Pertaining to the dual lens approach, the problem is that the light transmission emitted from a domed LED radiates out at a 125º angle thus hitting the aspheric all the way out to the edges where the light has to be bent quite hard back into a collimated beam. By de-doming the emitter a second aspheric can be stationed directly above the phosphor, pre-focusing the emitted light to the outer aspheric for a cleaner more efficient beam profile. Nothing humpty-dumpty about it. Quite scientific, enabling a lesser quality and cheaper aspheric to do greater and more efficient work.
Dale, you truly have a gift for taking the overly technical down to a level where most of us mere “flashmortals” can understand the concepts being described. I’ve had conversations with MEM on this topic and truly find it fascinating but when it comes to the optical theories involved it can be a bit overwhelming. Unless of course you have significant experience in lasers technology as MEM does. I’m just along for the ride but I like it and I’m sure this has helped some people understand what MEM is talking about. :+1:
I tend to agree with this assessment. For an engineer, you seem to be missing some of the simplest concepts. Maybe it is just so far out of your field that you are no more an engineer than we are in that area.
Simple concept: A parabolic reflector takes all light from a specific point (the focal point) and reflects it in parallel lines outwardly. A sphere takes all light from the central point and reflects it back to that same central point.
Simple concept: The focal point of a perfect sphere is zero dimensional. The topography of an LED die is at least two dimensional. Some are speculating that the new XP-G3 may be three dimensional. That means most of the die is outside of the focal point.
The further the light is produced outside the perfect focal point, the less it contributes to throw. This is true for both reflectors and optics. But, in the case of optics, only a very small percentage of the total light produced even gets into the beam at all, because most of it never hits the optic. And a lot of what hits the optic is hitting at the wrong angle for throw. A recycling collar is a hemispherical reflector, which receives most of the light produced by the LED and returns it to the LED for another chance at getting into focus for better throw. So, the percentage of increase in throw is in comparison to the original small amount of throw you get from the system without the collar. In other words, we’re not saying that it increases the actual lumens output of the die. We’re saying that it concentrates and redirects most of the light that would have been wasted back into the beam.
Imagine a highway with thousands of vehicles driving on it. If you have 12 lanes, your concentration of traffic will be very low indeed, and will actually look very sparse. But if you put all those cars into two lanes, you’ll see the traffic density go way up, and travelling speeds go way down as people have trouble moving their cars about. It will be bumper-to-bumper for miles. That’s what RA collars do. They concentrate the light into a smaller stream so that you get more light “density”, which is what we’re calling intensity.
I would prefer a discussion without open and subtle offenses.
I’m curious about this:
I guess the aperture used in this technique blocks some light which would otherwise find its way to the lense on the first go. Is the “photone recycling” in our standard LEDs sufficient to compensate this loss and even gain extra output - or do we need special LEDs? I can’t imagine just the reflection at the light emitting surface can do that.
A simple image for a simple explanation 
The blue is the reflective sphere, the green lines represent the limits of the aperture, normally any light outside of this is wasted, the red is light being reflected back onto the led, this reflected light further excites the phosphor, creating more light, or is reflected, hopefully going in the right direction, if not then it is reflected back to try again and again etc.
As I said a very simple explanation, there is also the blue light that missed the phosphor at the first pass being reflected back that excites the phosphor even more, but you should have the idea now 
Cheers David
+1
Well said. I’ve had enough of that garbage. BLF should be a friendly place to throw ideas and concepts around and ask stupid questions that are followed by helpful responses. It was so for a long time and should return to that. A bit of friendly banter is all good and even welcome but open hostility has no place here. This is a hobbyist forum about flashlights for crying out loud!
Now, back to all the stuff I barely understand. 
It is possible that a recycling aperture would block light that was already heading toward the lens, but that is not the intent. It would be considered a mistake, or poor design. Ideally, you would space the lens and aperture in a way so that the cone of light coming out of the aperture covers the back of the lens. Any light that would have hit the lens on the first go, still would. No special LEDs are needed for use with a recycling aperture. Under normal circumstances, MOST of the light leaving the LED is NOT adding to the throw. In that way it can be considered “wasted” light. If you get most of that light back to the die, you have already made huge gains.
