BLF recoil über-thrower

As far as I know, Hella doesn’t use any optics other than the reflector itself. Just bare LED
Here’s my idea for your parabolic setup:

I think this Spark design is considered as a semi quad recoil too with some spill allowed for peripheral vision. With some tweaking you can get something to begin with before start making any big reflector:

Spark SP6 (5xXM-L, 6x18650, Charger) Review: OUTDOOR BEAMSHOTS, RUNTIMES, VIDEO+ | Candle Power Flashlight Forum

….or cast bigger replica reflector out of it

That’s not gonna work anyway, because the plano side will act as a reflector at those angles and the lens would have to be extremely bulbous and will suffer from internal reflections.

You will spend a larger percentage of the surface area on that tiny bit of extra light though.
I don’t think it’s worth it.

I like collecting as much as possible too, but it is a thrower, an über thrower even, so it would be nice to have a very tight beam.
(edit) …without it being huge in diameter…
There’s still an option perhaps to make it zoomable by moving the LED towards the reflector (or the reflector towards the LED) so that you can collect a little more, but with a divergent beam, like an aspheric zoomie.
But still i doubt if those few % extra collection is worthwhile and / or noticeable.

Yeah, i was thinking along those lines for a bicycle headlight, LED-ifying an old incan headlight more or less like that.
But it’s not gonna be an über thrower like that, but better for vehicle headlights.
There’s no point in having a light saber on a vehicle.

This is true, that’s why low F number lenses are even worse than what I show in my spreadsheet.

Maybe if I have spare time in the future I will make a new spreadsheet for retro reflectors that can give the beam divergence and intensity and all that stuff.

If you have a reflector that collects 120 degrees (60 half angle) that’s 75% of the total light, not accounting for the losses of the beams or center circle blockage.
So going to 100% definitely won’t be a huge difference.

120 degrees would be F0.29

Hey @The_Driver, could you look this over and confirm or deny for me?
——————————————————————————————————————————————
If the center spot intensity depends on the front area, then these two parabolic reflectors would have the same intensity:

(ignore the uneven distribution of the rays)
Both are 120mm diameter, but one collects 180 degrees (100) of light, while the other collects 60 degrees (25) of light.
.
.
So we take the second reflector, and make a lens with an equivalent focal distance and diameter:

Now both the lens and reflector collect the same cone of light, both 25%, and create the same size spot because the surface of both optics is at the same distance from the LED.
Assuming nothing is blocking the reflector, they should have almost identical peak intensity.
.
.
Then, taking into consideration that a wavien collar would increase the intensity by 2.3x, there is no center circle blocking light, there are no “beams” that would hold the center LED in place, and it is easier to provide cooling to an LED behind a lens rather than on a long heatpipe in front of the reflector, I would assume that the peak intensity with the lens would be close to 3x that of the reflector.
.
.
Therefore, compared to a recoil reflector of the same diameter, the lens would be much more powerful for throw, albeit only using 50% of available lumens (compared to a reflector that collects 180 degrees, near 100%)
——————————————————————————————————————————————
Is this correct?

You slightly misinterpreted the numbers in smas test of the Wavien collar. The 2.3x increase in Luminance was only observed a low currents of 700mA (So the resulting absolute luminance was not really higher than is possible without a collar). At maximum current the benefit of the collar decreases slightly to a maximum of 2.2x.

Also the hole in the center of a reflector doesn’t make a very large difference in terms of size (because the size of a circle increases with the square of the radius).

So generally you can just use the 2.2x factor for comparison between lens with collar and reflectors (unless there are additional factors like ar-coatings, special reflector coatings etc.).

The cooling of an LED with heatpipes could actually be better than in a normal flashlight if the heatpipes are connected to large enough heatsink and the LED is soldered directly onto them. In reality I think it will just be very similar though.

Generally while I do find this discussion here very interesting I don’t think the concept here makes much sense.
To me normal reflectors (~75% light collection), aspheric lenses with pre-collimators and TIR-Lenses make sense when designing a flashlight.
The recoil reflector has no large benefit in terms of throw compared to a normal reflector of the same size.
The real thing to do is just to use high-quality reflectors with a large diameter. Nobody is really doing that.

Another difference between recoil and normal reflectors is that with a recoil reflector the light that is not collected is wasted. It is not useful spill light for walking around because it goes into the wrong direction. It either goes behind the light or newer leaves it because of the flashight body.

I think i showed ( BLF recoil über-thrower - #345 by Jerommel ) the focal distance is more even and generally larger with a recoil, resulting in an even, tight beam, better than in a regular set up.

But here’s another thing, learned from reality:
Why is a Brynite 158 a similar (better even) thrower than a reflector light with the same diameter?
It basically uses the light that is otherwise the spill of a regular light.

I just tried this with the SupwildFire 53mm reflector light:
Took off the bezel and the lens, then i put the Brynite lens on top (actually just fits inside the outer rim of the reflector), and it is in focus, like the Brynite would be.
The lens only focusses the spill from the Supwildfire.

Now the recoil setup would use practically all of it together.
This is also visible in my comparison between a Jacob A60 and the mirascope recoil setup.
The far from perfect recoil setup was clearly a lot brighter.

So, what are we missing in all the theory that suggests otherwise?
Maybe it’s the radiation diagram that is confusing?
I think we’re missing something.

Have you tried using a flatter reflector in your “regular setup”?
A flatter reflector will usually produce a more pronounced spot and less corona around it compared to a deeper one.

Well of course it’s similar when the diameter is the same. We have already covered that above. The apsheric should be slightly better because it doesn’t have the dead hole of the reflector in the middle.
Of course this only applies when both setups are perfectly focused, have the same LED at identical current and temperature and both lens and reflector are lit completely by the LED.

I was referring to the drawings here: BLF recoil über-thrower - #345 by Jerommel actually.
A flatter reflector collects less light though.
In the drawing i use the most popular proportions.

According to the radiation diagram of an LED the Aspheric lens would collect considerably less light, but it seems this is not the case.
Either way, the recoil set up will collect what a reflector light and an aspheric light would collect together.
And that’s consistent with the test set up results too.

Did you read what happens when i put the Brynite lens in the Supwildfire? (i edited the post a few times).

Oh I see now, missed that part. Still, 2.2x is very significant.

I was just calculating based on the design I had, which was a 11” reflector with a ~2” center hole and three 1/4” beams, compared to a 120mm lens.
The difference is 5x the area.

Oh I wasn’t talking about a normal flashlight, I meant like having the heatsink directly on the back of LED in a lens configuration, would perform better than having the heatsink at the end of a long 6” heatpipe.
Like this:

I think the most valuable part of a recoil reflector is that you can collect near 100% of light, which is impossible to do with any other lens or regular reflector setup.
The only other way to collect 100% of light is a TIR, but the problem with those is that they suck for throw because they need to be so close to the LED (unless you made a giant 12” TIR)
Going for a recoil reflector that collects less than 180 degrees kinda defeats the point of high light efficiency for the beam, because a lens of the same diameter could be doing much better (over 2x lux)
Unless of course you use a recoil setup that is so large that a lens would be super expensive and impractical, so basically larger than 150mm.

The idea here is that it behaves like a lens setup, where there is only beam and no spill :slight_smile:

Updated the recoil reflector drawing in this post and the OP.

…but i’m having doubts about that lobed radiation diagram again…
To explain my doubts i will make yet another drawing, showing (also to myself) that an aspheric light like the B158 only uses the spill of a reflector light like the HD2010 or that Supwildfire i mentioned (same reflector).
Yet both lights have a similar output.
So my original motivation for a recoil light is still the same: it will use both, practically all of it.

You need to know the depth of the reflector and the focal distance of the lens to be able to compare the % of light they collect properly.
Just the diameter isn’t enough.

Of course. :slight_smile:
But did you read about the B158 lens in the Supwildfire?

Oh no I didn’t read that part.
How do you know the output is the same? Has it been measured in an integrating sphere?
The reflector should have near 100% lumen output since when measured in an integrating sphere it will count both the spill and spot lumens.
The lens should have about 50% lumen output if it is F0.5 and collects 45 degrees half angle.

It honestly would not make much sense for the lens and reflector to have anywhere near the same lumen output.
EDIT- that’s not accounting for the fact that a reflector would probably be like 75% efficient while a lens is 90%+.

I know it’s approximately the same beam output (not counting the spill from the reflector light like you would in an integrating sphere of course) because i have compared the B158 with a C8 when they both had XP-L HI on 2.8 Amperes.
I’ll do some further testing with that light and the B158 lens though.

What do you mean with this?
Normal smooth Aluminium reflectors have ~90% reflectance in the visible range. This is very similar to the 90% transmission rate of good plastic lenses and 92% of glass (depending on type…).

Aluminum electroformed reflectors are about 90%, but a chunk of aluminum that is milled on a lathe or pressed/stamped and then polished is a lot less.
Most regular flashlights (especially budget ones) do not use electroformed reflectors because of the extremely high cost.
And then there’s the super budget flashlights which use a piece of plastic coated in reflective spray paint, which is just horrible.

Oh I see what you mean, the amount of light in the spot.
Yeah, the closer it is to F0.5 the closer they will be to eachother in terms of lumens in the spot.
The reflector will have a bigger spot/corona though.

I think they all use condensed aluminum vapor as a reflective surface though.
Iirc they do that in a vacuum chamber with some pure aluminum on a tungsten filament, vaporizing the aluminum which is then caught on the reflector surface.
Even on cheap plastic reflectors they do this.
The Jacob A60 is a good example of how good a plastic reflector can be.