Reflector vs. Optics: Light Loss

[jsr]

Which one, a Reflector or an Option (such as the TIR), loses more light when directing the light forward? For a fixed flashlight (same everything, but changing the photon management device), would more light come out the window with a properly made reflector or a TIR/optic?

I always thought a reflector was more efficient in this aspect (no data, just a thought) because I don't see any light coming through the reflector. All light is reflected outward toward the window. In an optic, even a TIR, from the pics I've seen online, I can see light when looking at the optic from the side, which to me would seem it's losing light that isn't being directed toward the window.

Thanks for the edjumacashun.

 

[Meduza]

Due to what i have heard, TIR is more efficient... but i really dont know...

 

[MillerMods]

I always thought a reflector was more efficient in this aspect (no data, just a thought) because I don't see any light coming through the reflector. All light is reflected outward toward the window. In an optic, even a TIR, from the pics I've seen online, I can see light when looking at the optic from the side, which to me would seem it's losing light that isn't being directed toward the window.

I'd like to know the answer to this question as well. I share your thoughts exactly.

 

[Long John]

I think a properly made AL-reflector is more effective. A plastic-reflector can be transparent, so it would loose light into the flashlight head.
An optic will not ermit the light out of the front of the light, which is loss to the side angle.
Another possibility could be a combination of i.e. an Aspherical lens and a reflector. But also in this case is also the reflector the factor for collecting the whole light from the side.
So at all I think a perfect reflector (like the MCR's) is more effective in therms of overall ermitting light.

Best regards

_____
Tom

 

[MillerMods]

I agree with that Tom. I've looked for hard numbers and have never been able to find any for chromed reflectors. TIROS specs are easy to find and are usually around 85% efficient. I would think chromed would be in the mid to high 90's range.

 

[NewBie]

Most of the coatings used in flashlight reflectors found on CPF are 70-85% reflective.

Enhanced Protected Aluminum is around 93% and Enhanced protected Silver is around 98%. I've not seen these materials used on flashlights typically found on CPF.

Some optics claim 90% is sent out the front.

http://www.mellesgriot.com/products/optics/oc_5_1.htm


One coating point of reference, and another note on oxidation of bare aluminum:
http://www.molalla.net/~leeper/refcoat.pdf


One of the threads on this, and some measurements that were done is found here:
https://www.candlepowerforums.com/threads/106790&page=3


Information from one of the threads I posted a while ago...


Reflector coatings, pass your pointer over the coating on the side:
http://www.optiforms.com/4000services/4100opticalCOATINGS/41401ocREFLECTIVITYmetal.html


Found another interesting reference for reflector coatings:
http://www.vergason.com/PDFdocs/Performance_Coatings.pdf


A reference for one brand of reflector coatings:
http://www.mellesgriot.com/products/optics/oc_5_1.htm


Another:
http://www.lambda.cc/PAGE69.htm


For the techical folks out there, Melles Griot has alot of interesting subjects to look at (notice the next button on the bottom of the page):
http://www.mellesgriot.com/products/optics/oc_5_1.htm


A cool AR lens coating comparision (run pointer over the coating at the side):
http://www.optiforms.com/4000services/4100opticalCOATINGS/41403ocREFLECTIVITYar.html


A reference for Bright Nickel coatings, placing it at 55% reflectance:
http://www.shd.org.yu/HtDocs/SHD/Vol67/No6/V67-No6-07.pdf


Most of the time, reflectors we have around here are Rhodium or Nickel Chromed.


Polished Aluminum oxidizes fast, some have used car wax to slow the oxidation in the past, when making their own.


Some useful information on AR coatings:
http://www.molalla.net/~leeper/ar_coa~1.pdf

 

[2xTrinity]

Which one, a Reflector or an Option (such as the TIR), loses more light when directing the light forward? For a fixed flashlight (same everything, but changing the photon management device), would more light come out the window with a properly made reflector or a TIR/optic?

I've seen some lights that have a TIR optic that "snaps" onto the LED, and that entire assembly fits inside a standard reflector. One cheap example of this are the Task Force lights Lowes sells (though since that's all plastic I don't think it's particular efficient compared to an ideal lens/reflector combo) This way, most of the light that would normally "spill" out of a standard reflector light is directed into the beam by the lens. Any light that "leaks" the optic is reflected forward by the reflector. Seems like the ideal combo to me, as it also produces a long-throwing hotspot with gradual transition to a dim spill.

In theory, I think both optics and reflectors can be near 100%, but this is using premium materials such as would be found in expensive optical equipment. I suspect for the purpose of maximum throw in a flashlight, a reflector/lens combo as described above is probably the best. For the purpose of maximum lumen output, a shallow reflector that produced a "floody" beam would probabyl be most efficient as most light would head straight out the front.

