAN - Reynolds Equation - pressureAndrew Kim2002/06/06 
38577Re: FLOTRAN - Reynolds Equation - pressureRoberto Porto2002/06/06 
38597Re: FLOTRAN - Reynolds Equation - pressureAndrew Kim2002/06/07 
38636Re: FLOTRAN - Reynolds Equation - pressureBarman Tambunan anies@2002/06/07 
38641Re: FLOTRAN - Reynolds Equation - pressureAndrew Kim2002/06/09