Aspheric lights - recycling sidespill back into the emitter

i think the problem w/ less efficient, is some of that light will reach the lens, and traveling in a random direction. it won’t be a clean beam. I like aspheric throwers because they have a clean beam….spill doesn’t reflect off fog/snow etc.

if that doesn’t bother you, i suppose it is worth a shot.

I tried the reflective tape trick once in a different set-up but what was the same was that light that was not getting out of the flashlight was reflected back to the led with Scotch reflective tape (in my set-up about 50% of the light hit the reflective tape). I was surprised to notice that output did not increase and that led tint did not change. Perhaps the reflected angle of the reflected light is not precise enough to make a noticable difference.

I think that’s exactly the point of the Waiven doodad - the hole is there so that the emitted/re-emitted/re-bounced around light has a relatively narrow angle, which is exactly what you want with an aspheric to avoid weird beam artifacts.

Personally I’m a touch dubious about the reflected light re-exciting the phosphor. I’ve read that LED makers are working on developing a way of doing this so as to reduce the heat loss caused by internally reflected light, but at present they go at great pains to get those photons out of the phosphor and not have them come back (the silicon dome, mainly). My take on the Waiven collar is that it simply bounces around the light emitted off the desired angle (30 deg?) until it find its way out of the hole and hits the aspheric, i.e. the efficiency gains come from collecting and using that previously unwanted/ wasted light, not from any other trickery. Then again, what I know about LED technology can be written on a 20mm star.

First I’d try with a fixed focal length aspheric and use some kind of OP reflector (that fits within that focal length) with the base drilled out and inverted over the LED. That in my mind would be the closest we’ll get to a Waiven collar.

well, the tint shift suggests that re-exciting the phosphor has some role.

I guess someone could estimate how much of the output increase is from redirected light, and how much is due to phosphor excitation with an integrating sphere.

measuring the overall output of an emitter when it is bare, w/ floody reflector, w/ throwy reflector, and w/ wavien collar should give some approximation, since the reflectors would give some idea of the IS’s deviation with respect to beam profile.

Nice .

Lol… I tried exactly that but it didn’t make a difference.

It’ll only work for aspheric light

true and I’m about as far an expert on this as possible. Still, at the end of the day, if someone on here can reproduce it I very much doubt they’ll care the whys and wherefores :slight_smile: I just remembered the name of the reflector I have - Boom SS. When I get another linear driver in I’ll bolt an XM-l to a block of alu and give it a whirl. Something like: emitter>space (15mm?)>piece of card/box with 20mm dia hole (?)> wall. See how much light hits the wall with and without a Boom SS upended on the LED. MIght fiddle a bit with the aperture on the Boom too. Should be a fun little project and if there are any easy gains to be had they’ll be apparent pretty quickly.

@ezarc - that made me laugh :slight_smile: The whole contraption might be too long to fit within the focal length of an aspheric though, as they’re pretty short (10-15mm?).

yeah, i’m definitely not expert, and like you I’ll take ANY gain in that kind of light…

since wavien collars are too expensive for most folks, we might not see budget copies unless we do that sort of testing, find an acceptable solution and it becomes fairly common.

iow, this stuff is still too far on the fringe for the chinese to hear about and want to copy. heck, there’s only few ‘regular’ fixed aspheric throwers available…

I’m still thinking that the right size christmas ornament, cut in half, is a good way to begin experimenting. In fact, I think the owner left some in the attic of the place I’m renting. time to go check lol

worth a try, it’s not like you have anything to lose! I bought that Boom years ago for not very much, then found that everyone panned them as being too floody, so it’s been sitting in my parts boxed unused since then. When I get my DrJones driver I’ll hook everything up and have a play.

I remember a thread on mtbr forum where trout and others experimented with making wavien type collars with a ball-end mill in aluminium and Im sure that some gains in output were achieved.

It would be great if a homebrew hack could be found cos I dont want to pay wavien $80 for a reflector!

yeah, they had some success but ahorton didn’t see enough improvement to make it worth pursuing. It also looked fiddly as hell. Perhaps part of the problem was that they focused (doh!) more on the phosphor re-excitation aspect, rather than collecting and channeling the side scatter.

I’m afraid that “collecting and channeling the side scatter” won’t work; the laws of optics forbid it. The wavien collar increases die luminance by reflecting light back to the die, where it adds to the original luminance.
Dedoming does something similar (—> thread)

I thought about this exact thing to do, too.
I used an aspheric zoomie and reflectivetape from an old constructionsitethingwithalamponit (is it called beacon?). To my surprise, I wasn’t able to see any difference in throw.
The main problem with this method is, that the sliding tube for the zooming mechanism is a pretty precise part, so that the zooming wouldn’t work that good anymore.

the Dr. is in!

I finally got around to reading your post. It will take a couple more reads for it to be knowledge, but I did learn a few things already :slight_smile:

I’ve been wondering - how different is the behavior in color emitters - particularly, red and green version of pt-54, sst-90, xp-e, xr-e. I’ve got a green sst-90 sitting here that’s going in an 7g5v2 (aspheric) and i’m wondering if I should dedome.

second question should get us back on topic. what are your thoughts on the size of the ‘recycling’ reflector? The experiments referenced above used a 5mm ashperic aluminum collar - but wavien’s is much bigger. Trial and error seems a very inefficient way to determine the ideal size - what does ‘science’ say?

To my thinking…the larger the reflector, the more accurate its surface and placement must be…and at some point, the intervening air space (and impurities?) could become a factor. But I can’t figure out why it shouldn’t be as small as possible…

Dedoming color emitters - hm, difficult to say, since there’s no diffuse phosphor layer. Reflected photons might not change their angle and thus never get out; the die surface roughness would help a bit though. I assume that the net effect is quite less.

The original wavien collar was intended for the SST-90, thus the size - but it was placed directly onto the dome (if I remember correctly). Having no air gap between dome and LED would be a nice thing to have, but the XM-L or even XP-G2 are quite small, and the dome is not that much bigger than the die, which makes things a bit difficult. Since the die is an extensive source, a small reflector might block some light that would have hit the lens, or not reflect some rays that don’t hit it.

A bigger back-reflector can more selectively reflect only that light that really won’t hit the lens, but it will have a harder job to really hit the die again.

W. Collar in DEFT X is pretty big:


Do you know anything about the collar in this pic? The reason I ask is the wavien callers I have look very different. That one has a much larger opening and appears to have some kind of plastic around the rim.