 

[lightrod]

......I suspect for the purpose of maximum throw in a flashlight, a reflector/lens combo as described above is probably the best. For the purpose of maximum lumen output, a shallow reflector that produced a "floody" beam would probabyl be most efficient as most light would head straight out the front.

I think this is a key point, and provides an advantage – however slight - to the Cree XRE for "lumen maintenance". It also brings in the equally important question of loss through the front window.

Let's say you have a reflector equipped light with a 45% beam exit angle, providing a relatively wide total beam angle of 90 degrees. Let's also say the reflector, which re-directs the LED output that lies outside the 45% angle, is 80% reflective, or "loses" about 20% of the light that hits it. For the XRE beam pattern, about 70% of the output/flux never hits the reflector, but exits directly out the front. The other 30% of the output lies outside the 45 degree angle and hits the reflector, with 20% of that light hitting the reflector "lost". So, 20% of 30% of the total LED output, or only 6%, is lost. With a LuxIII (broader) LED pattern, it works out to about 10% loss.

With the same analysis with a 30 degree exit angle (60 degree total beam width) from a relatively deeper reflector, about 60% from the XRE and 75% from the LuxIII hits the reflector, and the total is 12% loss for the XRE, and 15% loss for the LuxIII. Both exhibit more lumen loss in total than with the 45 degree angle, directly to 2xTrinity's point.

Now this begs the question of how much light is lost through the lens/window. For a wider angle light that will become the determining factor for the end result of total light loss…..

 

[Pinter]

This is a very nicely documented datasheet collection.
http://www.carclo-optics.co.uk/V2/pds/Datasheet105.pdf
Here you can find the efficiencies for all their products, for XR-E, K2, Luxeon III or even Luxeon V targets.
For XR-E optics efficiency is between 89 and 91%.

 

[Curious_character]

Playing with a Quickbeam-type light box, most of my lights seem to run around 75% efficient at getting light out the front. Some are a little better and some a fair amount worse. The comparison is done with the bare LED vs the light with the reflector and lens in place.

The MiniMag 2AA and 3AA LED lights have the interesting property of having the most light output just when the light comes on, when the head is screwed down as much as possible. The total output is down 20 - 30% at the point of best focus, and continues dropping as the head is unscrewed further. So they seem to be losing light behind the reflector. I was suspicious this might be an artifact of the measurement box, but other variable focus lights, including the Mag 2C with LED PR2 replacement, show a pretty constant total output as focus changes.

c_c

 

[CM]

Most of the coatings used in flashlight reflectors found on CPF are 70-85% reflective.

Enhanced Protected Aluminum is around 93% and Enhanced protected Silver is around 98%. I've not seen these materials used on flashlights typically found on CPF.

Some optics claim 90% is sent out the front.

http://www.mellesgriot.com/products/optics/oc_5_1.htm


One coating point of reference, and another note on oxidation of bare aluminum:
http://www.molalla.net/~leeper/refcoat.pdf


One of the threads on this, and some measurements that were done is found here:
https://www.candlepowerforums.com/threads/106790&page=3


Information from one of the threads I posted a while ago...


Reflector coatings, pass your pointer over the coating on the side:
http://www.optiforms.com/4000services/4100opticalCOATINGS/41401ocREFLECTIVITYmetal.html


Found another interesting reference for reflector coatings:
http://www.vergason.com/PDFdocs/Performance_Coatings.pdf


A reference for one brand of reflector coatings:
http://www.mellesgriot.com/products/optics/oc_5_1.htm


Another:
http://www.lambda.cc/PAGE69.htm


For the techical folks out there, Melles Griot has alot of interesting subjects to look at (notice the next button on the bottom of the page):
http://www.mellesgriot.com/products/optics/oc_5_1.htm


A cool AR lens coating comparision (run pointer over the coating at the side):
http://www.optiforms.com/4000services/4100opticalCOATINGS/41403ocREFLECTIVITYar.html


A reference for Bright Nickel coatings, placing it at 55% reflectance:
http://www.shd.org.yu/HtDocs/SHD/Vol67/No6/V67-No6-07.pdf


Most of the time, reflectors we have around here are Rhodium or Nickel Chromed.


Polished Aluminum oxidizes fast, some have used car wax to slow the oxidation in the past, when making their own.


Some useful information on AR coatings:
http://www.molalla.net/~leeper/ar_coa~1.pdf

Good stuff. I just bookmarked all these.

 

[NewBie]

On the window light transmission losses.

2 sided AR coated Glass:

Regular materials without special AR coatings:

 

[jsr]

Wow, I think my head's going to explode from info! Thanks Newbie and everyone else. I'll take a look at those links sometime.

