Fiber optic lighting?

[cyberwollf]

just LED lunars?  Yea that could easily be done all analog. Phototransiter, NPN transistor, and potentiometer, $2 in parts

[cyberwollf]

Heres circuit, that uses a comparator and relay (not sure if local radioshacks still sell them)

Tune VR1 to the desired darkness.



For the work and price, a $5 cheapo timer is my choice.  Unless for some reason your lights are on a random sequence??

[Dankicity]

Welp just ordered a 22.5" fresnel lens that's already curved @ $130 -- only one I've see that's not flat.  I read that I should put the fiber between the lens and focal point but I'm not sure if the light I don't catch will melt the cable.  Which is another reason why I opted for a second lens that shines in on the cable... which I think I can just use a plain on magnifying glass set at the right distance.

Also just purchased 10' of the EndGlow 50 strand .75mm per strand cable @ $3.45 ft.  Using the 22.5 inch lens I'm thinking you could support a bunch of these cables, estimated at a minimum of 8-10' per cable -- going pretty much straight down.  So about $500 for the fiber optics for a full rig.

Lighting reference -- http://en.wikipedia.org/wiki/File:Us_pv_annual_may2004.jpg  -- we get about 4-5 Kilowatt-hours per square meter per day.  Now how much light is that really?

And I finished up the concept for my home lighting which might be pretty close to what a reefer might light.


That break light circuit board had two rings and I figured jigsawing out the inner ring would let me slide in the cable and only power the outer ring.  Hoping to encase it in a standard light fixture.  Planning on ordering those today.

Still debating LEDs as I'm looking to get some of those Philips ones that are used in the Solaris sets, which are said to be the strongest -- until we get printable OLEDs from Kinko's or somewhere


Oh yea,  I do need to have a bottom cover for the fresnel lens because unless it tracks the sun the focal point will subtly move through the day and could possibly start melting or even burn your roof/house/occupants.

[cyberwollf]

White price did you find the phillips at?  I think you can get Cree XR-Es for $7 a piece, mounted on a metal core PCB.

[Dankicity]

I've been looking all over for them but most places I find just tells you about them and then links to someone that supposedly sells them but doesn't.  But I did realized that those high flux LEDs aren't just the LED but a platform with an LED on it.  Which is about 10x larger than a normal LED -- so my light fixture is out of the question for reefing.

I need to go back and read those posts you suggested.  I'm betting that will help me find a solution.

[cyberwollf]

yea the "star" they come on is a metal core circuit board to disperse heat.  You can't just run them like that either.  The stars have to be mounted on a larger heatsink.  3 watts is a tremendous amount of power to put through a doped semiconductor junction.  It generates a lot of heat and it has to be dissipated or the junction will destroy in seconds.

Nano-reef has more information than you would ever want.  DIY is catching on pretty fast in the nano world.  few hundred for a light that will put out 300+ PAR, and suppose to last 50,000 hours as long as cool properly.

[Dankicity]

Yea, I definitely need to do some more research on aquarium lighting.  Kinda feels the same as web-development,  just seems like there is always something your missing and should have known or had years ago... more so the latter with web.

Really I just want that lens so I can start playing with fire again.   ahh those were the days...

I just had a crazy thought of putting fiber optics in your sand bed.  I remember my bud saying something about how detritus(sp?) builds up under the top couple inches because it doesn't get light below that point -- Think I'm wrong on this.  If you put light down there would it help act as a larger filter?  Not cause as much trouble when siphoning the sand?

Or on a more realistic note: having shelves of frags stacked in a single tank with fiber optics lighting each shelf.  No more one shelf gets this light and this gets that... etc.  That would probably be the biggest benefit of using the central light source and fiber optics.

[Dankicity]

Videos old but the basis of my whole thought... basis... it is my whole thought.

(why do some of these videos not play or say video not found?)

"2 fibers (look like 2mm strands)  = 60 watt light bulb."  They "have 127 strands" fed from a 50-60" parabolic mirror.  == 3810 watt light bulb

So that's roughly 340 .75mm strands which would carry -- the equivalent of -- about 20 watts each -- or was it 10 watts...

Leaving my 22.5" lens capable of roughly equaling a 1430 watt light bulb -- depending on a wild list of variables that I don't have a clue on.

Their mirror has a UV and IR filter in the mirrors focal point reflector/mirror (haven't found what they call that..yet) which cuts down on heat in the fiber optics. Still yet to figure out why this is good besides maybe cool cables means more light. Narrator doesn't sound like they gave a good reason for it.

They also say that they use GPS to track the sun... LOL, you seriously need a GPS chip in that thing to track a massive ball of light above you? ... maybe use 3 cheap light sensors and get a Dev to script up something ONCE.

