CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   Understanding CFD Theory(Behind the buttons and meshing) (https://www.cfd-online.com/Forums/main/147537-understanding-cfd-theory-behind-buttons-meshing.html)

Jeeloong January 23, 2015 07:44

Understanding CFD Theory(Behind the buttons and meshing)
 
Dear CFD Experts.

I just completed my Master Degree last year 2014 and currently doing PhD research on both computational and experimental work on aerospace component. During first few months, I learned about ICEM meshing which I did not learn it during my final year as my final year project was purely experimental no CFD. I did took up CFD module on my final year (Level 4 module) which I learn to use the software to model steady flow in 90 degrees sudden expansion using k-e, k-w, RSM and LES model but not on meshing as well as some theories(which in my opinion that the course only offers maybe a few percentage of CFD theory!)

However, I am still in doubt about all this theory and things behind CFD. Despite having the knowledge(Very basic) it is still insufficient. This week I did a lot of meshing with many incorrect features such as does not take into account of Y+ and many assumptions (I ran sector model 60 degrees to save computational efforts) instead of 360 degrees (3D model). I thought the near wall is only to be refined after doing mesh independency and I was told they come along together.

Can someone enlighten me on how to gain a bit of organization and way of have a better practice in CFD. I know that Y+ is non-dimensionalized wall distance, experimentally viscous sublayer lies in region <10. But I am so confused on fluent reporting Y+ of let say 35 and why do we need to switch on Enhanced Wall Function or should we refine the mesh near the wall domain. Is there a guide where I can read on the theory.

I understand that CFD solves Navier Stokes Equation but sometimes I am confusing about those kinetic/turbulence terms. I did took up Turbulence topic and understand that dissipation of energy etc. But there are so many things going around now and I am kind of (So) lost. :confused:

Hope to hear from you all.

Thanks :)

Sincere Regards,
Jee Loong

spl January 23, 2015 09:00

Hi Jee Loong

The y+ necessary for your simulation depends on the turbulence model you are using. Some turbulence models use wall functions (k-e) and therefore a larger initial cell height is acceptable where as others require finer meshes at the wall to resolve the flow. This is covered in the fluent theory guide.

As for understanding the NS equation I suggest you have a look at Turbulent Flows by Pope. There is also a very good series of lectures on you tube from Boston University called "Computational Fluid Dynamics".

Out of interest where are you studying and what is your PhD research topic?

Regards

j_02 January 23, 2015 11:12

Quote:

Originally Posted by spl (Post 528781)
There is also a very good series of lectures on you tube from Boston University called "Computational Fluid Dynamics".Regards

This lecture series is very well done. It covers theory AND implementation, so you can put together some sample codes. The only downside is that it is largely based on the Finite Difference method, which is not as popular as the Finite Volume method. That being said, go take a look at these lectures because they are very helpful.

I am going to sound like a broken record, but I would highly recommend Versteeg & Malalasekera's An Introduction to Computational Fluid Dynamics: The Finite Volume Method. This book is a great introduction. It covers the Finite Volume method in detail and provides a very clear explanation.

The Versteeg & Malalasekera work also includes a chapter on turbulence that is a good introduction. As mentioned above, the work of Pope is much more detailed with regards to turbulence.

FMDenaro January 23, 2015 17:44

understanding well CFD theory requires a mix of fluid dynamics, numerical analysis and infromation technology...is a long path.
As good basic (and simple) textbook, I suggest also Peric & Ferziger

Jeeloong February 3, 2015 08:38

Quote:

Originally Posted by spl (Post 528781)
Hi Jee Loong

The y+ necessary for your simulation depends on the turbulence model you are using. Some turbulence models use wall functions (k-e) and therefore a larger initial cell height is acceptable where as others require finer meshes at the wall to resolve the flow. This is covered in the fluent theory guide.

As for understanding the NS equation I suggest you have a look at Turbulent Flows by Pope. There is also a very good series of lectures on you tube from Boston University called "Computational Fluid Dynamics".

Out of interest where are you studying and what is your PhD research topic?

Regards


Thanks, I am currently doing Hexa mesh for the domain. Also I had borrowed Versteeg book as recommended by many. I did not go into much about the theory during my 4th year as my FYP was not mainly CFD. Will try to look at the those you recommended thanks :)

Hi, currently I just started doing research with University of Nottingham Rolls Royce UTC department on analysis of gas turbine transmission system. My research topic is looking into fluid dynamics in transmission system. I spent my 1st and 2nd Year at Malaysia Nottingham and final 2nd years at UK campus for my Master.

What occupation are you doing by the way? :)

Kind Regards,
Jee Loong


All times are GMT -4. The time now is 05:43.