 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
Natural Science Forum / Physics / Optics / November 2005Giant prism for apartment airshaft?
Hi ya'll I live in a great apartment building in Olympia washington and I have a project I need help with. This building has 3' x 3' x 10' airshafts between each apartment that previously provided wonderful light and airflow. They have since been loosely capped with tin roofs that allow some air to flow under the eaves of the roof. This completely blocks the light. My friend owns part of the building and lives in down the hall. He lined the shaft with silver and has replaced the tin roof with a flat sheet of Plexiglas. It leaks and blows around in a storm. My Question: Is there any sub $300 way I can better his method of getting tons of light down this shaft? (I am thinking about those deck mounted prisms in sailboats that seem to throw so much light below deck) The philodendron I place in the shafts base will thank you most warmly. Sincerely... Tor Clausen ps. I posted a picture of the current roof top here: http://www.musicalfurnishings.com/shaft.jpg >Hi ya'll > [quoted text clipped - 24 lines] > >http://www.musicalfurnishings.com/shaft.jpg How about an array of smaller prisms? Edmund is closing out some deck prisms now. http://scientificsonline.com/product.asp_Q_pn_E_3038621 would that be noticably better that silver walls? t > would that be noticably better that silver walls? The problem with that is the bounces. If the light bounces around 10 times before hitting the bottom, and 90% reflects each time, the amount hitting the bottom isn't especially big. would that be noticeably better that silver walls? t I should also add this tin roof replacement needs to be water tight as the shaft is old and no longer drains water as it should (the gutter probably ends up in bookstores bathroom sink below me) >would that be noticeably better that silver walls? > [quoted text clipped - 3 lines] >the shaft is old and no longer drains water as it should (the gutter >probably ends up in bookstores bathroom sink below me) this would be in addition to the silver walls. Mount these things in the tin roof, properly sealed, then the silver walls will let whatever light gets in on down the shaft. > My Question: Is there any sub $300 way I can better his method of > getting tons of light down this shaft? (I am thinking about those deck > mounted prisms in sailboats that seem to throw so much light below > deck) A two mirror siderostat could put over a kilowatt of sunlight down a 3x10 foot shaft. Clear sky sunlight is roughly 1kw per square meter. That would consist of one stationary mirror that if you looked up the shaft you'd be looking parallel to the earth's axis off the reflection. The mirror and a flat glass window can weather seal the top of the shaft. A second flat mirror rotates at ~ 1 revolution per day "derotating" the earths rotation. It does not need to be sealed. Both mirrors can just be sheets of polished aluminum. No fixed optical assembly can come close for throughput. You might get 1/10 as much light if the entire shaft is coated in polished aluminum (even foil) with nothing more than a tilted window on top. Figure a 4-5% light loss for each reflection. $300? Probably if you buy surplus. Making two motors track the sun with photodiodes is probalby easier than using a computer to calculate the position. Either would work. The first mirror the light hits can be north and above or south and below the shaft, which ever is easist. Less mirror area is requiried if it's below. sun * win!\ * dow!*\ fixed mirror \ * !* \ \* |* | ----> north rotating \ |* | mirror |* | shaft >$300? Probably if you buy surplus. Making two motors track the sun with >photodiodes is probalby easier than using a computer to calculate the [quoted text clipped - 10 lines] > rotating \ |* | > mirror |* | shaft Hmmm..... Incoming angle....reflection angle..... How does that change over the day? Polar mounts can be obtained from the TV and satellite radio shops... but tracking must be done with a microprocessor, methinks. Am not sure if the diode method would work, would it? w. >>$300? Probably if you buy surplus. Making two motors track the sun with >>photodiodes is probalby easier than using a computer to calculate the [quoted text clipped - 13 lines] > Hmmm..... Incoming angle....reflection angle..... > How does that change over the day? > Polar mounts can be obtained from the TV and satellite radio shops... > but tracking must be done with a microprocessor, methinks. > Am