Prototype dual lens optics by elektrolumens??

[wechnivag]

I remember reading a thread started by Wayne about this double lens prototype that he built. Think this is some time back. Can't seem to find the thread now.:hairpull:

IIRC, the dual lens allow almost all the light to be gathered from the CREE, as compared to ~100 degrees with the best aspheric lens system.

The first smaller lens (meniscus) is placed very close to the LED to capture the max amount of light from it, and project it onto the second larger lens to be focussed into the tight beam aspherics are known for.

Does anyone else remember reading this? :thinking: Guess the prototype never went to production? This patent has a similar description of the concept. There is a sketch if you scroll downwards.
http://www.freepatentsonline.com/7175299.html

The reason i'm looking for Wayne's original post is to try to create a pure thrower that can capture almost 100% of the lumens from the CREE die.:devil::devil::devil:

 

[LukeA]

You have to find either a large collimating lens with a very short FL or a looooong flashlight head.

 

[wechnivag]

But it is my understanding that the beauty of the dual lens optics is that almost 100% light collection and collimation can be achieved using reasonable lens diameters and focal lengths!

http://www.candlepowerforums.com/vb/showthread.php?t=159372
This thread is living proof that this concept is very much plausible. The first lens collect as much light from the CREE, and passes it to the second lens, which then collimates it into laser beam. :devil::devil: I'm inclined to believe that with proper optimisation, the size and length of the dual lens system can be reduced to a very practical form.

http://www.flickr.com/photos/24250360@N04/2301159921/

Hope this can start the ball rolling on the required diameter and focal length for the two lens. Its actually rather like a condenser lens system, only the output has to be collimated rather than focused into a near spot. Hope optics experts on this forum have suggestions?

Actually am i making any sense? This is how i understand it to be...:thinking::thinking::thinking:

:thanks:

 

[wechnivag]

You have to find either a large collimating lens with a very short FL or a looooong flashlight head.

After almost a week of trawling through CPF using all possible search keywords, i finally found the thread i was looking for..

http://www.candlepowerforums.com/vb/showthread.php?t=12031

Wayne mentions using two lenses to provide an adjustable beam that goes from flood to spot. Granted it does not project an image of the die, more like a Megaray kind of beam.

Any idea how the two lenses are selected?

 

[ElectronGuru]

Any idea how the two lenses are selected?

I suggest heading over to his place to ask:

http://www.cpfmarketplace.com/mp/showthread.php?t=167887

 

[2xTrinity]

But it is my understanding that the beauty of the dual lens optics is that almost 100% light collection and collimation can be achieved using reasonable lens diameters and focal lengths!

The problem with that design is by placing a small lens right up next to the Cree, the die-image size becomes much larger. That means although more light is captured by the aspheric stage, it cannot throw as well as if there were only one lens. The advantage of a two-lens system is variable focus.

A better solution for a flashlight meant for throw is a TIR (total internal reflection) optic, which is essentially BOTH a reflector and a lens in one piece. Variable focus isn't an option with a TIR, but IMO that's more than made up for by the fact that the designer has almost complete control over the beam pattern, including hotspot beam angle, spill intensity, transition from hotspot to spill etc. Theoretically, TIR optics can collect over 95% of the bulb lumens and actually use them for throw, but most of the small cheap ones aren't designed nearly as well as they could be, and are only around ~80-85% efficient.

I would love to see some custom large TIR optics. A 52mm (maglite sized) TIR optic could achieve beam-angles similar to the best aspherics at that diameter, AND have nearly 100% light collection. The best would be if instead of using an aspheric for the final stage (eg a lens that is massively thick) they could accomplish the same effect by incorporating a fresnel lens instead.

 

[wechnivag]

The problem with that design is by placing a small lens right up next to the Cree, the die-image size becomes much larger. That means although more light is captured by the aspheric stage, it cannot throw as well as if there were only one lens. The advantage of a two-lens system is variable focus.

So i have another piece of my puzzle, the small lens close to the CREE to collect the light will make a larger image of the LED die, affecting the throw from the second lens. Have you tried this out? Would love to see some beamshots!

Is is true that a large ratio between outer and inner lens diameter will give better throw? As achieved here? Wayne's prototype mentioned the need for a large outer lens too. Megaray's outer lens does not seem that large. http://www.candlepowerforums.com/vb/showthread.php?t=159372

On the same note, i realised that the dome lens that comes on the CREE is itself like a positive meniscus, and i believe this is what is responsible for the narrower output pattern of the CREE, compared to lambertian outputs of the Luxeons and SSC emitters. There is some discussion and beamshots comparing a CREE with and without the dome.
http://www.candlepowerforums.com/vb/showthread.php?t=143195

I would love to see some custom large TIR optics. A 52mm (maglite sized) TIR optic could achieve beam-angles similar to the best aspherics at that diameter, AND have nearly 100% light collection. The best would be if instead of using an aspheric for the final stage (eg a lens that is massively thick) they could accomplish the same effect by incorporating a fresnel lens instead.

The tightest beams from TIR optics i have seen project about a 4 deg beam, though i have also seen one that is supposed to project the square LED die image as well. Like you said, shame that the focus is not variable, you need welding lens for close up work with a CREE 4 deg beam :laughing:.

Is there a reason why the TIR optics in the market are all 20-25mm? Is there no need for a larger diameter for better throw, unlike reflectors and lenses?

Also, one comercial dual/multi lens light i came across is the Megaray, the throw in spot mode is very nice, giving a parallel beam of light almost like the Maxabeam. Any idea how the lenses work? Actually die projection is not really necessary for ultra throw, just a parallel beam of light is good enough.
http://home.earthlink.net/~kenshiro2/

Hope to keep the discussion going and revive the interest in optics for flashlights, the aspheric frenzy is dying down.

 

[saabluster]

So i have another piece of my puzzle, the small lens close to the CREE to collect the light will make a larger image of the LED die, affecting the throw from the second lens. Have you tried this out? Would love to see some beamshots!

I have tried this out and can confirm that it does make the projected image of the die larger and reduces throw. I have been working on getting this setup perfected and as soon as I do I will post pictures.

 

[PEU]

From what I see in the patent aplication its classic zoom lens applied to the head of a flashlight.

Does it makes a good flashlight? dunno, does it makes a rugged flashlight? I'm doubtfull. Time will tell if we are going to see it in a production unit


Pablo

 

[wechnivag]

From what I see in the patent aplication its classic zoom lens applied to the head of a flashlight.

Does it makes a good flashlight? dunno, does it makes a rugged flashlight? I'm doubtfull. Time will tell if we are going to see it in a production unit


Pablo

Good point, but i'm inclined to think that the limiting factor now is the size and durability of the lens system, rather than practicality. The smooth flood light, and intense throw of a multi-optic light is very very practical.

There are a lot of patents out there, check this one by Eveready. Its not a very practical package, the lens assy is longer than the 2D batt compartment...
http://www.wipo.int/pctdb/en/wo.jsp?WO=2006/020574&IA=WO2006/020574&DISPLAY=STATUS

My 2 lenses should hopefully arrive today, i'll try to get some pics tonight!!!:twothumbs