CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   Self study of CFD! (https://www.cfd-online.com/Forums/main/2486-self-study-cfd.html)

 Reza Besharati August 16, 2000 15:57

Self study of CFD!

Hi I am trying to join the active comunity of CFD pople. I need help in order to plane the self learning schedule in a reasonable way. I have gethered some books during the last 10 years and believe me or not I am ready to fire the rocket! I am also a Mathematica user and programmer and I prefer to study the topics in conjuction with Mathematica capabilities, I think it is a good idea to sent my rssume and list of the books if some one is willing to help me do the tremendously difficult job. If any body wants to help me by any kind of suggestion I do appreciate in advanec and I am waiting for helping hands.

All the best for all of the best people

PS. I think I like the conceptual discussions first and then I think about formulation and problem solution techniques.

Reza

 John C. Chien August 16, 2000 21:37

Re: Self study of CFD!

(1). There are many aspects of modern cfd analysis, namely, the 3-D geometry and mesh generation, the turbulence modelling, the numerical solution algorithm, and the graphic animation and presentation of the results. (2). For a simple problem, such as flow over a 2-D airfoil or flow through a 2-D duct, the geometry is nothing but two curves which represent the upper wall and the lower wall of the airfoil or the duct. For this simple typical flow problem, the gemoetry and the mesh generation are not complicated at all. For self-study, this is the type of the problem to begin with. (3). For 3-D problem, such as the extension of the above simple flow cases, you will have to study the 3-D geometry modeling by parametric surface patches concept. In other words, you need to learn the foundation of CAD and the 3-D mesh generation. (4). The turbulence modeling is the core of modern cfd research. This part can be included as part of the study of the governing equations, such as the Navier-Stokes equations. There are 0-equation models, 1-equation models, and 2-equation models to study. (5). If you are interested in writing the cfd code, it is important to understand the numerical formulation and the solution algorithm. In this area, you have finite-difference methods, finite-volume methods, pressure-based methods, density-based methods,etc... Here, 2-D is always the starting point. Whether to use pressure-based or the density -based, depends on the problems to be solved and your personal interest. For aerospace applications, density -based transient approach is always used to take care of the compressibility. (6). Then, once you have the results, you would like to put it in easy to read graphic form, so that the useful features of the results can be easily deduced from the megabytes of the numbers. (7). Like testing, there are simple cfd problems as well as complex 3-D cfd problems. (8). My suggestion is: pick a simple problem as a vehicle to learn cfd. In this way, you don't have to spend a lot of time in each step, from the geometry to the finishing color plots.

 Mehdi BEN ELHADJ August 17, 2000 06:12

Re: Self study of CFD!

CFD people need a strong "Mathematica capabilities" to make best results. In all cases the idea is the most important thing for any work and Mathematics give possibilities to make wonders but in general it's very important to begin with a simple problem or study in order to undrestand steps of CFD.

Try to see the other forums of CFD codes and their web sites.

If you have problems you can contact me.

 Sebastien Perron August 18, 2000 08:21

Re: Self study of CFD!

Instead of using mathematica, I suggest writting your own CFD code and use a not too complicated numerical method (such as simple family + staggered structured grid + upwind scheme). 

 John C. Chien August 18, 2000 20:03

Re: Self study of CFD!

(1). Yes, good idea indeed. (2). I just want to add a few possibilities. Before writing your own code to solve Navier-Stokes equations, it is a good idea to learn the solution methods related to classical elliptic equations (Laplace and Poisson equation), Wave equation(hyperbolic equation), and parabolic equation(boundary layer equation). (3). Some numerical methods books in partial differential equations even have Fortran codes listing included. These types of simple equations have well defined solutions.

 Reza Besharati August 18, 2000 21:24

Reply to mr Chien's Re: Self study of CFD!

I realy appreciate your sharp and up to the point suggestions. What made me to write my question into forum was to have the chance to express my zeal toward commitment in any sort of CFD code development.

Indeed I tried to find some sort of link to people who might have a specific subject to work on and in this way learn the fundamentals of the specfic area in more detail and depth. This is what I wanted to realy say about alternative to the form first I decided to call for help on. I think I have the tenacity and suitable resources and of course an open mind to try to do some fairly good jobs in academic CFD problem solution.

I do thank all of the poeple on the forum for their attention to my question and I do welcom any kind of project definition for problem solution and code development activities that helps me to seriusly begin the path toward being a CFD specialist some day in future.

Reza

 John C. Chien August 18, 2000 21:50

Re: Reply to mr Chien's Re: Self study of CFD!

(1). Well, we welcome anyone who is interested in cfd, good, bad, ugly, new, and old. (2). You may want to put a printed copy of your goal on the wall, so you can refresh your mind from time to time. (3). It is nice to know that there is still someone out there who is interested in cfd.

 John C. Chien August 20, 2000 20:57

Re: Reply to mr Chien's Re: Self study of CFD!

(1). Your e-mail have received. (2). It seems to me that you have two goals in mind when posting the message. (3). But since the title of the message is "self study of CFD", we are thinking that it might be the main subject of dicsussion. (4). You also have expressed your interest working in cfd field with cfd colleagues. If this is your another goal of the message, then I would suggest that you post another message and make the title clear to the readers. In that way, you can easily get your message through. (5). otherwise, in most cases, cfd forum is just like a guessing game. (6). I am just guessing that you are interested in locating someone who has the time to work as a mentor in cfd area, or locating a cfd project which you can get involved so that you can study more about the cfd. If this is so, please post another message and state clearly what you are trying do do in the title. I am sorry that, my answer to your question was based on the title of the message of self study in cfd.

 Joel Cambo August 21, 2000 12:02

Re: Self study of CFD!

I can give you an example. I have a friend from China, arrived in U.S. back in September of 1998, with a Textile Engineering background (means not much fancy maths and physics), and barely spoke English. He worked with a professor in a local university. The professor had no CFD background. The project my friend worked on required some numerical calculation. So he picked up a couple of books, downloaded a few codes from the www, took one course in CFD. In six months, he wrote his own 3-D code, ran more than a dozen benchmark cases. By early this year, he has published 3 papers in CFD journals, submitted another 3. He got his MS degree a few months back and started to work with a CFD company. I would say he basically self study the CFD. He told me all he needs is a couple of books to start with. Then go to library at least three times a week and read most recent CFD papers on the daily basis. Of course, also spending 6-8 hours in front of computers. From his experience (and some of his friends), he said it is quite easy to self study CFD. So good luck.

 Ravishankar August 21, 2000 18:59

CFD for coning and rolling motion of projectiles

Hello:

Can u help me understand what role pitch-damping coefficient and roll-damping coefficients play in the aerodynamic stability of projectiles and the role of CFD in it?

 John C. Chien August 21, 2000 20:10

Re: CFD for coning and rolling motion of projectil

(1) you should be able to find several websites covering the subject. Just type the key word " pitch damping" into the search engine, such as Yahoo. (2). "damping" means resistance to the pitching motion, when the aircraft or the like is rotating along its pitching axis. If the aircraft has only one wing (no tails) and the axis of pitching is centered through the wing span, then I guess, you don't have damping. (3). I think, it is like driving a car with hands off the steering wheel, or riding a bicycle with both hands off (the front wheel is positioned in such a way that it resists to change in directions). The damping will reduce the pitching motion and keep the vehicle more stable (also hard to get out of its current course. bad for emergency, good for stability)

 All times are GMT -4. The time now is 15:11.