not sure if the diode method would work, would it? I wrote software which would do sun, moon and star tracking in 1982 using stepper motors for a 6809 processor. Equatorial sun position calculation to a few minutes of arc is under ten lines of code. EL/AZ tracking is about twice that. Any computer which can keep time has plenty of computational power. It's easy to find a 386 or faster used computer for $25. You can drive two stepper motors from a parallel port with <$20 worth of parts. You do need to know a programming langauge. BASIC is adequate. Sun tracking can also be done without a computer using four photodiodes separated by an X mask so that all will be illuminated only when it's directly pointed at the sun. At any other angle only one or two will be illuminated. Two power op amps and two dc gear motors drive the mirror. I'm not recommending the power for the motors comes from the photcells, though even that's possible. An equatorial suntracker has to be mounted so it moves at double the declination angle of the mirror but the same angle in right ascension. That's not difficult with with a chain or timing belt drive. An el-az mount using a "dumb" tracker would be more difficult, but the detectors couuld be fixed looking at the refelected sun image with separate sensors to "rewind" the mount each morning. My profession is building and operating fully automated research telescopes. http://www.fairobs.org Each moves to typically 1000 stars per night and centers to arc second precision to take photometric measurements. Finding and tracking the sun is not a tough job. Louis, Thanks for your ideas. I looked up a two mirror siderostat and I think I understand what you are talking about. I am looking at maybe a days bit of work, and while I do fancy myself handy, your skills far outshine mine. I perhaps at my own house I could pull such a thing off, but things need to be less "moving parts" oriented I think. We also get next to no sun in Olympia, but I do get 180 degrees of white sky that I am looking to jam down this shaft into the home office that I am working in right now. Its dark around here, so I would love to get some good light in here, I just worry that after ten feet, it will have fallen off so much. Bert, I like the simplest form of your idea, where I slant the plexiglas with a northern wall (maybe two feet) and line it with those foot square mirrors, six of em. I worry about having to much upright for two reasons, wind and blocking the non-sun grey days (which outnumber the sunny days in these parts) (Olympia) Charles, I also like your idea, because I would have the easiest time convincing the building owner to do it. He is in a bit of a bind, as the tin roofs are maintenance free and work, and some of these projects leave his roof looking horrible. My fear is that if made 10% of the tin roof riddled with those prisms, I would have maybe 15% of the light I would have with the sheet of plexiglas, but much more work and money. Am I wrong? >>> * >>> win!\ [quoted text clipped - 13 lines] >I wrote software which would do sun, moon and star tracking in 1982 >. Finding and tracking the sun is not a tough job. Thanks for the input. For tracking the sun one must have the Keplerian data for satellite nr. 1, or is there another way to track? You say it is relatively simple. Why don't you post your program here (Just joking) As for the OP problem of reflecting the sun into a window of the house, one must 1/2 the tracking movement? Angle deviation of the mirror will cause double movement angle of the reflected beam. Am not sure, time to make some drawings. w. > Thanks for the input. > For tracking the sun one must have the Keplerian data for > satellite nr. 1, or is there another way to track? > You say it is relatively simple. > Why don't you post your program here (Just joking) OK: In "C" (converted from basic in an era when computers didn't have much capability) void solar () { float halfpi; float twopi; float t; float r; float maos; float tlos; halfpi = PI / 2; twopi = 2 * PI; t = (mjd - 15019.5) * 2.73785079e-05; maos = 6.25658358 + 628.301946 * t - 2.61799388e-06 * t * t; tlos = ((maos + 0.03344051 * sin (maos) + 0.000349066 * sin (maos + maos) + 4.93169)); tlos = (tlos - ((int) (tlos / twopi) * twopi)); sun_rta = atan (0.91746 * (sin (tlos) / cos (tlos))); if (tlos > halfpi) { sun_rta = sun_rta + PI; } if (tlos > PI + halfpi) { sun_rta = sun_rta + PI; } r = (0.39782 * sin (tlos)) * (0.39782 * sin (tlos)); sun_dec = atan ((0.39782 * sin (tlos)) / sqrt (1.0 - (r))); return; } MJD is the modified Julian day with fraction calculated elsewhere from "clock time". maos and tlos are the calculated mean and true local solar time. The equations and coefficients were distilled from the "Alamanc For Computers" from the Naval Observatory. All angles are in radians. They are not corrected for the observers distance from the earth's center or for atmospheric refraction. It does account the ellipticity of the earth's orbit and the tilt of the earth's axis. It's good to a few arc minutes for this century. I just use it to calculate times of twilight and when the observatory roofs should open and close. > As for the OP problem of reflecting the sun into a window > of the house, one must 1/2 the tracking movement? > Angle deviation of the mirror will cause double movement> angle of the reflected beam. There's only one angle of incidence of a beam of light bouncing off a mirror with an equal angle of reflection, but it doesn't have to be in the same plane as the mounts which move the mirror. With an equatorial mount it is held in the plane of one axis so the reflected beam is along a meridian, adjusted with the other axis so the beam is parallel with the earths axis. The "double angle" only changes with the seasons, not with the earth's rotation. To send the beam any other fixed direction the mounts must move in two axis simultanously. That requres a bit more computation. Both axes will have constantly changing rates. > Am not sure, time to make some drawings. Before carbon or xenon arc lamps were invented clock driven siderostats were used for projection illumination. Clouds were the main difficulty. If the earlier open topped shafts provided light, why not just replace the cover with a glass window? If you wanted to be a bit more elaborate you might want to arrange a simple single pitch roof above the window facing south, and using it to protect a south facing mirror mounted at a suitable angle. In fact, you might be able to do both by mounting the mirrored fixture high enough above the window so that the window gets direct rays some of the time, and reflected rays the other. I'd think direct rays midday, and reflected/direct morning and/or evening (yeah, now I'm thinking about two smaller mirrors). By all means do something to make the walls reflective. Just replacing the roof cover with a window will give you (after the window is cleaned, of course) about 90% of the light you used to get with an open top. > My friend owns part of the building and lives in down the hall. He > lined the shaft with silver... Your friend must be rich! If you got enough light in the shaft when it had an open top, replace the solid roof with a window that is correcty sealed? Think of skylights that are in so many homes. It would stay cleaner if it was gabled rather than horizontal. When clean, yo'd get 90% of the original light down the shaft. Paint the shaft white or line it with foil and you'll gain a lot of light at your latitude. Suspend a south-east facing big mirror on the north side of the shaft so it won't cast shadows at the correct angle and you'll gain morning light: south-west facing for evening light. Make it manually rotate and tilt so you can make an adustment every couple of weeks and really gain AM or PM lighting. Your friend had the right idea, but it was poorly executed. Remember what Home Depot likes to say: "You can do it, we can help." Flashing, caulking, sealants, all that stuff will make the thing water tight. You can't just drop a piece of plexiglass over the hole and tape it down for the best results. snip how about a skylight? A well installed and designed skylight will not leak. Furthermore a skylight can be opened for ventilation whenever you want. I do not know if they make skylights with custom dimensions but that seems to be within the realm of the possible, look up the www, the yellow pages, take a stroll at Loewe's, Home Depot etc. 1 m by 1 m does (approx.) does not seem outrageous. Forget about the mirrors, the coelestat, siderostat etc. if you never see the sun with your perpetually overcast sky, what's the point. If you were in Arizona or Spain this would be a different story. >Hi ya'll > [quoted text clipped - 24 lines] > >http://www.musicalfurnishings.com/shaft.jpg Interesting problem. 