Reflector Decision for Aspherical Lens

sonikaccord · Dec 23, 2013

Hello everyone, I am building a torch from scratch using a Vero 29. I need help deciding which type of reflector yields the best possible performance:

Parabolic + Ashperical
Elliptical + Ashperical
Or only reflectors.

My goal is to have a beam with very little divergence from the light while capturing as much of the spill as possible.

Norm · Dec 23, 2013

:welcome:

An aspherical lens usually has no reflector or one that has been painted black

Norm

sonikaccord · Dec 23, 2013

Thanks!

I reviewed the datasheet again and most of my output is emitted within 120 degrees. Am I correct in assuming that the black reflector is for absorbing the spill? If that's the case, designing the optics would be tons easier.

Norm · Dec 23, 2013

You can try it with a reflector but from my experience you will see a lot of unwanted rings around the main beam.

Norm

DIWdiver · Dec 23, 2013

Yes, the black reflector is to absorb spill.

Any light that comes directly from the LED and gets into the lens, goes into the hotspot. Any light that bounces around inside the light and then gets into the lens, becomes spill. So to minimize spill, everything in the chamber between the LED and the lens should be not only black, but flat black. This will not increase the hotspot, but only decrease the spill.

Dark Laser · Dec 24, 2013

Why are systems with a condenser / pre-collimator lens not more popular? Spill light is catched and sent into the spot instead of wasting it. I understand that the setup is not the easiest, though.
By the way, has anyone lux numbers for a Maglite-size aspheric lens combined with a XP-G? With the deep reflectors and de-domed LED the lux numbers are quite impressive; so I wonder if an aspheric can beat that.

HarryN · Dec 24, 2013

Dark Laser said:
Why are systems with a condenser / pre-collimator lens not more popular? Spill light is catched and sent into the spot instead of wasting it. I understand that the setup is not the easiest, though.
By the way, has anyone lux numbers for a Maglite-size aspheric lens combined with a XP-G? With the deep reflectors and de-domed LED the lux numbers are quite impressive; so I wonder if an aspheric can beat that.

There have been a few lights built with 2 stage optics, but not really that many people are equipped to do it. The precision and custom alignment needed is beyond the tools that most people have.

I built a light with a 12 mm diameter 2 stage setup (reflector plus Fresnel lens) and it was a significant learning experience, and quite costly in test / prototype parts. Optical physics is not a strength of mine and it would have been much cheaper of course if I had the guidance of an expert in this area. For 50mm size designs, it is of course much easier, but still requires both optics knowledge and the tools, or enormous patience.

Lastly, 2 stage optics make the most sense when the apparent light source is very small and high intensity (which was my big downfall at the time). It is much easier to purchase larger area LEDs and build them up with somewhat larger single stage optics, but of course not quite as impressive.

The other path people tend to take is to just use an HID source instead of the LED for these kinds of applications, because those are easier to build up for "super throwers".

sonikaccord · Dec 24, 2013

HarryN said:
There have been a few lights built with 2 stage optics, but not really that many people are equipped to do it. The precision and custom alignment needed is beyond the tools that most people have.

I built a light with a 12 mm diameter 2 stage setup (reflector plus Fresnel lens) and it was a significant learning experience, and quite costly in test / prototype parts. Optical physics is not a strength of mine and it would have been much cheaper of course if I had the guidance of an expert in this area. For 50mm size designs, it is of course much easier, but still requires both optics knowledge and the tools, or enormous patience.

Lastly, 2 stage optics make the most sense when the apparent light source is very small and high intensity (which was my big downfall at the time). It is much easier to purchase larger area LEDs and build them up with somewhat larger single stage optics, but of course not quite as impressive.

The other path people tend to take is to just use an HID source instead of the LED for these kinds of applications, because those are easier to build up for "super throwers".

I have decent knowledge in optics. I understand the concepts and such. When you say "apparent light source," that means the size of the source before it hits the final lens? I've researched pre-collimators but it would have to be adjustable focus on both lenses as a prototype to get the best output.

Since we're on the topic, the LES (light emitting surface) has a diameter of 29.2mm with no optics. I'll see if I can pull up some equations and such to tighten the beam up in a theoretical sense.

alpg88 · Dec 24, 2013

waiven light colar.

sonikaccord · Dec 24, 2013

I checked out the Waiven light recycling tech. It sounds like a good idea esp since there is so much phosphor in the led that can be excited by that technique. I'll have to try it out once I get the led and build the heatsink.

