<html><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">Hi Mostapha<div><br></div><div>great that your model works, after all. Stupid me: If my guess at the non-working flow model would have been correct, you would have been flooded with warnings "afn not converged after 100 iterations". Oh, well. And I have noticed that res is a bit "finicky" with flow results sometimes, myself. Have you tried .xml output? Some afn data is available via that channel. The caveat is that it may slow the simulation down quite a bit and if you discover you need additional data, you have to re-run the simulation.</div><div><br></div><div>You have measurement results for the Atrium? What scope, if I may ask (and if you can / want to say)? That sounds very interesting.</div><div><br></div><div>Fine-tuning a model to meet measurements is generally a very difficult task, as there are dozens of places where you can change things. Some possibilities that come to mind (sorry, again a long list - just thinking aloud)</div><div><br></div><div>Flow:</div><div>- cp values. Did you play around with these / do you have measurements or a CFD simulation?</div><div>- leakage distribution? Do you have this? As written before, I would assume that any structure has some leakage at several places. Do you know the overall air-tightness when the top openings are closed? I think you can simulate an airtightness measurement with ESP-r, but haven't tried myself, yet</div><div>- wind speed and direction on site? </div><div><br></div><div>Thermal model:</div><div>- the meteo data (solar radiation, temperature); I assume you have an own weather data file</div><div>- convective heat transfer coefficients; maybe some alternative models are better suited for an atrium - ESP-r has quite a few models available</div><div>- long wave radiation exchange: The fictitious surfaces used for splitting up the atrium into zones for the air flow model do influence long wave radiation exchange unfavorably. However, this may not be a big issue in your atrium, depending on temperature differences between various surfaces (top - bottom?) - and can't be changed, anyway, as far as I know.</div><div>- solar distribution in "one-and-more-down" zones of the atrium. Did you do detailed solar distribution calcs?</div><div><br></div><div>(Jon, maybe you can add to the list?)</div><div><br></div><div>Best</div><div>Achim</div><div><br></div><div><br></div><div><br><div><div>On Jun 23, 2010, at 3:54 PM, Mostapha Sadeghipour wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite">Hi Achim,<br><br>By chance, I found that the stack pressure was NOT zero ever. The problem is only in the result visualisation in the esp-r result analysis...<br><br>I have tested your suggestions one by one. Confusingly, the results for the stack pressure was always zero! Finally, I have tried the exemplar model (alternative facade ...) and it was the same! I get curious to check some other variables to see how it is possible. Surprisingly when I added the volume flow rates to the selections, the stack pressures are changed to some new values! :-0 (See the attachments). Un/Fortunately, in my models' results it was the same story. ;-)<br> <br>By the way it seems that the AFN simulation can solve such networks. Hensen's thesis suggests an alternative stack pressure calculation for "an extensive network containing several internal nodes with "long" flow paths" on page 4-11 (page 58 of 197). I have tested this one as well but differences between the measurements and the simulation results still remain.<br> <br>All the best,<br><br>Mostapha<br> <br><div class="gmail_quote">On Fri, Jun 11, 2010 at 2:14 AM, Jon Hand <span dir="ltr"><<a href="mailto:jon@esru.strath.ac.uk" target="_blank">jon@esru.strath.ac.uk</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"> <br> If someone has a good atrium model that they have confidence in and which<br> is well-documented then we could add it as an exemplar model for use<br> in training workshops and/or self-study.<br> <br> Regards, Jon Hand<br> ________________________________________<br> From: <a href="mailto:esp-r-bounces@lists.strath.ac.uk" target="_blank">esp-r-bounces@lists.strath.ac.uk</a> [<a href="mailto:esp-r-bounces@lists.strath.ac.uk" target="_blank">esp-r-bounces@lists.strath.ac.uk</a>] On Behalf Of Achim.geissler [<a href="mailto:achim.geissler@intergga.ch" target="_blank">achim.geissler@intergga.ch</a>]<br> Sent: 10 June 2010 07:23<br> To: MostaphaSadeghipour<br> Cc: <a href="mailto:esp-r@lists.strath.ac.uk" target="_blank">esp-r@lists.strath.ac.uk</a><br> Subject: [esp-r] Re: Zero Stack Pressure<br> <div><div></div><div><br> Hi Mostapha<br> <br> in additio! n to the below, I think your approach with only one zone having external openings may be a general problem. The flow simulation can't really solve this, I think, because what you have is a possible flow path chain >in< the atrium (which is basically set up very good, I would think) which is singularily connected to the outside. Remember that one zone node can have only one temperature / pressure at one time step.<br> <br> Also, I would expect such a large atrium to have >some< leakage in the lower floor (entrance door etc.).<br> <br> Possible measures:<br> a) Split the topmost zone into two (or three) that each have a connect to an exterior node<br> b) also split the lowermost zone into two<br> c) as written below, add some leakage to the lowermost zone<br> <br> Do the thermal results seem plausible? Do you have a "nice temperature stack" in the stacked zones?