CFD - Computational Fluid dynamics

[bigben2k]

I found a link to a great, relatively easy to use CFD (Computational Fluid Dynamics) tool:

http://th1a8.eng.shizuoka.ac.jp/~hea.../Sns_100c.html

Has anyone tried it? Does anyone have any comments?

[BillA]

cute, but 21x21 nodes wont get you far
Ben, I heard an old expression which you might want to take to heart:

Before you can use the 'Tricks of the Trade', you have to 'Learn the Trade'.

I have suggested this primer many times, but here it is again:

"Using Computational Fluid Dynamics" by C.T. Shaw

had you the knowledge in this book, you question would not have been asked
and I can, and do, assure you that without at least grad work in this area you will not master it
(now I will hear from the engrs with 'only' BSs using CFD, but such is generally true)

very heavy sledding

[gmat]

Hmmm i've got way more than a BS but still i would'nt be able to use a CFD program at its limits, or even properly... This is a specialist thing. I wouldnt touch that without a lot of good reading and some evening courses

[BillA]

quite so

I have unlimited access to this software, and not a clue as to how to even define the most simple of parameters

Ben, you can look at the Jackel's pretty colored plots and never imagine just how much went into the setup, processing, or evaluation
- this CFD stuff is much more than learning some new words, it is an entire 'body of knowledge'

and even then one has no answers, mathematical constructs must be verified by testing
- the vendors will tell you how easy it is, do not believe them for a second

I don't really want to 'get into' your qualifications for this, but if you need to be asking -> you don't understand even the scope

leave CFD be, good thesis or dissertation work

[bigben2k]

Ya'll lack some optimism...

From everything I've seen, I agree that I certainly won't be able to understand everything about it, nor what makes it up, but...

I need something that'll help me visualize some elements.

I've been mucking with Saints2000 this afternoon, and here's what I found:

Software: it's buggy, and crashes unless everything is done right (i.e. it's not idiot proof)

21 by 21 is a joke. It's nowhere near accurate, especially for odd shapes (anything that's not rectangular).

It is limited to 2D (+time) simulations.

There doesn't appear to be any units to the input parameters, which appears to be by design, which would mean that it's all "in proportion". Since I'm dealing with fairly small bits of material, I am effectively introducing an even larger margin of error.

It will however allow me to get a rough (albeit very rough) idea of what the turbulence might look like. It should also allow me to get a "good enough" understanding of CFDs, so that when I try the next one, it should be a bit easier to input parameters.

As for how it relates to Radius, it doesn't help me one bit, but it will help me to visualize (read: get a mental picture) the flow, turbulence and (maybe) heat dispersion a bit more.

I personally would encourage would-be block makers to play with this.

What I was really looking for, is a free 3d modeling tool, so that I can calculate the pressure drop. I'm still looking.

[bigben2k]

Here's another one I'll have to try sometime:
Sinda/Fluint (specificaly, the FloCAD module)

http://www.crtech.com/


Links to other links:
http://www.cfd-online.com/
http://fluid.power.net/

Googling for "CFD" brings tons of results...