Combined lens/reflector optic studies

I came up with a simple method of suspending an optical element in front of the LED within a reflector:





MORE INFO

I posted some lux measurements on the MORE INFO web page linked above. I will bring in the beam shots:





From left to right: first a stock 3" SF "t" without any additional optics. Next is the combo with 1/2 ball lens all the way out, contacting the flat lens. the 1/2 ball is then progressively moved towards the LED until in the last image on the right, you have little effect from the reflector and a wide flood is projected due to the position of the 1/2 ball.

Interestingly, with the 1/2 ball situated where it just keeps you from seeing the LED, regardless of off angle view, you get a good lux reading as well as a beam like the second in on the right where the normal direct flood or spill would usually be much wider in angle with much less intensity. This normal flood can barely be detected in the "stock" image on the far left.

Unfortunately, I kind of screwed up the lexan lens when I removed it. In retrospect, instead of using .040" tungsten rod for the "uni-strut" I should have used some .062 nitinol that I have sitting around. Finally a perfect application for it and I don't consider it. /ubbthreads/images/graemlins/grin.gif

With this combination strategy, a new door is open for photon management. Shallower and wider reflectors could be used with optic elements to keep most of the light closer to the center. Given a certain sized reflector, I assume a mathematical solution might exist where a particular lens size and position could be determined for a near overlap of reflected beam and refracted beam portions. The rear end of NX05's and the 30mm could be removed and allow for partial management of the photons by the lens and partial by the reflector.

As anyone knows using the plastic optics, TIR is really MIR (mostly internal reflection). What I like about this combination shown above is that there isn't any TIR as part of the scheme.

If the flat lens weren't damaged and if the 1/2 ball had AR coating on the flat side facing the LED, it would be interesting to see the effect this would have on the lux readings.

Re: Combined lens/reflector: 5W turbo study

very nice!

i did somewhat of the same thing a long time ago with my modded brinkmann 3aa. i used an acrylic ball in conjunction with a mag and brinkman reflector. at that point (the ball right on top of the emitter). with such a small reflector, and the way the light was hitting, i got a nice center hotspot, with an anoying ring surrounding.

i've also experimented with nxo5s reversed on top of a mag lens in conjuntion with a reflector. this didn't work, but i did find that an nx05 directly on an emitter upside down yields a very nice beam...i would assume this is somewhat like the nx01.
i've also used upside down reflectors ontop of reflectors.

what i find extremely interesting about your method, is that it's adjustable...very nice! /ubbthreads/images/graemlins/thumbsup.gif

maybe if we used a glass 30mm optic upside down on top of a similar size reflector... /ubbthreads/images/graemlins/grin.gif /ubbthreads/images/graemlins/twak.gif

Re: Combined lens/reflector: 5W turbo study

Don,
Nitinol???- thermally adaptive/active optics?? /ubbthreads/images/graemlins/ooo.gif /ubbthreads/images/graemlins/ooo.gif /ubbthreads/images/graemlins/grin.gif
(I'm toast- off to bed- will reread tomorrow for some knowledge-gained!)

Larry

Re: Combined lens/reflector: 5W turbo study

i found the post i made when experimenting with very small reflectors and acrylic balls...

the images are temporarily down, but that should be better soon...

here it is

Re: Combined lens/reflector: 5W turbo study

In theory, the exact position and characteristics of the optical element depend on the emitter and the reflector.

The black line is the parabolic reflector. The red line is the edge of the off axis emission that contributes to spill, not collimation. The object is to collimate what would normally be spilled.

The violet line is the point below which the emission pattern does not utilize the reflector at all, normally resulting in "the black hole" when a reflectored emitter is brought close to a surface. The green lines perpendicular to the violet line represent the edges of the zone where there is no reflected light.

Anything extending beyond the red or green lines will interfere with light either before or after it is reflected. Since the reflector is already doing its job, this should be avoided. Anything not extending to the red line will not collimate what would otherwise be spill light.

There is only one place, the area between where the red and green lines intersect that a lens can be put to: 1) capture and collimate all spill light (yellow line) and 2) not interfere with any reflected light.

A condenser lens situated here, as long as its focal length was exactly right for the emitter position, might collimate another roughly third of the light that is currently lost in spill.

McGizmo, Very interesting idea! I think you've just dragged us into a new phase in optical design!

It just occurred to me that another possible way to "float" the optic would be:

where the gray area is a combination lens / optical element that refracts and collimates the "spill" light.

Re: Combined lens/reflector: 5W turbo study

Otokoyama,

Great drawings and considerations. I played with the light last night and it was kind of cool to be able to adjust the beam with a push and pull of the rod. Getting a bright flood beam out of a turbo head was a real first! With the lens in the "optimized" position, the refracted beam was bright enough for good medium throw and its fill around the reflected spot beam aided in illumination of targets larger than the spot beam's coverage. The turbo with this configuration is less of a search light and target acquisition is aided considerably.

It seems to me that in addition to the ease of assembly, the "uni-strut" mount has the least impact in terms of light loss.

In your drawing, I would think if you had a thin walled, transparent cylinder with low refractive index, you could place in in the green lines and it could host the lens element.
Your defined area of "non interference" dictates position as well as lens diameter but it still allows for various radii of both plano-convex, convex-convex, concave-convex, etc. lenses.

A shallow section of a larger diameter parabolic reflector might be viable now with the lens catching the significant amount of light that otherwise would have been spill. I think we might be able to get tighter beams in smaller overall heads with this combination.

Essentially what we have here, I think, is a case where we are using a reflector in lieu of the tube section and TIR portion of say the NX05 or 30 mm but still using the lens for refractive control of the light closer to axis.

Larry,

The operating temp within the flashlight would insure that the nitinol was always in its super elastic state which would make it an ideal strut material; stiff yet bendy with perfect memory. /ubbthreads/images/graemlins/grin.gif The tungsten is so brittle that a good shock load could easily result in Oops, me thinks.

- Don

Re: Combined lens/reflector: 5W turbo study

It is threads like this that keep me coming back to CPF! It is very gratifying to see the synergistic effects arising from the free and lightning fast exchange of ideas and experimental results. Keep it up guys. I love you.

Re: Combined lens/reflector: 5W turbo study

Ditto. This is great stuff.

Re: Combined lens/reflector: 5W turbo study

Don,

I see that you left a very clear fingerprint on the second picture. I'm afraid you've now been entered into the government's federal list of flashlight hackers /ubbthreads/images/graemlins/tongue.gif

Wilkey

Re: Combined lens/reflector: 5W turbo study

Wilkey,

Yeah, I didn't notice it until the film was developed and by then it was too late... Good thing I changed the ridges in PhotoShop. /ubbthreads/images/graemlins/grin.gif And while on the subject, I saw some fingerprints on your light too! /ubbthreads/images/graemlins/tongue.gif

- Don

Back on topic: I have a M3 reflector that I am trying this optic combo on now. It's a nice but relatively shallow reflector so it will be a good candidate for some refractive asist.......

Re: Combined lens/reflector: 5W turbo study

In some tests of this technique using a small glass condenser lens and a small fresnel lens held in place with a pipe cleaner, the beam was definitely strengthened. The focal point had to be chosen with care, or you get a perfect image of the die, bars and all. A problem with common lenses is that they have a slight prism effect, so the resulting image isn't nicely white.

Re: Combined lens/reflector: 5W turbo study

Well I buttoned up a 5W HD and a M3 reflector in one of Leo's proto McLux T heads and I should be able to try some different combinations since the mineral glass lenses aren't expensive and I can easily swap out the front end. I quickly put a funky 1/2 ball and flat lens together this eve and it's a step in the right direction but I think I need to make a slightly larger diameter 1/2 ball and probably drop it in just a bit. In the picture below, you can see that there is still room for wasted side spill:



The refracted beam is about 1.5 the size of the reflected beam and the corners of the square beam are rounded. I did a lux reading both with the 1/2 ball and without; 1460 and 1100 respectively. I think it might be nice to try a convex-convex lens that doesn't necessarily tighten up the refracted beam as much. I still like a flood portion to my beams but not as wide as the default spill.

I agree that you don't want the refractive lens in a position where it focuses the projected die image; especially the 5W grill!

- Don

Re: Combined lens/reflector: 5W turbo study

wow don! that looks very efficient!

Re: Combined lens/reflector: 5W turbo study

Don,

This is sort of related to your work here but it's perhaps more directly applicable to incandescent. Have you ever considered mounting a bulb assembly inverted at the focal point of the reflector? That is, similar in design to the large spotlights, where the bulb faces in, thus preventing any uncontrollable forward radiation. This way, all the bulb output could be collimated with little to no spill. I imagine this might only be useful for incandescents where heat is dissipated directly by the bulb. Do you think a "normal" bulb-forward reflector could be used?

Wilkey

Re: Combined lens/reflector: 5W turbo study

hmmmm, I wonder if this could allow a smaller head to yeild a decent beam with 5 watters. but make it non adjustable, make it fixed. It would be neat if this allowed Arc or slightly smaller head to make a beam the shape of what comes out of a 6Z?

Re: Combined lens/reflector: 5W turbo study

[ QUOTE ]
JonSidneyB said:
hmmmm, I wonder if this could allow a smaller head to yeild a decent beam with 5 watters. but make it non adjustable, make it fixed. It would be neat if this allowed Arc or slightly smaller head to make a beam the shape of what comes out of a 6Z?