 

[BPH]

NewBie,

You are unbelievable. I have been working for 20 years as an engineer in a research lab, and believe me I know how hard it is to get good data and information. You continuously amaze me by providing a wealth of high quality information presented in a very usable format. I guess the only explanation is you must really love this stuff, because money does not buy this kind of dedication and devotion. My hat is off to you.

-BH

 

[65535]

What about a TIR type optic with a metalized outer layer to help manage stray light.

 

[LightBright]

Hey Curious_Char - do you happen to have a Fenix P1D-CE to try? The tough part is testing it without the lens/reflector.

 

[BPH]

Great thread, great posts.

I don't buy that LED TIR optics get you close to 90% transmission.

My reasons;

1.)[font=&quot] [/font]Most PDF data sheets say "up to 90%", or more typically, "up to 85%". I don't want to know the max possible efficiency, what about the minimum or typical?

2.)[font=&quot] [/font]Most vendors use optical grade PMMA (Acrylic), and list the transmission as >90% for a 3mm sheet. This is much better transmission than I expected for PMMA, but let's give them the benefit. The optical path for a 20 mm diameter optic looks to me like 6 to 12mm and more for the larger 26 mm diameter optic. So, if the 3 mm test sample gets you just over 90%, what are the losses for a 5, 10 or 15mm path through the actual optic?

3.)[font=&quot] [/font]Just because the optic is called "TIR" does not mean you get total internal reflection, it means this is what was attempted in the design, and so there are losses here also.

4.)[font=&quot] [/font]Additionally, some vendors like Carclo list the "EFF" of their lenses (78-84% for 20mm Lux I) and they have a PDF which describes how "EFF" is defined, and it's a bit more complex than one might think. So I would expect the "EFF" numbers to be optimistic.

So, I think if you get a TRUE 80% of the LEDs lumens out the front of a typical LED optic, you are doing quite well.

I like the Cree XR-E, but the optic choices look slim. I have contacted several vendors and received data sheets for the Cree, but the performance seems to be far below the Lux optics. For example:

Carclo 10199 "tight" (for Cree XR-E), 12.9 FWHM angle with a peak of only 6.7 Cd/lm.

Carclo 10003 "narrow" (for Lux 1), 12 FWHM angle with a peak of 17 Cd/lm.

I am now realizing how critically important the optical elements are to the design of a light and how they must be matched with the specific LED you are using.

 

[Curious_character]

Hey Curious_Char - do you happen to have a Fenix P1D-CE to try? The tough part is testing it without the lens/reflector.

I have a P1D CE but haven't taken it apart -- and won't until I really need to. Maybe when the next super-Cree LED comes out. But the P1D CE total output is pretty similar to a number of other Cree and Seoul - fitted lights I have, so the efficiency is apparently not exceptionally good or bad.

c_c

 

[MillerMods]

Newbie does it again! A very nice resource of links for flashlight nuts!

 

[NewBie]

Great thread, great posts.

I don't buy that LED TIR optics get you close to 90% transmission.

My reasons;

1.)[font=&quot] [/font]Most PDF data sheets say "up to 90%", or more typically, "up to 85%". I don't want to know the max possible efficiency, what about the minimum or typical?

2.)[font=&quot] [/font]Most vendors use optical grade PMMA (Acrylic), and list the transmission as >90% for a 3mm sheet. This is much better transmission than I expected for PMMA, but let's give them the benefit. The optical path for a 20 mm diameter optic looks to me like 6 to 12mm and more for the larger 26 mm diameter optic. So, if the 3 mm test sample gets you just over 90%, what are the losses for a 5, 10 or 15mm path through the actual optic?

I am now realizing how critically important the optical elements are to the design of a light and how they must be matched with the specific LED you are using.

Keep in mind that there is a 4% loss at the input side and the exit side for your PMMA sheet. This is included in that typical 93-4% transmission spec.

Before adding, one needs to subtract the these index losses. Example, 3% loss input and 3% output loss adds up to 6%. A typical PMMA plastic has a 93% transmission, or a 7% loss. So, we take that 7% and subtract 6%, and there is 1% loss in the material.

Picking out a specific acrylic, such as a very generic low cost Cyro brand known as Acrylite that you can pretty much buy in most cities in the USA, or online from companies that distribute it like Tap Plastics ( http://www.tapplastics.com/info/acrylic.php? ), such as the GP or FF series, you will find that there is only 0.5% loss per inch of thickness.

It is explained more in detail for their specific plastic here (long download, huge file size), see page 4, under the colorless discussion on the right hand side, under Visible Light:
http://www.cyro.com/Internet/SiteCo...0E3A9/$File/1213F+UV+trans+TB.pdf?OpenElement

Some of the standard off the shelf molding PMMA plastics are a bit different, where they range from roughly 1% loss per inch (Daikon MH 254) up to 6% loss per inch (Perspex MS 499). See page 35:
http://www.lumileds.com/pdfs/AB27.PDF