Light pipe simulation

[Ina]

Hi,
I am new to IES, so I realise that my questions are going to be rather primitive...
I have to simulate the daylight conditions in a simple rectangular room of 14m2 (4x3.5m) with a single light pipe of 0.23m diameter. I have tried the example described in (Re: Sunpipe/Solar tube modelling on IES
http://www.iesve.com/support/search?key ... typhoto/0/). More specifically:
a. Created a “ceiling tile” component, a “diffuser tile” component with which I created a false ceiling with a diffuser in the middle and a “pipe” component, which I placed in the center, above the diffuser.
b. In ModelIT, I created a dome, made windows in all of its surfaces and placed it over the room, above the pipe.
c. In Radiance, with the room selected, I checked the “Use components” box, I set the surface properties (diffuser component=trans material with 75% transmittance, pipe component=metal with 98% reflectance, etc) and put the eye position inside the room, looking towards the diffuser.
The problem is that, in the Radiance simulation there is no daylight availability inside the room (illuminance everywhere on working plane is 0) and in the luminance image, the diffuser is bright, but the pipe (and the pipe's print on the diffuser) are dark.
I have tried making the diameter of the pipe larger, but again no daylight availability. I wonder if it is a problem of the pipe component (it has a floor that is not transparent maybe?) but don't know how to resolve it...
Any help would be useful and welcome!

[Marc Jensen]

The best way and most consistent method that I have found to model a light pipe is to not use the components and to model your roof around the pipe.


Ie create your light pipe first, glaze what would be there. Dont do it as a circle as that is a mission, draw your pipe as an equivalent area rectangle snapping to grid.
now draw in your roof around this pipe so essentially you have a glazed room in your ceiling and roof.

Remove any energy ie internal gains or air conditioning in that space and you have a simple light tube that you can control very easily and monitor what is inside it and the impact to the room etc.

[Ina]

Hi Mark,
Thank you for your reply.
I am not sure, but I have probably tried a similar solution with what you are proposing. I mean that I have created a room (the space), a cylinder above the room (the pipe - not as a component but as a room) with very high reflectance at walls and with openings on floor and ceiling and sometimes I have even tried a dome on top of the pipe (with holes on the dome and pipe floors). When the length of the pipe is close to reality (for example 1m) the illuminances inside the room are much lower than expected (by experimental data). I don't even want to try higher set ups...
Is what you were describing different from what I've tried?
Also, if anyone has any comment on the low light output, please let me know....
Thanks!

[Marc Jensen]

Hi Ina,

Can you generate an image from inside the room?

If you can render your image even quickly you can see where your light is coming in and how much.



Your image should look like that. I would recommend starting with your lowest settings initially and then increasing your detail.

If you have light entering through your light tube you should see it. I would also recommend looking at your placement etc.

If the image shows light entering your room then you will have an increased solar gain in the room or in the light tube etc.


Here is a building that I worked on last year:



Here is what it looks like inside:



Notice the light entering through the solar tube.

Having this system worked well and reduced the energy required for my lighting through the use of dimming setups...

[Ina]

Hi Marc,
I think you are talking about something more like a "light well", while I mean a light pipe, ie a very reflective tube, from 0,5-3m long, of 0,3m diameter, which might be the only opening in a room. (sorry, seems I can't post an attachment....)
I think the reason why I have very low Illuminance levels is because of the backward raytracing technique and I think a great diameter oversizing of the system would lead to more realistic results.
Thanks anyway....
I.

[Marc Jensen]

Hi Ina,

You need to also remember the software cuts off the number of reflections in your model.

Ie a particle of light will bounce off a reflective surface in reality more than three times however IES will chug if you set it to calculate large amounts of light reflects...

Probably your best bet would be to increase the diameter of your light tube and simplify your model as much as possible whilst still retaining the basics in your model.