CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   CFD: getting started (http://www.cfd-online.com/Forums/main/3435-cfd-getting-started.html)

Titus Mathe May 9, 2001 07:09

CFD: getting started
 
How do I get started on this CFD field? My knowledge on CFD is almost nill. Which is the best web site or book or journal to read for a start? I'm a chemical engineer by profession. I don't have access to any CFD packages.

Titus

John C. Chien May 9, 2001 11:45

Re: CFD: getting started
 
(1). Take a look at some websites listed in the forum-Resources section,by organization, company, or schools. (2). If you visit these places, you can get some feelings about the sample results of CFD process. (be careful about your conclusion, most of these result may not be accurate. they are just samples) (3). You can also take a look at the BOOKS section, and pick a popular entry level book on CFD. (4). To know CFD, you need to know the geometry modeling (something like CAD, but no the same), the mesh generation, and the numerical algorithms to solve the Navier-Stokes equations or governing equaitons.(this part is the core of CFD, and you must have some basic skill in the numerical analysis in partial differential equations.) (5). The other way to do at the same time is to find a simple 2-D code so that you can read it, or do something about it in programming. (6). good web-surfing to you.

ALI_ZARIF May 13, 2001 10:25

Re: CFD: getting started
 
HI I THINK IF YOU WANTS TO WORK WITH CFD IT IS BETTER TO START WITH ANSYS PACKAGE. OR IF YOU WANTS TO WRITE CFD CODES ITS BETTER TO PRACTICE WITH FORTRAN . BUY.

S.P.Asok May 17, 2001 02:01

Re: CFD: getting started
 
I myself being a newcomer can I add that it is better for a newcomer to restrict himself/herself in the beginning to anyone of the following CFD methodologies:Finite difference,finite volume and Finite element.

rupa dutta May 25, 2001 00:31

Re: CFD: getting started
 
hi

My suggestion will be :

(1) Do NOT go for any software package at this stage. (2) Start with basics - momentum, continuity, energy equations. (3) A book on Transport Phenomena by Bird, Stewart and Lightfoot (you must be knowing , being a chem. engr.)is a good book to start with. (4) When you are clear about the basic equations then only start reading numerical schemes. (5) Initially go for comparatively simpler schemes - finite difference(Taylor's series based) or finite volume (Patankar's book on Finite Volume is relatively easier to understand - but read it very carefully- it is not as easy as it will appear at the first reading) , Anderson and Pletcher's book is good. For Finite Difference- Karnahan, Luther & Wilkes's book is good. (6) develop a code (7) benchmark - Driven cavity flow

2ND Stage (1) books not dealing with numerical analysis but useful in an indirect way - Kays & Crawford , A. Bejan , Bermister, and others (2) take up a practical problem

* ( if you have never done any numerical modelling before - i would suggest you first solve numerically a heat transfer problem (conduction only)to get used to numerical techniques, discretisation, coding)

John C. Chien May 25, 2001 03:00

Re: CFD: getting started
 
(1). It is a good suggestion. I think, I like it.


All times are GMT -4. The time now is 06:38.