[/ QUOTE ]
Jon,
In an earlier post in this thread, Don speculates:

Essentially what we have here, I think, is a case where we are using a reflector in lieu of the tube section and TIR portion of say the NX05 or 30 mm but still using the lens for refractive control of the light closer to axis.

I believe that he is correct 100% on this. If this is correct, then the reflector+lens combo would offer no advantage over the TIR optic if non-adjustable. I suspect [speculation here] that at its best, TIR is a better means of reflection than metal or metalized surfaces. What I find exciting about the methods under discussion by Don and Otokoyama is the potential for adjustibility beyond what can be achieved by moving an entire TIR optic back and forth.

Re: Combined lens/reflector: 5W turbo study

Wilkey,

I think thermal management requirements precludes mounting a LED inverted unless the head were really big in diameter and depth to allow for a massive sink slug suspended by fins within the reflector. I do think a properly shaped reflector could be suspended in front of the LED that directed the light back at the parabolic surface.

Jon,

I don't think there is any magic bullet here. My sense is that tightness of beam will still be a function of diameter; larger = tighter. However, using the optic in front allows for a shallower section of a larger diameter reflector. My gut tells me that this combination may allow for a higher percentage of the luminous output to make it out the front end when compared to an optic that is relying on TIR. Certainly this should be the case when folks are lifting the 30 mm off the LED a bit to loose the grill.

One thing I do think is kind of cool is that it is certainly possible to treat a flashlight like a 35 mm camera where you could have the option of picking a lens or optic best suited for the task at hand. In the McLux T head, for instance, The 30 mm it was designed for can still be exchanged for the reflector and lens combo that are currently in place. I could also have a few different styles of front end optics to place in front of the M6 reflector. I can even fashion a drop in flood reflector for this head.

A cool design would be one where all of the various photon management schemes were integrated and sealed into the bezel cap. The front end of choice could be selected and screwed onto the head with a O-ring seal that the skirt of the bezel cap mated with.

Many lights have the option of beam shapers and filters. There is no reason a more sophisticated "plug and play" system couldn't be designed. No reason other than market viability that is /ubbthreads/images/graemlins/grin.gif .

- Don

EDIT: Actually an even better format for "Plug and Play" would be to go with a body and LA program like Mag or the Pelican M6. In this case, you would not be restricted by a head diameter and could just screw on a head with integrated optic, sized as required. SF has the turbo kit but replacement of lamp is also required. In the case of a LED, since most light is at 90 degrees and forward as Otokoyama has pointed out, a single LED LA could be used for a wide range of front ends. This way one could get max mileage from the single S rank 1W or X rank 5W that they won on EBay for a king's ransom. /ubbthreads/images/graemlins/grin.gif

Re: Combined lens/reflector: 5W turbo study

I couldn't resist trying this out using a 1" acrylic half-ball (a cheap condenser lens) inside a 1" acrylic tube attached to a lens. Following is a picture of the reflected beam with the refracted beam purposefully off center for comparison. Even with the prototype quality construction and non-optical plastic, the refracted beam adds perhaps 10%, though makes a nice picture of the die, barbeque grill and all. However, when this is merged with the reflected beam, the beam in its entirety simply looks brighter.

Here are pictures "inside" the beam. On the left is the beam without the refraction unit. You see only the LED and a bit of reflection. (The reflector is focused to infinity, and this is from 6 inches.) On the right is the beam with the refraction unit. The "blob" is the acrylic glue holding the tube to the lens blotting out part of the reflected beam, but the refracted beam is obviously strong.

To do this properly requires optics well beyond my humble means to design and produce, but the concept appears promising.

Re: Combined lens/reflector studies

I made another flat lens with 1/2 ball lens but this time I turned the refracting ball lens myself and left a stub for stand off and mounting to the flat lens:



The placement of the lens slightly encroaches on the reflector. The resulting beam is quite clean and the would be spill or non re directed light is now in a rounded corner flood beam that is considerably tighter than otherwise would have been the case. Without the refractive element, in the beam shot below, the flood beam would have extended out past the border of the image and its intensity would have been considerably less.



In the very center, you are actually looking straight down at the die itself so this the portion of light coming out the center is not being reflected or refracted. There is some useable light outside of the refracted portion which I suspect originates from here as well as light that is reflecting off the various lens surfaces rather than going through. There are no AR coatings on any of these elements.

I am real anxious to check this beam out at night in real world use.

- Don

Re: Combined lens/reflector studies

That is one fine looking beam, Don.

Would it be possible, holding the camera settings locked, to take a pic of the beam both with and without the refracting element (I.E. One with refraction added, one with plano lens) so that we could see the exact same source and reflector with and without?

I think that would help at least me to judge the level of improvement better ... /ubbthreads/images/graemlins/grin.gif

Good luck on this - I think this is a good avenue to go down.