Hello Francisco and all the others,

I won't be able to answer to your questions. However I'd like to offer my view to these issues:

I've recently spent a a couple of hours researching this issue on forums as well as playing a bit with NIST FDS for my work. I still cannot forget Andrew's fantastic presentation at Seattle 2,5 months ago. The peak moment of that was when he showed prepared model with CFD data. I believe it was 3D Analysis Grid, and by interactively changing X, Y, or Z coordinate everyone really get the sense of airflow and different pressures inbetween the buildings. That was really cool!

I'm not sure but I think this analysis were done by WinAir4, which is for some reason not available for any of us. So what about NIST FDS???

The way I see it's maybe possible to run it:
1. model stuff Ecotect, export as DXF,
2. use external utility DXF_to_NIST FDS,
3. be really skilled on how to set up all the scripting parameters and of airflow physics in general (it looked at least as "tough cookie" as Radiance, which has a long, long learning curve).
4. Visualise it with SMOKEVIEW utility (noone on forums said being able to load the results to Ecotect).

From my as architect's point of view it's not worth of it to try to do it right now. Maybe you know the proposed generations of building simulation tools by J.A. Clarke. For me it looked that CFD within packages like Ecotect right now is in "the 2 generation" valley. Although reality is achived at pretty good , usability curve has fallen dramatically low.

For example, let's compare it with Radiance. I think that Square One has done incredibly good job in enhancing Ecotect/Radiance (or Ecotect/DaySim) accessability. Transition in majority of cases works well, Control Panel is great, Image Viewer is great. There is still a big space for improvements, and you still need a lot of knowledge how to set up everything up and about lighting in general, but this is workable. I did one daylighting study and actually am quite happy and optimistic with a team of Ecotect/Radiance/DaySim. I wish CFD could be at the same level, hopefully soon, but as Wikipedia states the biggest problem CFD industy is facing is the lack of skilled people....

I think that the important question is what we can do right now? I think that there are a couple of things we can (and I'm sorry, I'm not HVAC engeneer, if there is any I'd like to hear his/her opinion):
1. At Weather Tool check the importance of using Natural Ventilation as a passive strategy (e.g. how often is hot/humid weather).
2. At Weather Tool/Ecotect draw a windrose and for (summer) months get a clue how can we catch these desired flows in our designs.
3. Draw old-school wind diagrams by a pencil and use old-school rules of thumbs from the literature (for example at Brown's and DeKay's "Sun, Wind & Light", pp. 17-21, or Norbert Lechner's "Heating. Cooling. Lighting. Design methods For Architects").
4. Analyse insolation (Ecotect is good tool for that) and be sure to be aware of (to reduce) heat gains.

As always in simulation there is a need for experience, intuition, ability of interpretation, healthy skepticism, design vision, ...

Best regards,
Zigurds.

P.S. I'd also be glad to hear other opinions.

Hello everyone,

BTW on Squ1 Wiki there is a page regarding air flow analysis:

http://squ1.org/wiki/Training_Air_Flow

There is intended to be a place for tutorial how to export to FDS (as well as more edge-cutting stuff). I'm looking forward to see it

Sincerely,
Zigurds.

Zigurds and others,

Thank you for the input Zirguds, i agree a lot on your comments.
I have worked hard myself trying to play with NIST.
Results: after some time figuring out how to work with the software (see attached .txt for my own "note to self" instructions), I am still unable to re-import any of the output files into Ecotect. So I use the Smoke View program, but I havn't got the hang of it yet.

As for WinAir - yes, Andrew uses WinAir4 - it works quite well in Ecotect and no...it isn't publically available (see post.

However, Ecotect never really intended to be its own CFD engine either, and as with many other third-party software, RADIANCE being a good example, it can only provide its export functionality to the best of its ability - not quite there with NIST, but I feel it is reallllly close....

BUT - As Zirguds mentionned...CFD really looks cool alright, but I fear that its application could actually be limited to small scale project. I see there are some test showing wind flow moving through an urban environment. These, again, look good, but a key user input into the CFD (with WinAir or NIST) is the wind direction. While this could be more or less estimated on an exposed terrain, i find it difficult to assess on a complex urban site, as well as its initial speed (alo a user input). So as far as my skills go, i can only see myself using CFD for small test of cross ventilation within a room or such experiments.

As Zirguds pointed out, I like to overlay my windrose on my Ecotect model by going to Calculate > Prevailing Winds. Then again, one has to be pretty careful with that rose...wind is usually recorded at some high distance from the ground level where speed is higher and not as affected by ground-borne turbulences.

