I accidentally invented a device to collimate any light source into a concentrated beam... what should I do with it?

I've been umming and ahhing about what to do with this for the couple of weeks since I accidentally invented the thing (I was working on a solar energy device), and I thought you guys may be your usual well-spring of help and advice. Basically I seem to have designed a device which will collimate an extended light source. Any light source. So you could, for example, turn a 100 Watt light bulb into a 100 Watt laser. Or a square meter of sunlight into a kiloWatt. I haven't made it, but simming on the computer gives an optimal blow out rate of about 1/2 meters per kilometer, so over a range of meters or tens of meters it's pretty much as accurate as a normal laser. The only upper limit I can see on how much power you could put through this thing is when the materials start melting, which would probably be pretty high. There's really nothing fancy about it, it's just a shape. You could print it out easily on a 3d printer and fill it with some kind of settable glass for the total-internal-reflection. It would weigh about as much as a 2 litre bottle of water and be about the same size. I was concerned about it being weaponisable, but even at a kiloWatt the lethal range would probably be less than 100 meters, in which case you might as well just get a gun. I don't know if anyone's already done something like this, I can't find anything online, and people kept telling me it was supposed to be impossible. So, um, yeah...

sort by: active | newest | oldest
daved86 months ago

Already done. It's called a "lens". In other news I invented a device to make almost any object literally frictionless on a flat surface. ... what should I do with it?

It's a rigid sheet with two horizontal cylinders under it placed side by side and you put the thing you want transported on top of the rigid sheet. I can't find anything online, and people keep telling me such a thing is impossible. So, um, yeah...

freemerk daved84 months ago

How come? Is your invention like a sled with cilindrical blades?

no, I was being ironic. What he descibed was a lens. What I described is one of the oldest devices in the world, a sledge with 2 rollers.

JohnathanW155 months ago

not an expert but I seem to remember from college astronomy lab and diy flash light builders forms that the moment you add mirror or lens you have losses. so claiming a 100watt light will be a 100watt laser makes you sound like a bit of a nut. also, i hope 8 years after you are posting this all lights you have are consuming something more like 5-20 watts if no get some LEDS bro

DeannaB321 year ago

What if you use microwaves to slow down the light?

How does light slow down light??

You get two single photon light beams very close to each other.
And if you are lucky enough they become interested in each other and form a relationship.
Just not sure where to find a male and a female beam of similar age....

I hear dissimilar ages are more accepted in germany

MikeT2496 months ago

Hi, am keen to hear if you have come closer to building a prototype or getting a patent? It would be a nice trick to build a 3D cutter of sorts.

RickV4610 months ago

I need something like this for my light show! Care to tell me more?

Yonatan241 year ago


A magnifying glass, perhaps?

hi, first off I know absolutely nothing about this but perhaps I can try to have one made. I have recently discovered I need to columate my blue filtered diving lights in order to take pictures of sea life which is phosphorescent under certain wavelengths. Have you made one yet? I will ask around here to see if someone could make one if you send the parameters, if that's OK. Thanks

I've not made one, will hopefully get around at some point to doing up a 3d model which could be printed or similar. No idea if it would actually work.
paolodime1 year ago

Hi SolarFlower, very interesting. I'm trying to create a light for video shooting, and I've been struggling for a long time on a problem that maybe you help me solve. I have a light source made of 4 powerful white COB (planar, lambertian) LEDs, that form a 80mm x 80mm light emitting source. What I want to do is to concentrate this source into a collimated beam, with the final purpose of using a floating Fresnel lens in order to adjust the angle of the final beam between, let's say, 10 and 60 degrees. This is very useful when you have to deal with studio lighting. Maybe you could help me? I'm spending time and money with little results... thanks! (and sorry for the poor English :))

zingo1563 years ago

I am extremely interested in your design, I have been working on a similar device. I have a very large concave mirror, it is about 1 meter in diameter, 1 square meter of light hitting earth = ~1000 watts. Exactly like a large magnifying glass if used with the sun will instantly ignite most items. If you could collimate the light from the focal point you could potentially make a laser. It could be used as a weapon: Imagine a giant concave mirror in space (if to heavy maybe a fresnel lens) if it could be collimated you would indeed have a true Archimedes death ray, of course I would prefer forwarding humanity instead. We could use this to beam energy to a specific point anywhere within line of sight. It could be very useful for heat/energy in remote parts of the planet or even for space exploration.

