We're seeing more and more low cost CFD packages that claim to be suitable for almost all for buildings/ hvac / facilities cfd applications.

I have seen a report where a company used a 'cheap' CFD to model the wind flow around buildings on a project - it looks like they used FLAIR.

It has been suggested that these packages are used only to create graphics for a system that you're already 100% confident is going to work and you merely want to communicate your ideas to a client.

Has anyone used FLAIR, Microflow, or any of the other cheaper CFD packages (compared to the likes of Fluent, CFX, Star CD, etc.)?

Why should I spend $k's on CFX when there are a number of low costs packages which claim to do the same thing for much less money?

(Note i use CFX and am perfectly happy with it but this question gets asked of me - and I don't have a full answer in response...)

General purpose CFD packages such as CFX, STAR, Fluent contain a whole range of models you might not need. You can skip the whole multiphase modelling for instance. But the license of (at least) CFX is modular. If you only need CAD interface/import, meshing, turbulence models, serial (not parallel), you might end up with a stripped, cheaper, license. But whether that's cheaper than a FLAIR I don't know. I do now that a lot of development is put into the turbulence models of CFX, whether the same holds for packages such as FLAIR....

Thanks for the response - my CFX licenses are quite 'stripped down'. I'm hoping someone will be able to provide me with a good comeback as to the merits of CFX vs say IES microflo or flair.

Put it this way - if i charge $x for my work using cfx and my competitor is chargin $y for his using ies microflow... how do i convince my client they should go with me (and cfx...)?

Dear Andrew,

These days is very easy to put together a "cheap CFD" package to solve flow in a box, and fairly Cartesian friendly geometry. The problem is to offer:

1 - A variety of tested/validated physical models to approach your modeling needs.

2 - Solution algorithms that are scalable, that allow you to predict the turn around time for your calculation.

3 - Reliable technical support on a wide spectrum of physical models.

The flyer for Microflow is very light on technical information. I do not know what is under the hood in regard to discretization algorithms, linear solver technology, validation of physical models, scalability for large models, parallel computing.

You will have to contact them directly, a setup a real configuration case you have done in CFX, and let see what it takes them to provide you with something useful to you.

Opaque

Hi,

As for selling the more expensive option to clients - I suggest you do a relevant benchmark simulation on CFX and Microflow/Flair/Whatever. Pretty much every CFD simulation code I know offers free trials so hopefully this will only require a little investment in time.

If the benchmark shows that CFX is more accurate or captures physics the others don't then you have something to justify the extra cost. If not then I guess you had better start moving to the cheaper code!

(BTW: I have just been quoted for my next year's lease of my ANSYS/CFX software and it is a 35% price increase on last year. I might start also start looking for cheaper options with price increases like that. Not happy.)

Regards, Glenn Horrocks

So we finally feel the 'benefits' of one major CFD player in the market?

Bart

Hi Bart,

I assume you are referring to the 35% price increase I have just been quoted. I have asked for the justification for the increase - we will see what they come back with.

In the meantime I will have a look at the alternatives. The CFD market has broadened considerably in recent years and there is quite a large number of good alternatives out there.

Glenn Horrocks

You will be missed in the Ansys CFX community :,-(

But all is not lost, just fine-tune your bargaining skills