Best,
Olivier

I would like to thank everyone for all the descriptions for possible workflows. Although, it is a little discouraging to spend so much time figuring out that the program cannot not really do what it claims to do. See http://www.squ1.com/ecotect/features/ventilation
I know there's a lot of continual improvement happening with the program, but customers need to have a better idea of how they should be spending their time while skilling up on the program.

Going forward, I would like to see a Ecotect to CFD interfaces with a wizard that guides the decision making process of the designer. I agree that the Radiance export is a great example of how Ecotect can be used with more specialized outside applications. There's no doubt that CFD images are cool, but bad assumptions going into the analysis just leads to more bad information.

Does anyone know if it is possible to get a non-official/beta release copy of Winair4? I am intending to use it just at a primary research level and would not be using it for a professional services.

Thanks for all the assistance.

The current verson of ecotect can export input file for FDS(file_export), but the user must modify input file by manual.
I think the function is not very perfect.
another problem, because ecotect can only setup a grid in a time ,so I think it is just fit for exterrior and simple interrior conditions.

I have read that Ecotect can't calc solar gains to a zone behind 2 walls or sheets of glass with a zone seperating those walls (Double Facade). I hope I'm wrong? CFD would be good but once implemented will there be other things that hold the software back from truly modelling spaces thermally correctly when air movement is invloved?

Can Ecotect calc cooling gains from wind driven ventilation for scenarios when someone opens a window and a breeze is available. Other software on the market can do this without the need for complex CFD integration.

Can Ecotect do this? If not, is full complex CFD a valid pursuit yet over vetilation gains from wind data already available in weather files?

>> ecotect can export input file for FDS(file_export),
>> but the user must modify input file by manual.
>> I think the function is not very perfect.
This is definately the case -- it can be done, but in the current v5.5 it is certainly not as easy as other exports such as to Radiance etc.
The reason for this is we originally wanted to use WinAir4 from Cardiff University, however when ECOTECT users started making enquiries about the code for download etc. they for whatever reason decided that they did not want to support its public use anymore.
We then changed tack and went for NIST-FDS, for which there has been a basic implimentation since v5.5.
v5.6 which is at beta and soon (Sept 17) RC1 (Release Candidate 1) will have a more refined implimentation of the one Zigurds saw in Seattle.
So to answer the original question for this post; v5.6 with a full release of this at the start of October 2007.

Simon I will head off now and see if I can find the other post you refer to with the solar stuff... if I get lost along the way (or more likely sidetracked) please just post a URL (node reference number) here and I will chase it up.

http://squ1.com/node/1965

Caroline, I note that there is a suggested way to get around this problem in this article is it OK?

Getting back to ventilation, is it too difficult to put in the necessary code to see the effects of wind driven ventilation into a space through a window?

My gut feeling is that if it was implemented into ECOTECT then designers would have a more complete tool set considering the advantages of knowing that a building that may not be performing thermally could be if wind driven ventilation was incorporated into the thermal analysis. I think it would be likely that you would gain a much larger user base...

Seeing that a building will be positive negative front to back concerning pressure is it possible to say, if I open this window I can safely assume that another window in the zone open on the other side will vent the zone and the corresponding change in the temperature and ACH will result in a cooler building...? You could say that two windows will need to be open in a zone and work on this premise for a simulation....

I understand though that this may lead to bouyancy effects not being accounted for. Ultimately this would lead to spaces such as double facades requiring additional code to account for the stack effect, but maybe in a later version....?

I'm intrigued as to how ECOTECT could have some of these features and how it would improve its usebility across an already large user base now that it seems that Winair is unavailable.

Simon,
The way Ecotect presently deals with natural ventilation and its cooling effect was described to me by Andrew some time ago.
I am attaching a sample page here, as I was not able to find the orinal message...

Best,
Olivier

Craines,

I tested the new version just after the download, but the problem is still remains, and I could not import the data from FDS.
Is it cause because I don't know the way or is the problem not solved yet?

Best,
Mostapha

Hey everyone!

I'm not that familiar with CFD yet but want to start getting used to it now. But before I download NIST-FDS and start working on it I'd like to know: can somebody really confirm that it is the right program to work with if it comes to building performance simulation?
For somebody with experience in that field it might be a stupid question but I don't know what a Large Eddy Simulation (LES)is and the program seems to focus on "lowspeed,
thermally-driven flow with an emphasis on smoke and heat transport from fires".
So, can anybody confirm that FDS is suitable for checking night time cooling, the effect of the wind pattern, ... ? Or is somebody of the opinion that FDS is not suitable besides for calculating fire scenarios?

Thanks for help,
Thomas