A laser is not a collimated beam of white light.

I am not sure why I said laser, I know quite well the definition of laser. I should have said something similar to a laser. It would just be a collimated beam of light. I am still thinking this is not possible. Quite a while back I made a (quite amature) solar pumped ND: YAG laser using the same 1 meter mirror. Similar to the one listed here: http://www.laserfocusworld.com/articles/print/vol...

More here:


Actually, a laser is a highly-collimated beam of light, and as far as a white laser goes:


In order to qualify as a laser, the output has to be coherent and is usually monochromatic (natural linewidth-limited). You won't get that from any old light source.

Clark57 Clark572 years ago


Is anybody going to reply to my comment that i made like 4 months ago???

Make an instructable.

EstiaanJ2 years ago

I have read your post and as an expert in electro-quantumdyanmic physics I am confident that what you've come up with is a device that we've thought theoretically possible for about 30 years now. Despite our theories we've not been able to stabilize the quantum fields produced by the Heisenberg-tesla waves. Our computers are simply not powerful enough to analyze the pseudo-geometric patters and compensate for the flat nature of the earth.

I would love to talk to you about this, I believe our research can greatly contribute to your invention.



OK, I'll bite, what are Heisenberg-tesla waves?

Heisenberg – Tesla waves,
more commonly known as Zero Point Energy (for historical political reasons I do
not wish to go into) [E=hf/2] is the
result of Planck and Einstein looking at differing quantum problems concerning
light and not realising that the mathematical formulations they devised were
not the same [bosons and photons respectively].....

Planck formulated his results and equation for spectral line bosons [E=n.hv]
Einstein formulated his for photons of energy momenta [E=hf] with both terming
their quanta as a 'frequency' of quanta...

Tetryonic geometry we can now see that the true relationship and formulation
for both is n.hv = E = hf.... where bosons are ODD number equilateral quanta
and photons are EVEN numbered quanta of mass-energy momenta

continue using the same old equations for spectral lines bosons and EM
photons is akin to saying 1 =2 or ODDS equal EVENS...

now reveals Zero Point Energy to be nothing more than a charged boson of
mass-energy momenta [a 1/2 wavelength photon],

in turn
unifying quantum mechanics and special relativity; and that bosons [quantum
inductive loops of mass-energy momenta] are the true building blocks of
everything in our Universe - energy integers if you like.

without geometric grammar produces scientific gibberish.....

That's kind of the answer I was expecting.

What has that got to do with focused polyfrequent EMR?

I need something looking like a fluorescent tube, but collecting light from all angles in stead of emitting it. The collected light must then somehow organize itself to exit in parallel through the ends of the tube. The amount of light over a wide spectrum will be about 1000W if it is a 2m tube. Can we do it using your method?

This design, if it works at all, wouldn't in any way address the taking of light in. If you could get it into something like a light pipe / fibre optic it would, if it does anything, then collimate it.

Thank you for the prompt reply!

Let us presume, for the moment, that your accidental invention works and I can actually pipe the light into a 20mm diameter light pipe, but deflecting at all angles, how much space do you recon the collimator will need?

About the volume and mass of a 1/2 litre bottle of water.

esar012 years ago

Does your design involve a convex mirror with it's virtual focal point at the same focal point as the parabolic, reflecting the beam back towards the parabolic requiring a hole for the beam to emit through?

SolarFlower_org (author)  esar012 years ago
No, it's a little more complicated than that, tho does just work through multiple reflective surfaces.

Have a look at my website creation at hotawesomestuff.com

I hope you like it

Well...SOMEWHAT of topic,it sounds unlogical,but if i concentrated a powerful light source foucusing on the center of a multipule sided crystal or rectangular prismic crystal,is there a chance i could get a disco-*screws up* a chance of beams of light emerging from all sides?

tc5703 years ago

umm... , I don't think so. collimating the light from a 100 watt bulb would give you collimated light from a 100 waTt bulb, not a laser. lasers are collimated, yes but not all collimated light IS A lasers (all dogs are animals but not all animals are dogs) Any light can be collimated any form of radiated energy can be in fact and perhaps you have stumbled onto a better method of doing so. I hope you have, good luck with it.