Cyber.. how tough would it be to make sure LEDs balance out any fluctuation with the natural light?  Close to doing the lunar lighting sensor?-- outside having


.. LEDLvl = TargetLvl - LightLvl; ?  Man I wish it was that simple

[cyberwollf]

I could probably work up a circuit that could compare what a photoresister sees to a bank of LEDs.  If you tune it to sunlight at noon, varrying number of LEDs could turn on during cloudy times to supplement.

As far as tracking the sun, I think you could get away with a single axis of movement and timer, and adjust the base to the trajectory of the sun maybe once a month.  As long as you are in the proper direction.

[cyberwollf]

wow, I should of watched the video before i posted   That was impressive.  The fiber you ordered, Is it of the same quality in the video, or is that some super expensive stuff he has?  Starting to make me think its possible.  There are several Folks on RC that can do light spectral tests and whatnot once you get it going to see if there are any deficiencies in photosynthetic ranges.

[Dankicity]

I'm assuming that the one I bought is of a lesser quality but they are both used for lighting and I'm hoping that the difference is < 5%... 10% max which still leaves me with at least 1000 watts coming through.

Actually been downloading courses trying to find a couple lessons that deal with fiber optics.  I know they all use the same drip method for stringing up the cables (thanks "How's It Made" but the quality of the plastics is probably a biggie....I get that it similar to good and bad chocolate in that one type melts to a grainy paste while the other melts to a smooth liquid.

[cyberwollf]

The only classes I had dealing with Fiber was Physics of the incident rays of light.  for each material there is an incident ray angle that with all very close to 100% is transmitted and 0% reflected (used to get light into the fiber) and a second angle at which the light needs to be reflected inside the fiber so that 100% of light is reflected and very little it transmitted out (loss). I think these only come into play during the manufacturing of the fiber.

[Dankicity]

http://www.atmosphericzone.com/images/uploads/media/az-tech-endglow.pdf

Spectral Attenuation Max = 250 DB/KM  -- still trying to understand how this relates -- laymen -- to light.

Wikipedia does a decent job of highlighting the differences but I'm still missing some specifics.

http://en.wikipedia.org/wiki/File:Optical_fiber_types.svg     -- Single mode looks to be the best but the one I ordered is Graded -- If I read it all right.

[cyberwollf]

if running through 20ft of fiber = 0.006096 kilometers, so thats 1.524 dB of attenuation? (is that total attenuation?)  Not too bad, in power applications you use 3dB of losses as the Half power point (point at which you have lost half of your power). so you are below that, and dB is logarithmic so you are pretty well off.  The pic you posted was in reference to the impulse response of fibers.  That is much more important in the analog communication world (you have to know the impulse response so you can determine how it will affect your modulation) than "light distribution"  When does the lens and fiber come it?  I'd like to check it out.  There has been alot on interest in a club PAR meter, so that might be an option later for tests.

Also, is the fiber your bought actually hollow as the wiki pic suggest? or is solid like the DIY fiber video you posted?

[Dankicity]

Yea, I believe that is the most loss found when tested between temps of -40c to 60c.  Also of note was that the cables shrink... falloff starts at about 35c at 1 % to beyond 5% @ 60c and higher.  That could cause some problems and probably why they filter out the IR, since anytime things expand and contract you probably loose some structure of the fibers...cracks etc.

The PDF shows it as having a pvc core surrounded by the fibers and then casing.  I'm guessing that I'm going to have to strip a couple inches of the casing and core so that I can compact all the fibers, which could be a real pain when using a bunch of cables that then need to be recovered. -- that could spin up that loss to 3db and higher real quick.

I'm using the lens as a replacement to their parabolic mirror (same prices range) -- also skipping the 2nd mirror where they have the filters.  Thinking that having the curved fresnel lens will let me catch the sun without tracking it -- as much.  If not I'll have to go back to a design more like theirs, using the IR filter but not the UV -- if no problems found that way.

-- Just finished watching the Berkley Physics 10 lessons on google.  Sadly I've watched them before but I remember it having a lesson that specifically dealt with fiber optics... but of course I didn't find it.  -- strike that breif mention
http://video.google.com/videoplay?docid=-7784318168349672868&ei=gJGZSdzrH43m-AH-gNGBDQ&q=Physics+10+-+&hl=en#24m26s

[Dankicity]

Fibers came in ... no core (whew)...  lens will probably arrive tomorrow or Monday.   

It's just sweet how you can put one end under your shirt and still see light coming through the other end.  Took a 3" magnifying glass to it... a 4" or 5" would have been about the right scale... but the principle works.

[Dankicity]

Lens came in today.  Quick $100 lesson why a parabolic is better... Focal length.    The fresnel lens has roughly a 34" focal length.  It focuses all 22.5" into about an inch but I'm not sure how many would like a 3'x23" cylinder sticking above their roof/backyard.   That and the focal point varies, somewhat wildly, with a moving light source even though it's domed.

