Optics engineer needed! NEED the beam!

[cowboyfoomp]

Hi guys, I know this is my first post, however I have been a CPF lurker for quite a while now, couple years if I remember correctly. However, I begun my first homebuild and I need an optics guy for some lensing help!

I am using a ebay "10W" LED for my source. From the looks of things, it's a 9 chip on a single die with 6 chips that have gone inop. Regardless it's REALLY bright. I love it. However I need to lens it down to something useable.

Unlensed with no reflector, the spec's say it's 140 degrees, and from the looks of it, I'm not going to argue. The emitter section is 14mm from the specs, however given only 3 of the 9 chips illuminate it's kinda off center. But for my purposes, it's close.

What I'm attempting to do, is create a beam of light about 3-5" in diameter that stays collumnated to about 100 feet. Saying it may start at 3" at the unit, then reaching 5" at 100 feet. Make sense? I don't want any spill over and would like to refector it for as much brightness as possible.

Now do it for $5.00 LOL! I'm kidding, but I'm scrounging parts from slide projectors, binoculars and the like for the optics. I am willing to spend the money if I know it works though.

ANY help will be GREATLY appreciated!!

Thanks a bunch guys. CPF rocks!

FOOMP

 

[effulgentOne]

5" at 100ft is going to be impossible, especially with the multi-die LED you have. It sounds like you want a laser.

 

[TorchBoy]

5" at 100ft is going to be impossible, especially with the multi-die LED you have.

+1.

It sounds like you want a laser.

+1.

:welcome:

 

[cowboyfoomp]

Okay, so I may have been exaggerating a bit about the size and distance a bit. I'm SURE it's possible, however I don't have access to the math, the engineers, and the cutting shop to make the lenses.

What I'm looking for is a fat, columnated beam with very little divergence over distance. No spill over or artifacts, or as close as I can get within a reasonable cost range.

More like a long throw spotlight seen at concerts but in a small form factor. LED sized.

I know my source could be better, and if someone has a better idea than my half broken multichip led I'm all for it! It's gotta be crazy bright, and LED. (I thought about HID but it's not dimmable)

I understand the challenges I'm facing, the lack of high powered LED from a single chip, no single source for the desired products off-the-shelf, and my own optical ignorance. Which is why I'm asking for help.

Thanks guys!
Foomp

 

[SemiMan]

It is heady reading, but to start .... http://en.wikipedia.org/wiki/Etendue

Semiman

 

[Lumencraft (Matt)]

Hi cowboy,

Welcome to CPF.

To me it sounds like the closest thing you are going to get to what you want is an ashperical lens. DealExtream sell several different sizes. A dome shaped lens like this will concentrate the beam of light down to a square the shape of the emitter die at a great distance (not 5 inches at 100 feet ) but still very small. There is going to be some spill but not like what you normally see from a light. There is no way to be sure what effect this will have on the light you described, but with Cree single and quad die emitters the result is quite good.

This is just me speaking from my personal experience, having spent money on cheap lights trying to make them be something I liked, and never being quite satisfied with the result. You may want to look into getting a light that is a bit better quality, with all the parts working before you spend any money on "upgrading".

I am not in any way trying to discourage you (quite the opposite really) I just dont want you to have to spend money on something only to be disappointed.

Good luck with your build.

 

[Magic Matt]

I'm not an engineer, I'm just thinking outside the box a bit... but I can't see why it wouldn't be possible. It would need some seriously good engineering though. I think you'd have to pre-columate the beam from the multi-di LED, then use a secondary lens to shape it. I'm basing that on the fact we have a 4-bulb projector at the observatory that projects a mock-up of the sun or moon about 6ft across at around 75ft, and those are overhead projector bulbs. I've not studied it closely, but I know there's some sort of lens assembly between the bulbs and moon image, and another after the slide of the image and the front of the unit. It may be multiple lenses after the image, as I couldn't see into that part of the unit.