SolarFlower_org (author) 4 years ago
So after several years now of sitting on this design with the half hearted background intention of maybe commercialising it one day it's now become fairly obvious that I'm never getting round to doing that.
So I'll be opening up the design, publishing it in full on my website along with a bunch of other stuff in the coming months.
I'll post a link here when that happens.
Wisaam4 years ago
How did you made it, what shape is it?
I could see a parabolic dish feeding a bundle of fiber optic cables being used to light up an underground installation like a subway station or mine. If it weren’t for the lack of cooperation of weather this could be a great energy saver.
I am told that they say that a man made crystal power storage is impossible; not so. Feed enough power or light into a device from which it cannot escape on it's own, and find a way to meter the output slowly and you have a replacement for a metals or chemical battery.
I have for a long time thought about combining piezoelectric crystal generation and semiconductors to create power source.
rhodge-14 years ago
Edison made hundreds of devices; we still use the results of the light bulb and the phonograph as well as a number of other things like electric equipment of all kids; Go For It.
I am told that they say that a man made crystal power storage is impossible; not so. Feed enough power or light into a device from which it cannot escape on it's own, and find a way to meter the output slowly and you have a replacement for a metals or chemical battery. Do the release in a 'at once' manner and it goes BOOM.
lemonie8 years ago
You cannot turn a 100W light bulb in to a 100W LASER, for more than one reason (ask if you need me to explain) L
Indeed, one quite simply cannot, through optical manipulation, turn any non-coherent light source into a coherent beam of light (i.e. laser beam), any more than you can't turn apples into oranges by confusing the two. But I have two questions to ask here just for my own information, as follows. Question 1) Is it possible (disregarding the issue of diameter for now) to optically manipulate a large, non-coherent light source -like, say several 200W blacklight tubes like those used in tanning salons- to produce a fairly collimated flow of light? Question 2) Is it possible to reduce the diameter such a collimated flow of light to one that is narrower than that of the of the original light source/lamp?
Yes, no, this will collimate, not coherent-ise.

Question 1)
As long as you can get it all into a light pipe or similar then, according to my simulations, this device will collimate it, yes.
However trying to do anything with light sources that large might be a bit tricky...
Probably better to go with bulbs, as small and powerful as possible.

Question 2)
Sure, just whack a couple lenses on the beam to focus and recollimate it.
Do you have any sketches? Even a simple sketch? I found one system from RPC Photonics that could work, but I'd have to modify it and try to scale it down. Can your system be scaled down? If neither system can be made small enough, I'd use several lenses and just deal with the light loss.
SolarFlower_org (author)  Deep-Thinker4 years ago
I have a 3d model, but am still not sure what to do with the thing, and whether I want to open it up...

You have a link for the RPC system?

My device, if it works at all, would be arbitrarily scalable.
No, I don't have that link. Now that I think about it though, since you haven't patented it, it probably wouldn't be a good idea to show anyone. Here's what I would suggest (maybe you already know all this...but I'll go ahead anyway, if you don't mind): get the materials you need (light source, mirrors, lenses, or whatever your invention takes--making sure though that any lenses, molds, or curved/parabolic mirrors have the correct focal lengths and are to scale), build a prototype, and test it. There's no good in having an invention if it always stays in a workshop, after all. :) If it works, great! Patent it and then sell your idea. If it doesn't work, just make adjustments and try again.  If you need to get any kind of professional help or anything like that, don't tell anyone anything until they sign a nondisclosure agreement.  Just be sure to measure everything to exacting accuracy and precision, because the slightest error could throw the collimation all off balance. Snell's law and various other equations should work for the equations, as long as you remember that different wavelengths of light have different indices of refraction. The other thing to remember is that the light from a LED or incandescent lightbulb radiates outward in an arc, while the light rays from the Sun are nearly parallel when they hit any surface on the Earth because the Sun is so far away. Therefore, while the same device might possibly collimate different types of light sources, the focal lengths of the optics you use would have to vary. Parallel rays and narrow spread beams can be efficiently collimated with just lenses, but for beams that spread widely (such as SMD type LEDs and light bulbs) lenses cannot capture all of the light unless they are very large, and parabolic mirrors only collimate the outside portion of a beam, allowing the center portion to spill out, uncollimated. Unfortunately, that's the difficulty with collimating highly divergent light. RPC Photonics uses a special shape to collimate LED light...but I don't see how it works better than a plain parabolic mirror.  Another company claims it uses a combination of a parabolic mirror and a lens/lenses to collimate 90% of the light.  Perhaps that's true but they don't show us how they do it.  I have even tinkered with the idea, and I think my model would be able to collimate 98-99% of the light from an SMD LED (the only purpose it's meant for), but like you I still have to gather the materials, test it, and hopefully eventually patent it.  But like I said, it's only meant to collimate SMD LED light, so it wouldn't be versatile beyond that specific purpose.  I'm definitely looking forward to hearing whether or not your invention is a success, and I wish the best of success to you!
There was a time when all the scientific data of the day said you cannot travel faster than 35 miles an hour or you would suffocate but they built the steam engine anyway and the train was borne.