Think of it this way. These numbers are made up, but this is an example of what can be happening. Say you have a flashlight that 40% of the light from the LED hits the lens at the appropriate angle to make up the beam for throw. The rest of the light (60) is wasted. Even if you ultimately only re-collect 70 of that lost light into the beam, that would be 42% of the original output of the LED. Now, the beam has 82% of all the light that the LED is making, which is a gain of over 100%!
I guess you are missing the central point:
a parabolic reflector is perfect to focus parallel light to a single point (or the other way around). But here is no parallel light. The light is emitted in the center of a (hemi-)sphere, radially, and reflected to the center again (apart from the light which is leaving through the aperture hole).
Here you will find further explanation:
https://www.osapublishing.org/oe/fulltext.cfm?uri=oe-21-15-18414
It has always struck me as somewhat odd how the reflectors are designed in the first place. The parabola’s focus point is virtually always determined by light entering the parabola from the outside and creating a hot spot at the focal point. This is great if you have a tiny spot of light source for an LED to reverse the process, but we don’t. And the dome on the LED changes things even more. So a high percentage of light escapes out the front of a reflector in a direct line from the dome, without ever coming into contact with this shiny mirror. This light is the spill, the unfocused emergence of light as radiated directly from the emitter and not collimated. This is why when it comes to throw in a reflectored light, the large reflector collimates a tidier beam.
When we start bending light with an aspheric, things become interesting. The glass itself creates limitations, as witnessed all too well in my multi-thousand dollar Pro lenses on my cameras. You can’t just pop a $3 aspheric in and expect superior results, it’s simply not gonna happen. But still, it takes a very exacting process to properly align lenses to work in the desired fashion, the more lenses the more complicated it becomes and the more chromatic aberrations introduced to the rays of light traveling thus. The complexity creates expense, which of course is why we usually see reflectors.
I’d love to understand it more, just like I’d love to have the spare change lying around for some top glass to go in a flashlight. But, alas, my spare change is pretty much tied up in the hundreds of cells I have on hand for the multitude of lights already in my possession, not to mention my photography kit. Perhaps I should be looking for a malfunctioning Sigma 12-24mm lens on the cheap, it’s got a tremendous aspheric in it that’d certainly be worth pilfering!
Unless you pay buckets of money and have the glass overproduced to avoid edge infarction.
Edit: Case in point… Zeiss Distagon T* 15mm f/2.8 ZF.2 Lens at $5 short of $3000.
This is a photo of an Actual lens cut in half!
15 elements in 12 groups, fwiw
Sharpie, look at the first image in your link. There is parallel light heading to the mirror. But in our case the LED is located in the center of a (hemi-)sphere and emits the light radially.
Transformers!
ROFLMAO!
[quote=Sharpie]
Sharpie, a RA does not try to focus stray light to a direction. It simply reflects the emitted light directly back where it came from.)
Reflector. Light from point A goes to point B depending on angle. (A spherical parabolic or concave or any shape can do this with differing results and efficiencies
RA. Light from point A is reflected back to point A. (A flat mirror can do this with a laser, but it’s a different story with an LED because it emits light in a wide pattern.)
I haven’t followed this whole thread so my observation here may be irrelevant, but I think one thing you may be missing is that one of the primary functions of wavien collars and the like is to use that “misdirected” light to excite underutilized phosphors and thereby create more output from the emitter than it would normally provide.
Sharpie, you bring up some interesting points, and to optimize an LED/RA system lots of these points should be considered. But it is an observed fact that dedoming (some) LEDs significantly increases the intensity of the flashlight beam, and use of a hemispherical RA also increases the beam intensity. I don’t think people here have all the answers, but let’s start with the facts, that these modifications do work to increase beam intensity. They don’t necessarily maximize the total flux output, but beam intensity is the main concern for some flashlights.
Sharpie,
de doming, light recylcling, pre collimator lenses and multiple lenses system really works
Saabluster 1,3mcd monster here
Gaston 1,3 mcd monster here
Mem 1,5mcd here
Ervin Anastasi 1mcd here
They are good for breaking candela records but usability?
When someone makes this with all characteristic of good modded zoomie I’ll buy it