I am looking into CFD programs for the architecture firm that I work for, the powers that be are interested in a low cost product that can have a autodesk model inserted in.

The results that we are interested in are solar heat gain on facades and internal heat gain that can result in venturi effect cooling. these being displayed in model still shots or mpeg format for viewing.

I have looked at airpack and while it seems to be top of the line the cost is scaring the bosses off.

If there is anyone out there that could give me some suggestions or point me in a general direction it would be great.

I can point you in a general direction, but it may not be what you had in mind: I would carefully consider the two options to either pay for a CFD program, or to have the work done by a contractor. The way I see it, your decision should be based on your employees' current expertise in CFD applications, and on how important CFD is for your firm. If your employees have only little CFD experience, and CFD is really just going to be a minor temporary part of your firm's work, I probably wouldn't spend a lot of money on buying a software license, training, support... etc.

However, if you have made up your mind on buying a CFD program rather than contracting out, look for software that's specifically written for architectural applications. There are some CFD codes out there which are quite versatile (= do a lot of things you'll never need), quite powerful (= complex), quite robust (= inaccurate and/or slow), and very dominant in the CFD market (= expensive). Don't look for those.

Ask around in other architecture firms that have experience with CFD software, search online for university research programs in that area (not to get their research codes, but to get free advice on good low-cost commercial codes).

I think Mani's advice is basically sound. What I would add to that though is that this doesn't really sound like a simple problem. In other words, you either have to go for an empirically tuned "design" code, which is not really CFD-based, or alternatively you have to invest quite a lot of resources into developing a competent in-house CFD capability using commercial CFD software. The third alternative is just to farm the work out to somebody who understands the problem.

Generally (i.e., not just applied to architecture), having a good in-house facility might be a very idea in the long term, but don't underestimate how much effort it will take to reach that capability. There are some neat "Engineering Fluid Dynamics" software packages around that may make CFD look easier to get into, but don't make the mistake of thinking that you can just plug the CAD geometry in and get good information out of it. There's a lot more to it. Having someone around who really understands his fluid dynamics and heat transfer makes a world of difference.

I did similar simulation in 1999 using PHOENICS, don't know how much it is now. But it was an open code, you better have some CFD knowledge to use it.