One of the origins of the saying.

If an old grey haired scientist tells you can do it, you can do it.
If an old grey haired scientist tells you can’t do it, he is wrong.

I say he should go for it too.
SolarFlower_org (author)  Deep-Thinker4 years ago
I have a 3d model, but am still not sure what to do with the thing, and whether I want to open it up...

You have a link for the RPC system?

My device, if it works at all, would be arbitrarily scalable.
SolarFlower_org (author)  Deep-Thinker4 years ago
I have a 3d model, but am still not sure what to do with the thing, and whether I want to open it up...

You have a link for the RPC system?

My device, if it works at all, would be arbitrarily scalable.
Yes it is called a parabolic dish it will collect the light and focus the light to a single point, however at that point it will begin to diffuse. The trick is finding a way to keep it focused beyond that point.
Exactly, it is one of the traits that keeps us from having Light Sabers LOL the diffused like "goes on". and the "force" would not only not act solid (see Star Wars movies) but would only exist with any usefulness at one single point in space.
To quote Albert Einstein
If an old grey haired scientist tells you can do it, you can do it.
If an old grey haired scientist tells you can’t do it, he is wrong.

Build a prototype test it and get it to work then patent it.
Lazars are lousy weapons so don’t worry about that.
SolarFlower_org (author)  Kiteman4 years ago
Josehf was I think simply encouraging me to go ahead with the device as is, which is a light collimator, that the discouragement of a thing not sounding like it's likely to work shouldn't put one off building a prototype and finding out, just in case.

I did mean to imply that this device will make light coherent, just that it would form a laser-like beam of varying wavelengths of light.
Remember to have fun and take notes you never know what you may find and learn.
As mentioned a light cannot become a laser but there are huge applications. One that I have noted is to pipe the sun into optical fibres (simly filter the infrared and ultraviolet out) and then pipe light into buildings, giving the occupants natural light more than one storey down. The cost and efficiency is important and has to be better than a solar panel plus controller plus LED lights.
wtg30005 years ago
i'm looking to make a collimated non-coherent "spotlight" that can send the light over 3 or 4 metres. unlike a PAR bulb (parabolic aluminized reflector), which still has a fair bit of divergence. a PAR bulb (http://oee.nrcan.gc.ca/publications/equipment/lighting/6571) will probably be the source.

any help appreciated.

SolarFlower_org (author)  wtg30005 years ago
This is still on my list of things to prototype, I'll take a basic proof of concept crack at it, then re-evaluate what I want to do depending on the results.
It's possibly something I'd want to patent, but am thinking if effective it would make quite a good ultra cheap opensource 'laser' cutter, which could be of use to people.
Um, sure. Guess so.