3M used to make a product for just this application. It was a sheet of plexiglas into which were embossed humdreds of rows of tiny prisms. They deflected sunlight so it would go straight down a shaft (probably at noon on the equinox, though perhaps at a lower angle). This solution was tried at the Minnesota zoo in a tall building with long light shafts. The plants went crazy. Now the question is whether they still make this product or if you can buy it elsewhere. If not, go for the skylight.  SignatureBest regards, Steve Eckhardt skeckhardt at mmm dot com > Interesting problem. 3M used to make a product for just this application. It > was a sheet of plexiglas into which were embossed humdreds of rows of tiny > prisms. it probably doesn't help the original poster much but its still an interesting optical problem. I'm not sure that 3M still make the product, now that they're concentrating on their large diameter light pipes. However, the sheet of plexiglas with prisms was probably one of Lorne Whitehead's inventions. Lots of reading at http://www.phas.ubc.ca/ssp/ssp_publications.htm particularly the papers from 1982 and 1984 > it probably doesn't help the original poster much but its still an interesting > optical problem. I'm not sure that 3M still make the product, now that > they're concentrating on their large diameter light pipes. What about their light pipes instead ? AFAIK that has a very low reflective loss. Steve > > it probably doesn't help the original poster much but its still an interesting > > optical problem. I'm not sure that 3M still make the product, now that > > they're concentrating on their large diameter light pipes. > > What about their light pipes instead ? AFAIK that has a very low > reflective loss. They're good. You have to have them in the diameter they make them in, which is 250mm from memory. They're on the 3M web site. There's another company http://www.sunpipe.com/ who claim that the 3M film deteriorates in UV over a year or two steve, I found this about the zoo's skylights.. http://www.biology.mcgill.ca/Phytotron/LightWkshp1994/6.6%20Kneipp/Kneipp%20text.htm tor I'm curious about the use of a polished surface within the shaft vs. a white painted surface. I know that a polished aluminum sheet absorbs a great deal more IR than does a white painted aluminum sheet. How do the same compare for reflectivity of visible light? Scott Kelley My gut says the refective is better for getting the light down the tube, but I know not by how much. Its my understanding that a very bright white surface is the most reflective surface possible, but it is not directional, which is desired in this case. (mirrors get awfully hot in the sun) tor > I'm curious about the use of a polished surface within the shaft vs. a white > painted surface. I know that a polished aluminum sheet absorbs a great deal > more IR than does a white painted aluminum sheet. How do the same compare > for reflectivity of visible light? > > Scott Kelley > My gut says the refective is better for getting the light down the > tube, but I know not by how much. Its my understanding that a very [quoted text clipped - 3 lines] > > tor You won't find a paint or metallic coating which comes close to being as reflective as the total internal reflections of a a fiber optic cable. It's not practical to fit the light tunnel with a high refractive index medium. So the percentage of light which will get through a painted or lined pipe will depend on the the number of reflections. which is dependent on it's lenght to diameter ratio and whether the refletions are specular or difuse. For a long skinny pipe it won't be much unless the light is well collimated to begin with and the reflections are specular. With difuse reflections the light won't propagate down the pipe. With the piple lined with aluminum mirrors: and assuming a 45 degree solar zenith angle. a 1 meter light pipe eleven meters long would have ten reflections. (reflectivity)^ (number of reflections) .9 (typical for al. foil) ^10 (ten reflections = .348 That's 34.8% of the light entering the pipe. That may be enough to be useful but it's not nearly as efficient as one or two reflections directing the sunlight parallel to the shaft. Louis, Thanks for that last entry. Very clear. I am gathering that it is only sunlight that I can work to improve its voyage down the shaft. I am mostly working with overcast light, so its looking like I will just live with what light I can get - about 35% by your calculation. Thanks! Tor >Hi ya'll > [quoted text clipped - 24 lines] > >http://www.musicalfurnishings.com/shaft.jpg Giant prisms get real heavy and real expensive realiy fast. Jim Klein |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2009 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.