I have been trying to model ETFE cushions for EP and Radiance. For Radiance I am introducing manually results as produced by LBNL but for EP I have my doubts if a representation of a single layer using “windowMaterial” component is enough as ETFE cushions have a different behavior than that of a glass window. Especially when it comes to model more complex ETFEs with extra (shading/printed) layers and varying behavior during time I think that any simulation that is currently available is not representative enough. Any ideas/thoughts?
I am attaching a paper “ENERGY MODELLING OF ETFE MEMBRANES IN BUILDING APPLICATIONS” from H. Poirazis et al explaining the behavior of ETFEs (p. 701) with with a frit intermediate layer as a reference.
For Daylight modeling you can use Radiance’s image-mapping. I published a paper recently that discusses the process (http://www.simaud.org/2015/program.php#a-new-approach-to-modeling-f…). They haven’t send it online but I can email you the paper if you’re interested to read it in details.
For EnergyPlus, I would say the best solution is to generate a BSDF material using the pillows and image mapping process that I mentioned above and apply that to the surface. I’m not an expert in either creating or using BSDF materials though. Andy Mcneil has recently published a tutorial on BSDF which might be useful for this case.
Hi all, thank you for replying. Mostapha thank you very much for both links they are very useful. I am not familiar with BDSF materials either but I will try to set it up for E+. Can I use the “radBSDF” material component to export it also as an EP material or I should export it in another way? Finally is there any chance you publish the custom component you created for this study in HB?
Hi Tasos (Sorry I misspelled your name up there!),
I have never used BSDF in EnergyPlus but I think the structure should be different. Looks like you can use Window to generate BSDF definitions for EnergyPlus (https://windows.lbl.gov/software/window/7/7.3.4/w7_releasenotes.htm). Then you should be able to copy the string and apply it as a construction to Honeybee surfaces. Please keep me posted. We can see if there is a better way to implement it to Honeybee.
I developed the component while I was working for the company and technically they own the code. I may re-write a similar component in the near future but for now I don’t have the ownership to make the decision.
Thank you for your reply, indeed you can create. I exported some material already and will apply them. I will update the post soon. Is a pity but understand the copyrights issues! One thing that bugs me is that the evaluation of the results is more based on common sense rather than accuracy since I am not really sure if E+ will actually reads the created BDSF correctly. Do you have any sort of checkpoints for this? for example when you create material (in HB) and write them in the material library how do you check if E+ runs a correct simulation with the new material? (except from a qualitative evaluation)
I was wondering if you had any success with linking BSDF materials to EnergyPlus. I would love to be able to use the vast database and ease of Window 7.3 in my simulations.
As for the non BSDF idf files that window can create, is there an easier way of importing them in HB than simply copying all the values manually to a new material construction?
I still haven’t tried as I got stuck in some other projects but is first on my list for the upcoming week. I am not sure though what to do with new materials .idf’s that I have generated. Only thing that I have tried is to copy the generated Text file inputs (from the idf Text editor) into GH like Chris is doing in this example but still need to test if it works properly. As for the nonBSDF material I would try to create a common directory and save the properties in a .csv and read them in GH from there to assign them “automatically” to new constructions? Still though a workaround and probably not less of a work than inputting values manually.
I am trying to model an ETFE canopy and I was wondering if someone has radiance material definition created for a ETFE. I will be trying with Window and BSDF file but thought I’d ask first.
I tried importing from WINDOW with the new components of the new release and I get results in the right directions. My biggest question mark though was the actual modelling of the cushions in WINDOW as I couldn’t find a direct way to model the ETFE layer itself. If you have any updates with this please let me know, I am highly interested
Do you mean you managed to import typical glazing with Window+BDSF? I think I’ve managed that as well. But yes, the problem in this case is that there is nothing in the database of Window for ETFE. For my part, what I did was a very simple custom radiance material in HB, using quite high numbers of transmittance (smth ETFE is famous for?).
I’m lucky enough to be doing outdoor studies so I mostly care about shading of an area almost 30m below. In which case I’m assuming small impact by modeling them as a simple radiance surface. Sorry I can’t help more. I will look into it in the future, perhaps I am lucky and I am able to drill some info out of a supplier.
I also did not use Mostapha’s method since it involves a number of custom made components and radiance functions I don’t have the time to learn now. However, I was wondering if anyone has modelled ETFE as simple radiance materials to see if I’m close.
For the printed pattern I used intermediate panel surfaces made from Lunchbox and made them opaque (for the first test). Is this crude method anywhere close to what you’re doing? Is it smth I can use for a simple outdoor radiation study?
Prior to assigning the frit pattern, did you model the cushions and assigned its layers to a trans material? Also may I ask how did you define the trans material properties?
@mostapha curious if you’ve rewritten and added a similar functioning component to the custom component you’ve outlined in the paper? (Also is there a video showing an example honeybee script using any time of frit?) Thanks!
One thing that bugs me is that the evaluation of the results is more based on common sense rather than accuracy since I am not really sure if E+ will actually reads the created BDSF correctly. Do you have any sort of checkpoints for this? for example when you create material (in HB) and write them in the material library how do you check if E+ runs a correct simulation with the new material? (except from a qualitative evaluation)