sugarandfat gmail
email alert went to spam folder -- strange. i have a very, very cool thing i'm developing (way better than a cutter) and a part of it collimating a bulb into something like a laser (small but doesn't need to be a dot). we should talk -- there could be some synergy here if you can do a quick-and-dirty POC. i'm available this weekend. probably need an nda on both sides.
PKM8 years ago
As far as I know you can't concentrate light from a large diffuse source (note- not the same as a wide beam from a point source) back to a single point. I have wondered whether something like the Winston cone was possible, after seeing the Gomboc I occasionally think anything is possible with enough maths :) I envisaged a funnel-shaped reflector with a spiral running down the sides that would effectively "ratchet" incident light from a wide cone to a small aperture, for light pipes etc. Feel like posting any sort of diagrams of your device? No-one's going to steal your IP :)
SolarFlower_org (author)  PKM8 years ago
> No-one's going to steal your IP :) Ha. If I get around to patenting it, yes. My design won't bring light back to a perfect point, but will get it pretty close. Funnel? Spiral? LIke a shaped Fresnel?
Yeah, pretty much. Imagine a fresnel lens vacuum formed to the inside of a funnel (a horn-shaped one rather than a Winston cone shape), but with a slight twist. I'm not sure why the twist, I figured the principle probably wouldn't work in two dimensions but might in three...
SolarFlower_org (author)  PKM8 years ago
Do you have any diagrams? (I may steal your IP...)
Lftndbt8 years ago
Did I hear you say, it is still in the design phase? Better get a crack on as my effort is built and in the semi final stages callibration. I can throw a 1w LeD's light output, onto a surface over 2 km's away, as a concentrated dot. Sun rays can be amped dramatically also.
SolarFlower_org (author)  Lftndbt8 years ago
Pff, LEDs are cheating. They're practically a point light. How concentrated the dot and how much amped the sunlight? I'd love to see your design, though if it's not patented yet I'll understand...
LeD was just an example, incandescent or any other light source can be amped. Hmmm, sunlight. You know how you can burn something with a magnifying glass? Well this little item combusts any flammable material instantly. If directed at a solar panel reading 1.5v output in normal light, it can produce an output of 9.5v. That's just a basic example. I do not use solar panels in the device though. I am still developing the devices, solar tracking ability ATM.
SolarFlower_org (author)  Lftndbt8 years ago
Kewl. I'm working on solar tracking at the moment as well, though mine needs to be built out of scrap...
kelseymh8 years ago
As I noted below, you can find some existing prior art very easily with a Google search for "light concentrator collimator." Among those hits, you'll also see some good articles explaining the thermodynamics of collimation and why some things are possible while others are impossible.
SolarFlower_org (author)  kelseymh8 years ago
I found plenty of things for collimating LEDs or other very small light sources, or unidirectional light, but that's easy. I haven't managed to find anything which will collimate light out of a lightpipe; ie shooting around all over the place, or from an extended or lambertian source. Someone's going to have to explain the implications with the second law of thermodynamics for me. Keep in mind that this is still a long way off 'perfectly collimated', if that makes any difference.
kelseymh8 years ago
Be a little careful with your terminology -- "laser" refers to a coherent beam (i.e., a single continuous plane wave at a fixed frequency). What you describe is high-efficiency collimation of a regular incoherent light source.

I'm not sure we (myself included) said this was impossible, just difficult :-) You will get some beam spread (you quote 1-2 meters/km, or 1-2 milliradians), so the intensity will still fall off proportional to 1/r2, but with a small proportionality constant.

I'm curious as to whether what you've really done is to reinvent the Winston cone. Take a look at that article (and look up the many users of Winston cones in Google Scholar) and see if it relates to your work.

By the way, the reason I didn't mention Winston cones when we were discussing your project, is that I didn't think of them as "collimators," but just as light collectors you hook up to a phototube :-/
SolarFlower_org (author)  kelseymh8 years ago
Yes, must stop using the L word. -Non-coherent collimated beam- I don't want to give too much away, but it's a bit more comprehensive than a Winston cone, though based on vaaaaguely similar principles. As I understand it a Winston cone won't really collimate at all, even if it's really long.
Ah, good point. What we (I'm a particle physicist) use the Winston cone for is to collect light and concentrate it down to the surface of a PMT or photodiode array, which is mounted at the narrow end of the cone. I sort of had a picture where you put the light source at that narrow end, and ran the ray-tracing backwards to send the light back out. But I guess you're right that it wouldn't actually be collimated. I just did a Google search on "light concentrator collimator" (I presume you did so as well), and there are a number of relevant matches, including both research articles and existing patents. I suspect that you'll need to do a fairly extensive prior-art search in order to prove originality and non-obviousness.
lemonie8 years ago
Sorry, I missed it the first time - why did you post this ("ignore")? L
SolarFlower_org (author)  lemonie8 years ago
Sorry, for some reason I've been having difficulty posting here. Trying to sort it out.