looking for digital windtunnel...

[Eamon]

Hi,

just wondering if any of you know of any method of simulating natural ventilation in architectural models?

I think this topic could be pretty relevant as more and more designs are executed via the BIM method (ie. archicad and revit), as well as there being various sustainable design legislations coming into effect every day.

What I'm looking for exactly is a method of running (natural) ventilation simulations on models exported from archicad, something akin to a digital windtunnel, to verify various ventilation strategies etc.

I've access to 3ds max in case particle flow can manage something like this. The only other program I know of is realflow by next limit, but there doesn't seem to be any sort of support forum on their website. There is a new program (x-flow) mentioned, which would seem to be exactly what I'm looking for, but again, info is limited.

Thanks in advance

[William Alexander]

RealFlow is probably about the best solution...however LOL

 

Maya's fluid dynamics particle system would get you very close pretty quick, imho. It just 'works' and works fast.

 

The question is do you want to create the effect or actually try for a 'simulation'?

 

WDA

[Eamon]

Thanks wda,

I guess I'll have to take the plunge and download the realflow trial.

what I'm looking for is something that gets within 90% accuracy but looks convincing - it's for a thesis project next year - sustainable design will probably have to be considered...

I tried out a demo of the excellent ecotect (its acoustics analysis tools are fantastic for laymen/architects as opposed to engineers http://squ1.com/ecotect/features/acoustics ) but couldn't really get the same interactive feedback for ventilation...it does export to third party apps however

[Sketchrender]

Have you tried the new verasion of ecotecs software.

 

 

http://squ1.com/ecotect/features/ventilation

 

 

 

I have not tried it, I posed the same question about a year go, and even went and bought cinema d4 , because i saw a lovely streamline simulation of a car on there promo website , and it turned out it was all hand drawn splines , and a particle system applied............

 

anyway, I rang arup engineers in dublin, and they said they send most of their simulation out to a specialist department in England.

I asked them about Ecotec and one of them said that one piece of software that dose everything under one heading dosen't sound right.

 

So There is no quick and fast answer for this topic.

I once post this question on the microstation website a while ago too.

I was told that light travels in straight lines and the calculations are simple compaired to air flow and water.

 

Realflow could be the answer ...but is it accurate, and acredited?

 

Let me know how you get on will you .

 

thanks

 

Phil

[jackb602]

I worked for a very high-tech architecture firm several years ago, and to simulate wind behavior for a proposed structure, they sent it out to a specialized consusltant. They tested it by putting a physical model in a windtunnel. They then added what looked like salt to the windtunnel and tracked how it was deposited. Though it's not as slick as a CG fluid simulation, I think it's more accurate. If you can find the name of a consultant (I can't remember who we worked with), try giving them a call. They'd probably be happy to talk to a student.

 

Jack

[AJLynn]

We did this in my undergrad building tech lab, with a program called Phoenix Commander (though Phoenix was misspelled and I don't remember how to correctly misspell it). You either modelled in primitives in its interface or imported CAD files, you could assign objects to be hot or cold, or air flow intakes and outlets, and it would run its simulation and then spit out numbers and visuals. You had to let it run for hours (I was using a P3/700 at the time) and you knew it converged by watching graphs of the calculations progress and waiting for them to stop moving. But it could converge to an incorrect solution, and you no good way of knowing if it had. The nice thing was that you could visualize temperature as well as air flow, but aside from that it was a pain.

 

I was always told you could run the same kind of thing in Femlab. And there was a piece of software you could get free from one of the major HVAC vendors (I think it was Carrier) for doing heating and fresh air loads for a building by room, given plans and HVAC plans.

 

How technical do you want to get here? If you want visuals and something that will get it pretty much right, you can use Realflow and it Max interoperability (Realflow 4 is out and they do student pricing, but, well, it's a Nextlimit product). X-Flow is like Realflow for applications where more power and accuracy is needed, they say you can use it in engineering, and last I saw they wouldn't even say what it costs unless you emailed them.

[DavidR]

watching

[imgumbydammit]

Watching as well....I know of virtual windtunnel app for evaluating airfoils (xfoil) but not for building envelope ??

 

 

IGD

[Eamon]

Thanks for the info guys.

As it's for college work, I guess it should look fantastic, whilst accuracy isn't as important, if that isn't too flippant an answer. The methodology is more important than whether a few decimal places are incorrect - I suppose a comparison would be building a little card model of an apartment interior to measure luminance levels, but not actually bothering to take into account carpeting etc.

To be honest, I'd say between 0 and 1 percent of buildings are actually designed this way in UK/Ireland anyway, at least at sketch stage. A quick google search for brise soleil manufacturers in Ireland brought up a tech specs page with all of the projects showing a nominal 45 degree pitch for the fins! (compare with http://squ1.com/ecotect/features/shadingdesign )

Proof that sustainable design can be tacked on afterwards!

Phil, about ecotect, well this post was prompted after trying out the demo, but the display of air flow within the model is the one thing it can't calculate. If you read through the link you gave me (very seductive visuals), you can see that you need to export to a third party CFD program, the results of which you can import back into ecotect to get those fantastic visuals.

Now, there are only one, or maybe two of those programs freely available, so I downloaded this one

http://www.fire.nist.gov/fds/

but didn't get very far with it. It lacks a GUI and actually runs off a command line, which is so alien to me, it's not even funny.

But once I get somewhere with it, I think I'll purchase the student version of ecotect, but that won't be until around christmas.

Andrew, if x-flow pricing is by enquiry only, it's probably very high-end, and so beyond my means. I've emailed them anyway and am waiting for an answer. But I think they did a special offer of a free personal learning edition of version 3 through 3d world mag a while back, I'll have to look into that.

Basically I'm trying to use the computer for something more than putting in lens flares in photoshop. It's hilarious watching final year students drawing a semicircle on the site plan with a little smiley-face sun and calling it a site analysis!

Anyway guys, these inquiries will be kind of on hold, until September/October for various reasons, least of all, the fact that I'm messing with somebody else's computer at the moment, but I'll post any progress I make around then.

[DavidR]

Watching as well....I know of virtual windtunnel app for evaluating airfoils (xfoil) but not for building envelope ??

IGD

What apps do you know of for evaluating airfoils? I'm interested. Thanks.

[imgumbydammit]

What apps do you know of for evaluating airfoils? I'm interested. Thanks.

 

google for "profili" and "xfoil"

 

profili is excellent at making templates for cutting foam wing cores to be vacuum laminated with carbon fiber...fiberglass and kevlar.xfoil is part of the program and will show polars for lift/drag coefficients so you can see how airfoil changes might affect those parameters.

 

I use this for making wings for RC gliders and RC electric powered airplanes

 

48 year old kid..that's me

 

IGD

[Koper]

one should not under estimate the power of 'particle flow' from max, anything can be simulated within this process. The only prob. is that one really needs to know what you'r doing, first, in max, second, how different forces react within a real environment, but luckily one do get systems within max that can help you out such as the wind, gravity and fortex forces. Once one starts with a simulation it is quite easily updatable, depending on ones memory assigned to the simulation. and everything can be adjusted easily within the particle flow window once it has been added.

 

Check out 'particle flow' within max reference and tutorials under the help menu.

[AJLynn]

Is particle flow actually powerful enough for simulation? Can you set up laminar flow formulas and use real physics? That would be really cool. Some of my friends in school were trying to get this to happen last year and couldn't figure it out in Max alone.