CAD-free CFD
 

Hello all,

Are any of you using CAD-free CFD? I have read a lot of paper about it. The idea is to place CFD upstream of the CAD in the design process. The CFD engineer provide to his CAD colleagues dimension range (length, diameters, thickness, etc...) and limits within which he can design the product according to the CFD analysis. And then the final design comes back to cfd to check it is within the ranges and results. It is reported to decrease the time to market much more than "integrated CFD in the CAD" where endless discussion occurs. This allows for a margin for errors and assumptions from the CFD side. The knowledge of these margin and errors comes from experience or expertise only. It is a more realistic approach in the manufacturing world.

What is your view on this?

Guten Abend,

Herbert

Re: CAD-free CFD
 

Every situation is unique. When applicable, I prefer this technique. Sometimes I try to get in on the proposal phase of a job so that if the job comes through, I can hand a list of requirements down to the mechanical engineers and the D&D staff. It's not as disconnected as "here's my requirements" and then down the line sometime getting "here's the result". You still need to be involved in the design process, and occassionally you're going to need to run some models to see if everythings on track at the moment, and you're still going to make changes here and there. The benefit comes in with the fact that you're giving the ME and D&D a direction to go with the system. The company I'm at has done a lot of work on electronics... occasionally there was some aero work, but the ME on the program would make some best guesses, and if it worked, an aero guy wouldn't even see the model. We'd only get dragged in way down on the design process when the proverbial sh!t hit the fan. And then we'd go in, and cause them to rethink the whole thing which was a huge hit on money/schedule/manpower... recently the company as a whole has started to realize that if they get the input before hand, then they can avoid a lot of problems. The other benefit to CFD analysis is that it's usually one engineer, and possibly one or two support people (usually a D&D who can work the models so that they can be pulled into Gambit quickly) so that it's comparatively a low cost part of the program if done before the program gets fully staffed. Most companies have used this structure for electrical / software / RF designs in the past, and it's only recently that CFD has become a well known tool that they are starting to incorporate this model with aero design as well.

Just my opinions and experiences though Jason