I was able to get most of the light (60% or so) from an ordinary LED focused onto a single .75m strand that I had led into the hall.  The single, uncovered strand was able to match the LEDs strength but only within about a foot from the wall.  I'm hoping that is just because it's just a single -- .75mm -- strand without anything to direct the exiting light.

I just ordered two 24" mirrored acrylic pieces that are supposedly easy to form in an oven ($40 for 2 compared to $100+ for a single parabolic mirror).  As well 12' of a single 3mm strand.  A single 490ft spool of 3mm fiber would about cover this and everywhere I look it's priced around $380.  Giving you about an inch in diameter bundle (36 fibers) at 12' in length.

You want bigger fibers because the casing that they send the fibers in are JACKED and you're better off with double sided tape and arranging them yourself (light interference from tape?).  I  used 20 strands to match what I got with the 50 strand cable just by binding them together more tightly...  and the fibers in the cable didn't like staying even on both ends even when gripped tightly.

[edit] 3mm cable 0.2 DB/m  ... 0.8 @ 12'

[lazylivin]

Dankicity, been enjoying your thread. Keep the info and your progress coming.

[cyberwollf]

I don't know exactly how much light we need here, but there is a mythbusters show on a Archimedes weapon burn off. They make several mirror variants that focus light without parabolics.

[jeremyt]

you should post the pics of  the cable  pretty sweet looking

[Dankicity]

Thx Lazy.  I started getting a bit worried that I'd be that guy spouting nonsense like a fire hose

Just watched that Archimedes Revisited -- Just so happens to have been on TV  -- That mirror cutting doesn't look that bad but the two guys had the same results with a much smaller parabolic mirror and it had a much shorter focal length.  With the flat mirrors you're not compacting the light like with a parabola,  so the focal point of flat mirrors is always the same size as the mirrors you used.

The guys idea of spinning plaster to get parabola form is a great idea.  Use a string to trace half of the shape onto cardboard, cut it out and tape the edge.  Put some plaster on a turn table and use the cardboard to get the right shape as it solidifies.  Sand the plaster to a smooth finish.  Then just place the mirrored acrylic on the shape and place in the oven.. or outside with my lens (why I got 2 pieces).  Mirrored Tape or small mirrors might work as well.

Only now you need the 2nd lens that focuses the light back onto the fibers from above the parabola.  I need to looks around for what can be use to let the IR out and still reflect everything else so I don't cook the fiber.  Hopefully they sell acrylic like that.....

[Dankicity]

Left is the lens on top a dinky lamp.  On right is the result of a single unfocused LED shining on 20 uncovered .75mm strands.  Taken from my iPhone while holding everything in place.

good article on those hybrid lighting guys.
http://www.photonics.com/Content/ReadArticle.aspx?ArticleID=33892

[Riderc82]

I've been following the thread it's over my head but I'm interested in how the finally product will come out.

[cyberwollf]

As for the IR, There should be some low cost gel filter for photography somewhere.  I know all digital camera have an internal IR filter.  Shooting IR film used to be pretty big few decades ago. See whats in the photography world that could be put before the focal point.  that way you could still filter it out, but not burn through the gel with the main power of the focal point. 

[Dankicity]

Found some better cables - 2% loss per ft... 0.12 db/m.  Now I just need to find where to order it.  Assuming these guys only sell spools.
http://www.advancedlighting.com/products/fibercable_megabrite.html

and I think these are the types of filters that I'm looking for.  Not sure if it's narrow, small or wide band.  Need to read more.
http://www.jdsu.com/products/custom-optics/products/bandpass-filters/filter-narrow-bandpass.html

It has to only reflect the visible spectrum.  Photography lenses let the visible light through and absorb the IR, so they're backwards... at least that's what I understood from the ones I looked at.  I'm guessing I'll have to order one.  I did manage to see one that did look close, 2" lens for $40.  Hope I can get it to let light through from above as well -- looks like what the hybrid guys did.

If this cost $1,000 total to get running how long would it have to run to be economical?  If it could (ex.) replace 8 x 4' T5's for a 100 gallon tank (4'x2'x2' I think).

Fixture - $400,
Fixture Lifespan - 5 years

Ballast # - 2x
Ballast Cost - $40
Ballast Lifespan - 1 year

Bulb # - 8x
Bulb Cost - $30
Bulb Watt - 60 watt
Bulb Lifespan - 6 months

Electricity Cost - $0.08 kw/h  (pretty sure I'm off on this number)
Electricity Durr  - 12 hrs (10hrs on display & cumulative 2hrs for reverse cycle sump)
Electricity Usage - 5.4 Kw/h a day ...  (seriously only 43 cents a day?.. $150 a year?)

Rough Total = $800 a year to light a 100 gallon tank w/ 25 gallon sump... let me know how off I am.  In all probability actinics will still be needed but who knows how badly.