<br> <br> Best<br> Achim<br> <br> <br> --------- Original-Nachricht --------<br> Von: "Achim Geissler" <<a href="mailto:achim.geissler@intergga.ch" target="_blank">achim.geissler@intergga.ch</a>><br> An: "Mostapha Sadeghipour" <<a href="mailto:sadeghipour@gmail.com" target="_blank">sadeghipour@gmail.com</a>><br> Cc: "Aizaz Samuel" <<a href="mailto:aizaz.a.samuel@strath.ac.uk" target="_blank">aizaz.a.samuel@strath.ac.uk</a>>, <a href="mailto:esp-r@lists.strath.ac.uk" target="_blank">esp-r@lists.strath.ac.uk</a><br> Betreff: [esp-r] Re: Zero Stack Pressure<br> Datum: 09/06/10 23:13<br> <br> Hi Mostapha<br> <br> looking at your .a! fn, I think the problem is that all your connections to external nodes are in one zone (01).<br> <br> To get an afn up and running, a basic requirement is that for every zone you can go from an external node to a zone ... to >the< zone ... to a zone ... and so on and then to a (different) external node again. I think this does not hold for your afn. Add a connection type "tiny", e.g. a crack a factor 10 or 100 smaller than your _crk types and connect the lowermost zone(s) to an additional external node.<br> <br> Then something should happen and there should be pressure differences.<br> <br> Best<br> Achim<br> <br> <br> <br> <br> On Jun 9, 2010, at 2:47 PM, Mostapha Sadeghipour wrote:<br> <br> Hello Aizaz,<br> <br> I think the problem is not the node heights in my case, however I'm not sure. The .afn file is attached.<br> <br> Regard! s,<br> <br> Mostapha Sadeghipour<br> <br> </div></div><div><div></div><div>On Wed, Jun 9, 2010 at 3:54 PM, Aizaz Samuel <<a href="mailto:aizaz.a.samuel@strath.ac.uk" target="_blank">aizaz.a.samuel@strath.ac.uk</a><mailto:<a href="mailto:aizaz.a.samuel@strath.ac.uk" target="_blank">aizaz.a.samuel@strath.ac.uk</a>>> wrote:<br> Look up how node heights are defined in exemplars>network<br> ventilation>alternative facade vent.....<br> <br> Theory can be found in Jan Hensen's thesis (section 4.3.2) available at<br> <a href="http://www.esru.strath.ac.uk/publications.htm" target="_blank">http://www.esru.strath.ac.uk/publications.htm</a><br> --<br> Dr Aizaz A Samuel<br> Department of Mechanical Engineering<br> University of Strathclyde<br> M324 James Weir Building<br> 75 Montrose Street<br> Glasgow, UK<br> G1 1XJ<br> <br> T:+44 141 548 5765<br> F:+44 141 552 5105<br> <br> On Wednesday 09 June 2010 11:28, Mostapha Sadeghipour wro! te:<br> > Dear esp-r users,<br> ><br> > I'm modeling an atrium with a closed glass roof. The atrium has nearly 16m<br> > height and it only connects to the outside through 3 cracks on the roof. I<br> > divided the atrium into 5 vertical zones, and adjacent vertical bypass<br> > zones. They are connected to each other through fictitious faces.<br> ><br> > Common orifice components connect zone nodes to each other vertically, and<br> > door component connects zone nodes to bypass nodes horizontally. The<br> > attached image shows the Air Flow Network in my model more clearly.<br> ><br> > I supposed that esp-r includes the stack effect whenever delta height<br> > between the nodes and the component is not zero. So I expected to have a<br> > stack pressure in the vertical connections. But when I check the results<br> > the stack pressure in the connections are zero!<br> ><br> > I'm also curious! how esp-r consider the stack effects and the gravity? The<br> > components dP are set manually and there is no "g" in the equations? Where<br> > I can find the PSTACK equations that is used to calculate DP:<br> > DP=PRES(IDPS)-PRES(IDNE)+PSTACK(ICNN).<br> ><br> > Thanks in advance.<br> ><br> > Mostapha Sadeghipour<br> <br> <br> <atr.afn>_______________________________________________<br> esp-r mailing list<br> </div></div><a href="mailto:esp-r@lists.strath.ac.uk" target="_blank">esp-r@lists.strath.ac.uk</a><mailto:<a href="mailto:esp-r@lists.strath.ac.uk" target="_blank">esp-r@lists.strath.ac.uk</a>><br> <div><a href="http://lists.strath.ac.uk/mailman/listinfo/esp-r" target="_blank">http://lists.strath.ac.uk/mailman/listinfo/esp-r</a><br> <br> </div><a href="mailto:achim.geissler@intergga.ch" target="_blank">achim.geissler@intergga.ch</a><mailto:<a href="mailto:achim.geissler@intergga.ch" target="_blank">achim.geissler@intergga.ch</a>><br> <br> <br> <br> <br> <br> <br> ________________________________<br> <br> _______________________________________________<br> esp-r mailing list<br> <a href="mailto:esp-r@lists.strath.ac.uk" target="_blank">esp-r@lists.strath.ac.uk</a><mailto:<a href="mailto:esp-r@lists.strath.ac.uk" target="_blank">esp-r@lists.strath.ac.uk</a>><br> <div><div></div><div><a href="http://lists.strath.ac.uk/mailman/listinfo/esp-r" target="_blank">http://lists.strath.ac.uk/mailman/listinfo/esp-r</a><br> </div></div></blockquote></div><br> <span><3_windows_legacy_StkP-01.gif></span><span><3_windows_legacy_StkP-02.gif></span>_______________________________________________<br>esp-r mailing list<br><a href="mailto:esp-r@lists.strath.ac.uk">esp-r@lists.strath.ac.uk</a><br>http://lists.strath.ac.uk/mailman/listinfo/esp-r</blockquote></div><br><div apple-content-edited="true"> <span class="Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: 2; text-align: auto; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-border-horizontal-spacing: 0px; -webkit-border-vertical-spacing: 0px; -webkit-text-decorations-in-effect: none; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0; "><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><div><a href="mailto:achim.geissler@intergga.ch">achim.geissler@intergga.ch</a></div><div><br></div><div><br></div></div></span><br class="Apple-interchange-newline"> </div><br></div></body></html>