Jump to content
  • Advertisement


This topic is now archived and is closed to further replies.


CFD and me

This topic is 5248 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

Are there any CFD (aka aerodynamics) systems that don''t require a masters degree to understand? I''m having trouble believing that water flow through a curved pipe or over a flat plate requires a vector processor and a university tenure. Are there any intro algorithms out there for this? I know PV=nRT, but that''s about it....

Share this post

Link to post
Share on other sites
Guest Anonymous Poster
It rather depends on what you are trying to do; if it''s just some kind of simple simulation of total flows, why aren''t bernoulli''s equations sufficient ? if what you''re really after is a complete description of the flow field then CFD is probably the only solution ... and yes, a relevent degree will go a long way to making it comprehendable...

Why not tell us in a little more detail what you are trying to do/achieve

Share this post

Link to post
Share on other sites
The good news is that the answer is yes, as AP mentioned.

The bad news is that the answer is, it depends! Depends on how comprehensive you want the flow simulation to be and whether or not you want to deal with off-design/non-ideal conditions. Do you want just to be able to predict the basic flow properties (velocity) in the non-fully-developed portion of the flow (short lengths of pipe only)? Do you care about the thermodynamics? Do you care about the fully-resolved part (where the flow can become supersonic at the centerline depending on boundary conditions)? Do you care about velocity variation through the cross section (in reality the fluid is stationary at the walls due to skin friction, and a maximum roughly at the centerline)?

I personally have a Bachelors and Masters degree in aerospace engineering, and completed the coursework for a Ph.D related to CFD. You certainly can do a very basic simulation using just Bernoulli's equation (watch out for large elevation changes, which require conversion of kinetic energy into potential and vice versa). This approach will give quite reasonable results for shortish pipes where the boundary layer does not fully develop. Sophomore or Junior year Bachelor's degree classes in aerodynamics/fluid dynamics or hydraulics should get into fully developed flows and that will help simulate flows in much longer pipes. Very detailed analysis of the boundary layer (velocity and temperature profiles across the cross section) are something that would be covered in Masters or Ph.D level courses.

Of course, it is possible to teach yourself the advanced stuff, if you read the proper books/papers, write experimental CFD codes, etc.

Flat plate is a similar deal. If you're only interested in simulating the plate at low angles-of-attack (almost aligned parallel to the wind), you can use a very simple equation to estimate lift that is a linear function of angle. There are rules of thumb for this. And you can also predict drag (parabolic drag polar) and pitching moment using similarly simple equations. There are some threads in the forum archives on this. But, if you're interested in the nonlinear, behavior of the forces due to friction, onset of turbulence, boundary layer transition, separation, and hysteresis, laminar separation bubbles, etc., then things get more tricky. You can use table lookup of some nonlinear effects (assuming you can get the tables of data), and that is fairly simple and doesn't require a degree. But the underlying prediction requires a complex experiment or complex CFD calculation.

You might not think it, but the simulation of a flat plate in supersonic flow also is not that difficult as long as you are talking about frictionless flow. It actually could possibly be done on a GPU in a vertex shader. Of course, much is ignored!

Graham Rhodes
Senior Scientist
Applied Research Associates, Inc.

[edited by - grhodes_at_work on January 13, 2004 6:01:32 PM]

Share this post

Link to post
Share on other sites

  • Advertisement

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

Participate in the game development conversation and more when you create an account on GameDev.net!

Sign me up!