DIWdiver · Dec 24, 2013

A source with a 29mm diameter is only good for flood lighting. Any optic, either lens or reflector, large enough to create any kind of tight beam would be huge. If you want a flood, then you want a shallow reflector just to push the widest of the spill toward the front. If you want any kind of beam, pick another emitter.

sonikaccord · Dec 24, 2013

Thanks for that advice. I will still get the emitter to experiment with, but I won't have my hopes up for a tight clean beam. I've looked up emitter sizes and they should be as close to a point source as possible for max throw.

However, could I use a converging lens to shrink the "apparant source?" Then have the aspheric take over? I realized that matching the focal points would result in a large system, just wondering if it's possible.

inetdog · Dec 24, 2013

No it is not. You cannot get rays in all directions from all parts of the emitter to converge in one place. If the emitter beamed the light strongly in one direction you could do better, but not with an omnidirectional emitter surface.

Sent from my XT1080 using Tapatalk

sonikaccord · Dec 24, 2013

Awesome, I have more research to do. You guys have been great.

HarryN · Dec 25, 2013

Sonikaccord - I spent a great deal of money and time on exactly trying to accomplish exactly what your question asks. Of course others told me that it would not work, and I still tried anyway. Yes, I read the physics books that explain it, but I still could not bring myself to intuitively accept the answer.

What I can say about it, is that while I still don't intuitively understand "why" it is true, my experiments (and wallet) made me a believer that the key to "throw" is a high surface brightness, compact light source. I just could not get a similar "total lumens", but larger light source to achieve high lux levels, even with multi stage optics.

OMG is a custom flashlight company that makes some flashlights with amazing throw. While not cheap, buying one of those is still likely cheaper than attempting to build your own, of course, not quite as satisfying though.

sonikaccord · Dec 26, 2013

I have seen the OMG DEFT-X and it is a beautiful work of art. That's also where I saw the Waiven RLT used. I think this will end up turning more into a learning experience than anything. I have built my own LED flashlight from off the shelf parts, but I think it will be very educational to build my own from scratch, play with a couple of lens designs and see the theory in action.

I have been searching through other people's threads seeing what mistakes they made, what worked, what they would've done differently, etc. Experiences like HarryN's really help because they have done it already.

TEEJ · Dec 26, 2013

The only ways I'd see an increase from standard uses:

if the LED was aimed BACK at the reflector, and the reflector focused a point source where the aspherical lens could then project it from the flashlight. IE: Use the reflector to simulate a smaller point source than otherwise possible.

if the LED was projected as a SMALLER image onto a second aspherical lens that then used THAT image as a smaller point source, and so forth.

As mentioned though, that's a lot of precision aiming going on.

DIWdiver · Dec 26, 2013

From what I've read, the consensus is that you cannot create a smaller image of the original source with increased intensity. It's not just impractical, it's theoretically impossible.

TEEJ · Dec 26, 2013

DIWdiver said:
From what I've read, the consensus is that you cannot create a smaller image of the original source with increased intensity. It's not just impractical, it's theoretically impossible.

Tell that to the ants I popped with a magnifying glass.

What if I simply mounted the LED at an angle, so instead of a flat block, it was at a 45º angle. The intensity would be the same, but the apparent size would be cut in half. It would also tend to send more photons at the reflector wall and fewer out the front as spill.

sonikaccord · Dec 26, 2013

That's what I was saying earlier. Minify the apparent size of the source so that the focal point of the first lens is at the focal point of the second lens.

Sent from my HTC6990LVW using Tapatalk

Reflector Decision for Aspherical Lens

Newly Enlightened

Retired Administrator

Newly Enlightened

Retired Administrator

Flashlight Enthusiast

Newly Enlightened

Flashlight Enthusiast

Newly Enlightened

Flashlight Enthusiast

Newly Enlightened

Flashlight Enthusiast

Newly Enlightened

Enlightened

Newly Enlightened

Flashlight Enthusiast

Newly Enlightened

Flashaholic

Flashlight Enthusiast

Flashaholic

Newly Enlightened

Similar threads