I have a skylight shaft that I've like to maximize the light coming down. At the bottom is a stained glass window. I could paint the shaft sides gloss white. But what about silver? Is there some reflective sheet metal that I can buy that is even better? A glass mirror would probably be a bit difficult to get up there.
Thanks, Don .
I think glossy white paint or silver or even mirrors will not pull more photons out of the sky than plain white paint. White paint will do the job.
-B
Mylar mirror?
Bright and shiny. I saw one of the DIY shows where they installed a skylight that used a round flexible duct to send the light from roof to the room ceiling. It was bright silvery coated like a mirror. Maybe you could get some highly polished stainless steel sheet metal or a bright foil lining. I have no idea where you would buy it, but the silvered Mylar film would work well. The stuff they make balloons from or line potato chip bags. Ed snipped-for-privacy@snet.net
It would take a very good quality mirror to reflect more light than a good white paint. Cheap mirrors may only reflect about 70% or less of the light.
Buy a high quality white paint and look for one with a high LRV (light reflectance value). IIRC, matt white should actually have a higher reflectance than gloss.
Mike
Sure, but I'm looking at a table that says white paint has 80% reflectance, vs 86% for aluminum foil and 92% for aluminized Mylar. The mirrorlike Mylar can minimize the number of bounces in the shaft, compared to diffuse white paint, further increasing the intensity at the bottom. A south-facing reflective sunscoop over the skylight can also help.
Nick
It is not a question of pulling photons out of the sky. It is a question of moving the ones that you do have in the most efficient manner. There are much better materials than white paint to do the job. Depending on the particular pain and gloss, it is maybe 75 to 80% efficient. I don't have the specs on a mirror, but I'd guess it to be over 90%.
Remember the foil reflecting material used over car dashes? Use that or similar for max reflectance. However, max reflectance may not give you the max output at the bottom because they are so directional. Think of a mirror...you have to angle it around to reflect the source light where you want it to go. For your purpose, a non specular (not glossy) white or aluminum paint would probably be best due to the usual shape of a skylight shaft.
-- dadiOH _____________________________
dadiOH's dandies v3.0... ...a help file of info about MP3s, recording from LP/cassette and tips & tricks on this and that. Get it at
Considering you can see through aluminized Mylar, I have my doubts about this claim.
Consider what they cover telescope observatories with. The domes are almost always painted white with a high-titanium content paint. They are trying to reflect as much heat and light as possible in order to keep the observatory from heating up during the day. When have you ever seen them cover it with a mirror finish?
Only if you can get it nice and smooth.
Mike
Paints can be up to 90% reflective. These are high-quality titanium oxide based paints, of course. There are other oxides as well, but the Ti stuff is probably easier to find.
Mirrors are highly variable, depending on the quality of the coating. Getting over 90% is quite difficult and expensive. I remember paying quite a premium to get a telescope mirror that was over 90% - the standard quality (still higher than, say, a bathroom mirror) was in the mid 80s.
Dollar for dollar, paint is a much better deal and easier to install.
Mike
Not according to the usual lightwell calculations.
I doubt that. Can you find a specific example that makes your claim true?
Nick
10 RHO=.9'lightwell wall reflectance 20 W=4'lightwell width (feet) 30 L=8'lightwell length (feet) 40 C=4'concentration factor 50 EC=.8'concentrator efficiency 60 TFLUX=8000*W*L*C*EC'total flux entering well (FC) 70 FOR F = 1 TO 8'floor number 80 H=8*(8-F)'lightwell depth (feet) 90 R=5*H*(W+L)/(W*L)'well cavity ratio 100 E=EXP(-.0443*R/RHO)'transmission 105 PRINT F,E 110 ATTN=ATTN+1/E'cumulative attenuation 120 NEXT 130 ATA=ATTN/8'average attenuation 140 PRINT ATA'average attenuation 150 AREA=32^2*8'floor area (ft^2) 160 ER=.9'secondary reflector efficiency 170 PRINT ER*TFLUX/ATTN/AREA 180 ER=.6'eg 2 wall bounces, and a bottom reflection... 190 PRINT ER*TFLUX/AREAFloor Lightwell number Transmittance
1 .0056940 2 .0119144 3 .0249304 4 .0521655 5 .1091535 6 .2283978 7 .4779098 8 1Average diffuse well attenuation: 41.9
Illumination from diffuse well: .268 footcandles
Illumination from mirror well: 60.0 footcandles(?)
Nick
Full sun is about 10,000 footcandles. We can see in 0.1 FC, ie we can "see through" something that's 99.999% reflective.
That's not easy to do.
I disagree. Wrinkles make little difference.
Nick
What is the paint they use when they paint flat roofs? Mine was just painted silver, and checking out the job just now I find it very reflective. Certainly paint would be the easiest. Are these titanium oxide paints readily available in my local hardware store?
Don .
Can you line a skylight chase with mirrors (or even specular reflectors) so they reflect the skylight opening to the ceiling opening? Didn't think so.
-- dadiOH _____________________________
dadiOH's dandies v3.0... ...a help file of info about MP3s, recording from LP/cassette and tips & tricks on this and that. Get it at
Think "non-imaging." Direct sun aside, the question is: "Can a lightwell with diffuse reflective walls move more lumens into a room than one with specular walls, given a skylight that collects a certain solid angle of diffuse isotropic sun?"
Intuitively, light rays enter the skylight from all directions. If the lightwell is vertical, vertical rays from overhead sky pass all the way down the well with no reflections in either case, like shining a laser down a pipe. Slightly off-vertical rays are reflected from the walls far down in the lightwell. More tilted rays hit the walls nearer the top.
Specular reflectors pass these rays downwards with little intensity loss. Diffuse reflectors scatter the light all over the place--down, up, and horizontally to adjacent walls...
We could make this more mathematical, if you like.
Nick
I can see thru some aluminized mylar in a dimly lit room. My guess is that these are no more than 50% reflective. High- reflective aluminized mylar is harder to get and more expensive. If you're going to recommend Al mylar, you have to be specific about the quality and reflectivity.
My solar filter is nowhere near 99.999% reflective and I see the sun thru it dim and safe. Don't forget that there is absorption. My solar filter - nickel-steel coated glass - gets quite hot when in the sun. The titanium-white painted mask around it doesn't.
It's a myth that mirror finishes (metallic coatings, polished metal etc) are great reflectors of heat and light. Good quality white paint can outdo mediocre mirror finishes easily and are often a lot cheaper.
Mike
Roof paint. :-)
The full moon looks almost white, but is something like 10-15% reflective and is quite black (check it tonight - it's a blue moon, BTW).
Better check the actual LRV of the paint.
Many good quality white paints use TiO2. Cheaper ones use clays and stuff. Contact the paint manufacturers for this info. They can tell you which ones are the highest reflectance. This data is available for architects and interior designers.
Mike
Agreed. But the difficulty would be to get the specular reflectors to collect the light from source and direct it to where desired. Keeping in mind that it is a skylight in (presumably) a home. Actually, if I were the OP and wanted to light a stained glass window in the ceiling I probably would have scrapped the skylight in favor of artificial, controllable light.
Not me, I = R is about my limit :)
-- dadiOH _____________________________
dadiOH's dandies v3.0... ...a help file of info about MP3s, recording from LP/cassette and tips & tricks on this and that. Get it at
Think LOTS of optical fibers.
Or a sunlamp in the shaft.
Hmmm. Can one get a REVERSE photocell?
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.