 

[HarryN]

As noted, it is possible. The DEFT flashlight is really close to what you want to accomplish. My suggestion is to consider buying one, take it apart if you must, and then make your own if that is still interesting.

The goal you state is perhaps the most complex thing to undertake in DIY flashlights, in spite of its strong alure.

 

[TorchBoy]

... a 4-bulb projector at the observatory that projects a mock-up of the sun about 6ft across at around 75ft, and those are overhead projector bulbs. ... there's some sort of lens assembly between the bulbs and moon image, ...

You've lost me.

 

[Gunner12]

I'm studying to be an engineer, but not optical. The multidie LED is harder to focus then a single die LED.

A big elliptical reflector could work. Keep one LED at the first focal point, and have the second focal point at 100 ft away, but the tightest focus will be at that focal point and that point only (won't be 3" at the unit).

The best I can think of is a single die LED (smaller the die the better) and a sufficiently large reflector, optic, lens. Maybe even a combination.

What are you using this for?

 

[Magic Matt]

You've lost me.

It was a typo I amended later. It projects the Sun OR the Moon, depending on which glass slide you put into it. Nearest thing I could think to describe it would be like a slide projector you'd have for 35mm slides... but for 6 inch square glass slides.... and for use in a large auditorium or theatre. It's pretty big, and it gets very hot.

 

[Al Combs]

What I'm attempting to do, is create a beam of light about 3-5" in diameter that stays collumnated to about 100 feet. Saying it may start at 3" at the unit, then reaching 5" at 100 feet. Make sense? I don't want any spill over and would like to refector it for as much brightness as possible.

FOOMP

There are some Sam Brown books available in pdf format that would help to visualize this problem. They are available from Anchor Optics, a division of Edmund Scientific that sells these books. Try the pdf "How to Build Opaque Projectors" from Anchor's Educational Documents list. On page 4 of the pdf (page 6 actual) is a table to figure focal length, spacing, magnification ratios and the like.

To have the beam start at 3 inches and diverge to only 5 inches at 100 feet is only a 2 inch divergence. I think that as the effulgentOne put it, that calls for a laser. I picked an example that has your 14 mm LED enlarged to 5 inches at 100 feet. That's a magnification ratio of 9.071 to 1. Basically you can do this using similar triangles. That assumes a simple infinitely thin lens. The total distance from the LED to the image is Mag + 1. So the LED to lens distance or 'A' is =100/(9.071+1) or 9.929 feet. The actual focal length of the lens= (A*M)/(M+1) or 8.943 feet. That's a 2725.9 mm focal length. :sick2:

The other problem is the 'f' ratio of the lens. The CVI Melles Griot aspheric lenses have been very popular in aspheric Mags. The 01 LAG 123 has a 52 mm diameter and a 37 mm focal length. That's an 'f' ratio of 0.71 to 1. For your 8.9 foot focal length lens to have that same 'f' ratio would be 12.57 feet in diameter, weigh several tons and probably cost more than 5 bucks. So if it's 1 foot in diameter it's only f=8.943 and has 158 times less light gathering power.

That is to say the intensity of the spot you gain by the image being so small, you loose with the low surface brightness from the slow speed of the lens. Not to mention needing a small crane to lift your 3 meter lens tube. It's a balance issue the Sam Brown book should be helpful with.

Good luck :welcome:

 

[Walterk]

Try this: send the light through the eyepiece of a rifle-scope !
It is a narrow beam and projects the cross. Not best for efficiency probably, but it can give a cool effect. And hey, you have 10 watts.

A telescope would make a narrow beam also. The diameter of the eyepiece is what limits you in choice of lightsource.

I think its easyiest to just try and see what works, make a test rig and work systematically. Its all about focus length.
I think most companys just experiment with reflectors and lenses till they have something cost-effective.
For us to find the optimum!

BTW; Really old skool slide projectors have really nice good aspherical lenses. Shielding the beam by enclosing it with a tube makes a sharp edge.