All I found out F3D actually have very limited post-processing capabilities. for example cut plane contour plot can't overlay in surface contour plot in 3D. no solide temp and flow filed cut plane....so wonder if F3D user eve request any enhancement from flowscience?

Oh, yes they certainly do! And along with work on its own post-processor along those lines, FLOW-3D also has direct interfaces to CFD visualization packages such as Ensight, Fieldview and Tecplot. Compared to those, FLOW-3D's post-processor is very simple and easy to use.

Thanks for feedback, but i believe no one is willing to spend addition money just for post-processing at lest not for my company. In addition, those pro-processing software mentioned is not cheap as well. this is major drawback with F3D. I agreed F3D post-processing is easy to use but it is not sufficient for example: allow user to specific data extraction points in 3D domain and 3D countour plot as mentioned early. So are you using 3rd party software for post-processing?

no, i am not using these packages although i have used them.

There is a reason they are expensive - it takes time and effort to make them what they are. In FLOW-3D it is always a balancing act - there should be basic post-processing capabilities, but can't make it grand since the main effort goes into CFD. And then a direct reader to the data files is provided, so people can interface to it, if they want to. Having said this, I believe, work is going on to enhance the post-processor along the lines you mentioned.

Michael seem like you are from FlowScience, are you? I'm actually agreed on comments posted early that it is not a smart choise to spend extra money to those expensive post-processing software, expecially with current world economy situtation. One thing i would like to point out is user friendly doesn't mean limited/simple capablities, as F3D always seeling user friendly point, but with very limited solver and post-processing capabilities if compared to others Generic CFD codes.

Now you got me, Jack! I agree with the post-processing limitations in FLOW-3D, but what do you mean by very limited SOLVER capabilities?

Michael

Michael the limiation that i'm refering to as below (only highlighted few of them)

1. Solver is not capable for steady state problem, even though F3D published some alternative way to do it. but overal solver speed is slow 2. Solver is too dependent on TLEN lenght for turbulent flow model. results can be significant out if worng TLEN values been used. Fluent have more options for turbulent and stable. 3. No implicit Split langragian options make simulation work limited by advection time step. 4. Mesh aspect ratio of 1:2 or 1:1 make it difficult for model consist of very small and huge geometry. Multiblock mesh error also limiting factor.......

Michael pls advice how many CFD developer availble in flowscience to develop GUI, pre-processing, solver and post-processing?

These are good observations, Jack, but I would not call them severe limitations. Besides, we are working to remove some of them, including the TLEN dependency and adding implicitness to the Lagrangian VOF solvers.

Some other things you mention are a matter of philosophy rather than technical shortcoming. FLOW-3D is an essentially transient code. Accurate modeling of free surface is paramount, hence stress in explicit advection solvers. Then meshing. Every meshing technique is a compromize of ease of use, accuracy and required resources. I think we are in good shape here.

In short, I can't imagine FLOW-3D's limitations are more out of wack than those of other code's.

What do you use FLOW-3D for?

FSI has 25 people total.

Michael what is FSI stand for? you mean total software code developer is 25 people only? I used flow-3D and others CFD codes (Fluent, CFDesign & Star-CD) for various transient, steady state VOF and conjugate heat transfer. Btw, F3D has the limited post-processing capabilities as compared to others code mentioned here.

Sorry, Jack, FSI is Flow Science, Inc., the company that develops FLOW-3D. It has 25 people TOTAL, including developers.

I am interested to know what kind of physical and industrial processes you use FLOW-3D for? Are you in the US?

Michael it is very suprise to know F3D is a very very small company with total of only 25 ppl. so how your company compete with others CFD company that have easy >100 developers today?

There a few other successful small CFD companies around. You don't generally need many people to create good software. In fact, I think, too many people working on the same code make things harder. What do you think?

Michael I don't really agree with your statement, CAE trend is moving forward to solve more complex physic/problem and Full couple thermal-mechanical-fludis-electric will be a future trend. So, suggest you try or compared others CFD codes such as Fluent, CFD-ACE+, Star-CD, CFD2000,.. before draw to your conclusion.

Jack & Michael i'm did conducted CFD tools benchmark some time back before downselect one the meet my requirement. I do agree with Jack comments on Flow-3D and significant improvement is needed as compared to others CFD codes that mentioned.

ED

Jack,

Thank you for the advice, but it is becoming a little bit overbearing an somewhat pointless. Let me just conclude this discussion by saying that FLOW-3D has unique capabilities that other codes do not come close to matching and we are moving forward too.

You still did not answer my question on what you do with FLOW-3D? I would like to know where these opinions are coming from?

Hi Ed,

Could you please be a bit more specific?

Michael I'm actually expect more open mind discussion and believe others CFD users that have similar experience as mine can share it in this forum for more feedback.

I agree, Jack.

Jack, I am sorry, I agree with Michael. Software is not only interface. We use CFDesign and FLOW-3D. They are different things. Where is the gravity in CFDesign? It does not exist. Don´t you think it is a huge limitation more that post processing?