Where does the light go in a Supfire F3 zoomie?

I was answering a question from cajampa in the XP-E2 Torch thread and while I was at it I dissected the F3 to see where the light losses go. I thought it was nice enough information to start a topic about.

The supfire F3 is (apart from the stock switch, I had to replace it with a small Omten plus do a spring bypass ) a well build 18650 zoomie with a relatively long focal length lens (so a lot of travel on the slider, a small hotspot and a big difference in light output between flood and throw).

The light is modded with a 2.10A AK-47C(6x7135 chips) lineair driver and a dedomed XP-E2 R3 1D0 led. I blackened the brown led-board-screw-in ring for a better beam when zoomed-in. In my integrating sphere I measured the -100% setting- light output 30 seconds after switch-on: fully assembled (slider in flood and throw setting), with the lens removed (slider in flood and throw setting), with the head removed, and with the screw-in ledboard-fastener removed. And last I already had the bare led output outside the flashlight on my cooled copper mount (although the cooling regime was a bit different on the mount compared to the after-30second-situation in the flashlight). These are the results:

complete zoom out:280lm minus lens 'zoom out': 231 lumen minus head: 381 lumen minus black ring: 413 lumen (and bare led: 427 lumen)

complete zoom in: 90 lumen minus lens 'zoom in': 81 lumen

Apart from these numbers I tried to single out the light loss by absorption/reflection of just the lens of the F3. I did that by holding a bare ledboard with a dedomed XP-E2 at the entrance of the sphere, and then again with a lens (held with a forceps) in front of it, in both situations making sure that the complete light-cone was entering the sphere (who said that I don't have steady hands , holding a lens at the tiny edge with a forceps, and the ledboard, and making a picture with my phone camera with my other hand, I did not drop the lens in the sphere!) . I measured twice, the measured light drop was 11% and 12% resp.

So now you know where all that light goes in this zoomie (if you did not expect that already): when zoomed out, from the 100% of the bare led output, 11% is absorbed by the black ring, 15% gets lost inside the head, 8% is absorbed/reflected by the lens, just 66% makes it out of the flashlight. The lens helps bending the light so that a bit more gets past the bezel, but it also absorbs/reflects some light. When zoomed in, the light that gets lost inside the head increases drastically, it is 65%, so including the 11% of the black ring and the absorption by the lens, just 21% of the light makes it out of the flashlight (although it is all in an intense narrow spot).

I don't think in other zoomies it is much different, although if they have shorter focal length lenses the light loss when zoomed-in is more limited.

Very interesting finds, now I know WHY I don’t like Zoomies! :smiley:

The aspheric itself can be and IS a big limiting factor in getting the light out of the host. Even expensive ones can be like this. I have one that was another member’s project that he paid $80 for, and it just sucks! The best I’ve managed through the aspheric is 161Kcd, with the stock reflector in place I’m seeing over 300Kcd. This project was frustrating enough that the entire thing was dumped in my lap in disgust. And eventually, flat out given to me! So now I too have a project that is frustrating and seemingly impossible. But I’ll figure something out, and the new member MEM is giving some fabulous information on aspherics lately! :wink:

Great info djozz.

Do you have any plans to replicate the tests with a TIR optic (aka Led Lenser style). As I’d love to know how they fair in comparison.

Yes Dale i feel your pain ....

When members give me 180Kcd lights for free it's certainly frustrating .

Give me your tired, your poor.

Your huddled masses yearning to breathe free,

The wretched refuse of your teeming shore.

Send these, the homeless, tempest-tost to me,

I lift my lamp beside the golden door!"

The project was intended to hit 1 Million Cd or 1,000Kcd, so you might be able to see how 161Kcd could be frustrating after spending a couple 3 hundred dollars trying to achieve this goal. Custom heat sink made by another member, boards designed and drivers built, to no avail.

I guess if I were to sit back and just look at the flashlight it would be a gift to enjoy, but I can’t do that. I HAVE to pick up the challenge and find that mega long distance it was designed for.

(Shhhhhh! Then I’ll send it back to him!) I say it like that because at this particular time he doesn’t ever want to see it again.

Very interesting djozz :slight_smile: this is what i was wondering about.
I think much of this can be extrapolated to most zoomies.

The more i learn about aspherics the more it looks to be that multiple lenses is really needed to get the lumens and spot size up effectively.

But i have to say i misunderstood what you meant by zoomed out every time you wrote it, i find it easier to use zoomed & flood to describe the two modes, because what you describe as zoomed in, is what i would call zoomed out :~

Did you guys know that you can coat lenses with AR at home with a spray & the oven, i haven’t tried it because so far i haven’t seen a AR coating that doesn’t make the tint much worse, but for more throw maybe it could be an DIY option.

+1

i just replaced the s2+ blue reflector with a TIR optic, would be nice to know :slight_smile:

I wonder, what would happen to these findings if a wavien collar type device were used to aim the core light into the aspheric and recycle the side emissions back into the phosphor?

The problem with a zoomie is that the used light cone differs when zoomed in and out. So the wavien collar opening ideally should vary with it. If you use a fixed opening optimised for throw modus, in flood modus you loose a lot of light and the flood is narrowed. If you use a wider fixed opening optimised for flood modus, there's not all that much gain from the collar.

Thanks Jos, now that you’ve said it it makes perfect sense. And that also helps me in my thinking towards a fixed aspheric I’m working on… :wink: