Reflector, efficiency?

Nowadays, most reflectors use aluminium as reflection material. In optical instruments, the reflection coating may be aluminium, silver or dielectric. The efficiency of aluminium in instrument is around 85%, silver may be as high as 95%, while advanced dielectic coat can be 99%.

My question is, anybody has some idea on the reflection efficiency of flashlight's reflector? some engineer told me it is around 70%, is it true? is it possbile to apply some higher efficiency coating?

As far as I know a fairly low estimate of around 70% efficiency is probably correct. That's why I wish we would see more flashlights, especially LED lights, using optics, which offer potentially greater control over the light, and potentially much higher overall efficiency (over 90% is certainly possible with TIR optics) In addition, in the case of lights designed for throw, LEDs especially can waste a lot of light as spill as much of the light is projected forward, never even hitting the reflector in the first place. The use of an optic instead captures ALL the light, rather than just the component that is emmited sideways.

I love the flawless beam of reflector, although it may has lower efficiency. If the 70% is an accurate value, is it possible to apply some better coating tech? I heard that "enhanced" aluminium coating can achieve 90% reflection rate. Futher, for high-end flashlight, dielectric coating may be applied.

2xTrinity said:

......... The use of an optic instead captures ALL the light, rather than just the component that is emmited sideways.

Click to expand...

I have yet to see a TIR optic that didn't appear quite bright from the side, looking at the optic itself. That is light that is escaping and not being reflected by the reflector portion of the optic. These TIR optics are a combination reflector and lens. I believe PK mentioned that SF using some quality optic material has achieved better efficiency with their TIR optics over the SF reflectors. There is some great glass, plastic and coatings for lenses as well as reflective surfaces to be sure. The real gain, tempered by cost, to a flashlight is beyond me.

The Pelican recoil reflector is exposed essentially to all of the light produced by the LED and it does a nice job of collimation. With a good reflective surface, such a reflector may be a winner in terms of percentage of light managed as well as percentage of light transmitted beyond the flashlight.

It all boils down to A) getting the light out of the flashlight and B) directing it where you want it. I believe the interest in this thread is primarily in the process mentioned in A. If one looks at efficiency with both A and B combined, that is to say percentage of light output delivered to target, a good solution might be a surprise if one is only considering A or B.

McGizmo said:

It all boils down to A) getting the light out of the flashlight and B) directing it where you want it. I believe the interest in this thread is primarily in the process mentioned in A. If one looks at efficiency with both A and B combined, that is to say percentage of light output delivered to target, a good solution might be a surprise if one is only considering A or B.

Click to expand...

Take the part about light out of this, and it's true for so many things.
-Winston

I care more about A) getting the light out of the flashlight, some spill light is quite useful for general use. When looking at a reflecting surface of reflector, we could note some are brighter than others, we called them "higher quality reflectors".

I've tried the new optics (creed) of KL5 and L1, they improved much but the beam is not as nice as that produced by a good reflector.

agree to what McGizmo typed.
That light all around and behind an optic just shows there is no chance 90 % of the light going into comes out front.

Imho optic/reflector are totally equal when it comes to efficiency:
reflector looses much light at the reflecting part,
optic has a media transition at the entry (some 7-10 % loss), some have a total reflection at the upper part, then all have a reflection just at the same part as a reflector and finally another media transition at the end.

I still have not seen any optic where I liked the output more than with a good reflector

McGizmo said:

I have yet to see a TIR optic that didn't appear quite bright from the side, looking at the optic itself. That is light that is escaping and not being reflected by the reflector portion of the optic.

Click to expand...

Have you seen one of Inova's TIROS optics in action? Officially, they're 90% efficient, but Inova does not have a reputation for honesty in advertising.

(I'm not about to freeze-pop my TIROS lights to visually check their efficiency myself.... )

funder said:

Nowadays, most reflectors use aluminium as reflection material. In optical instruments, the reflection coating may be aluminium, silver or dielectric. The efficiency of aluminium in instrument is around 85%, silver may be as high as 95%, while advanced dielectic coat can be 99%.

My question is, anybody has some idea on the reflection efficiency of flashlight's reflector? some engineer told me it is around 70%, is it true? is it possbile to apply some higher efficiency coating?

Click to expand...

It was measured at 70.7% (or ~65% if you include the lens). Note it's not 85% because of losses from anti-oxidation additives in the coating, so unless you are willing to seal the flashlight head in an argon atmosphere (which would make it rather difficult to change the bulb)...

Today I refered one of my friends who are making relector telescopes, he give me a test report of his 10 inch reflector mirror, it is coated by aluminum and protected by an SiO2 coating. The test report has it that the reflection rate from 400nm to 750nm is better than 90%.

Why the reflector of flashlight only has 70%?

bfg9000 said:

It was measured at 70.7% (or ~65% if you include the lens). Note it's not 85% because of losses from anti-oxidation additives in the coating, so unless you are willing to seal the flashlight head in an argon atmosphere (which would make it rather difficult to change the bulb)...

Click to expand...

That idea isn't actually that absurd. I can buy argon-filled high-power incan lamps with built-in reflectors for about $2 at the hardware store (a typical PAR lamp). While swapping an entire lamp assembly might get expensive for incans with short life, something like that with longer-life LEDs might actually make sense, although short of a huge economy of scale like household light bulbs it probably will be prohibitively expensive. Household bulbs also appear to use the same cheap 75% efficient reflector surface as well, when comparing the stated lumen numbers... (Of course, a well-placed PAR lamp is a lot more efficient than throwing a standard omnidrectional bulb under a lamp shade...)

agree to what McGizmo typed.
That light all around and behind an optic just shows there is no chance 90 % of the light going into comes out front.

Imho optic/reflector are totally equal when it comes to efficiency:
reflector looses much light at the reflecting part,
optic has a media transition at the entry (some 7-10 % loss), some have a total reflection at the upper part, then all have a reflection just at the same part as a reflector and finally another media transition at the end.

Click to expand...

While It's true that given the existing optics and reflectors commonly used in flashlights, the two are about the same. However, I believe there is a lot more affordable room for improvement with the optics. We've yet to see any high quality custom Cree optics (ie AR coated glass, NOT plain epoxy) for example. Crees are especially well suited as they naturally emit most of their light forward to begin with (less light to be lost to the sides). Another option may also be to use some sort of combination optic AND reflector to harness that little bit of light that "leaks" from the optic, and turn 65% of it into usable spill...

Well Argon is only maybe 45 cents per cubic foot at the welding supply store so it's not the price of the gas.

It's the requirement for an inert atmosphere at every stage of manufacturing and storage between the reflector coating and assembly. I can't see a 15% difference in light output so it would be a considerable expense for no visible benefit to the end-user. And if the atmosphere were compromised even for a moment at any stage the losses could easily exceed that 15% as the coating tarnishes--perhaps even to the point of being visible and causing warranty problems.

The extra expense is justifiable for a laser or instrument but probably not for consumer lighting, where it's more important to reduce costs to be able to offer the product at a reasonable price.

funder said:

The test report has it that the reflection rate from 400nm to 750nm is better than 90%.

Why the reflector of flashlight only has 70%?

Click to expand...

just a guess:
because noone really has had a good actual reflector tested (?)
and because those 75 % have been common some xx years ago and its still inside the minds (?)

Reflectors, lenses, internal reflectors; they are all viable and all can be enhanced with the proper use of materials and coatings. On that, I think we all will agree. These are used typically to redirect the flow of photons from the light source. Once we are satisfied with the new direction the majority of these photons have taken, we can address the minority of photons who aren't showing up where they are supposed to and see what can be done to change this. Improving the reflectivity of a reflective surface is a no brainer; provided it is cost effective and doable. Improving or adding an AR coating on an element to enhance the throughput and exit of light through it is also a no brainer; provided it is cost effective and doable.

In specific terms of reflector efficiency, one can direct a stream of photons of known power at the surface and measure the power reflected. For sake of argument, lets say a reflective film has been measured with a 90% efficiency. This does not mean that a reflector that uses this film will deliver 90% of a LED's output. If the LED's stated output was limited to that which in fact hits the reflector then perhaps it would be close.

In simple terms, can todays flashlight reflectors be improved? No doubt! Would such an improvement be viable? No idea! Is a reflector better or worse than a TIR optic? IMHO, yes: both. :nana:

Newbie did complain about reflector manufacturers that speced 93%. They tarnish over time just as you'd expect, maybe even to 65% before you even get them depending on storage conditions.

And he also explained the AR coating reduces interface losses on lenses and optics.

T1 (old model) TIROS at work.
No wonder to me, just as I guessed: everything lighted up by the light that got out of the optic NOT at the front, no other light source present.
imho no way this could be 90 % effective (say more effective than a reflector)

funder said:

...The efficiency of aluminium in instrument is around 85%, silver may be as high as 95%, while advanced dielectic coat can be 99%...

Click to expand...

Thank you for this topic, funder.

I used to drive a Ford Model T for -all- my daily transport needs. I racked up about 25 thousand city-miles before one day the car got crashed.

Its headlights were electric, with the then-new for 1922, "H" lenses. The reflector bowls were large and silvered, and for my first year of driving, the original plating (aways thin, shame) was far less than optimal.
It was darkened by sub-plate tarnish of the copper strike coat.
That is to say, one could polish the silver bright, but it was no longer opague to the copper underneath.

I had my car's reflectors done over in a high-tech aluminizing, touted as superior to silver. It wasn't.
And because the aluminum coating, perfect-looking as it was at first,
was not proof against corrosive gases, it failed very soon.

So I had the bowls replated in pure silver. The light output improved noticeably, I think, better than the aluminum.

Plus, silver was polish-able. But, even though the nature of Ford's headlight design precludes sealing the assembly from gases and water vapor, I never needed to polish the fresh-plated reflectors. I drove the coupe for a year and half of so before the loss. This car was kept out of doors year-round in South Florida.
I guess that the low-sulfur fuels of today were a help?

Those 1915-basis headlights, represented with their 1921 "H" lenses, the first scientifically designed, focused, beam-shaped auto lamps ever.
They shone with modern bulbs, as bright as modern cars' standard low beams.
My bulbs were made by myself from -axial- filament, mere 20W halogen bipins.
The beam pattern of the Ford H lens system was a flattened rectangle, perfectly anti-glare.
They knew their onions*, those scientists at Ford's labs.

*catchphrase of the day: "she sure knows her onions!"

Point: silver is cheap, low-tech and polish-able.
And I doubt it would necessarily tarnish for a very long while in today's quality flashlights.
It just works and it is white, and yet,
silver is not an option today? Why?

funder said:

...silver may be as high as 95%, while advanced dielectric coat can be 99%.

Click to expand...

Q: what is "dielectric"?

Common mirror use aluminium or silver as reflective meterial, the dielectric coating is something other than metal, the meterial used in dielectric coating, for example Ta2O5, has no reflectivity for itself, but when many many layers combined, it can achieve very high reflection rate. For example, in some high-end (zeiss FL 8*32) roof prism binoculars, 70-layers dielectric coating applied in pachan prism. Another example is the diagonal for telescope:

http://www.televue.com/engine/page.asp?ID=57

Silver has better relectivity than alumimium, however, the silver relfective surface, if is exposed in air, is easy to react with Sulphur in the air and result in faint into black.

Dielectric is the best reflective surface known so far, the only problem is, it is expensive, is it worth to upgrade? I have no idea.