Anything out ther brighter at a distance than our flashlight?

[get-lit]

Surface brightness works in conjunction with optical etendue to produce candlepower and throw:

Relative Candlepower = Relative Intensity x Relative Light Gather x (Relative Optic Focal Length)^2
Relative Throw = sqrt(Relative Intensity) x sqrt(Relative Light Gather) x Relative Optic Focal Length

What is really beginning to ring a bell here for me is that the collar increases luminous intensity while possibly maintaining enough light gather for the trade off to be beneficial to throw just has been suggested; because the same amount of light is gathered through the light paths, albiet with an additional loss due to the reflectance loss of the collar as well as the reflectance loss of the emitter itself. So for example if 50% of the light is captured by the collar and 50% exits the aperture directly, and the collar has a reflectance of 85%, and the emitter has a reflectance of 75%, we retain a total 80% of the light gather:

50% direct aperture light + (50% collar light x (100% - 15% reflectance loss - 25% emitter reflectance loss)) = 80%

or

.5 direct aperture light + (.5 collar light x (1 - .15 reflectance loss - .25 emitter reflectance loss)) = .8

So when replacing a collimating lens with an equivelent diameter collet and smaller diameter collimating lens of matching focal length, we would then have traded 20% light gather for 58% more luminous intensity, resulting in net gains in relative candlepower and throw:

Relative Candlepower = 1.58 x .8 x (Relative Optic Focal Length)^2
Relative Throw Distance = sqrt(1.58) x sqrt(.8) x Relative Optic Focal Length

We would then have a net 26.4% more candlepower and 12.4% more throw!

If the losses could be minimized, these gains could be much more! What I wasn't realizing before is that the light gather of the extra side light is still for the most part retained with the use of the collar, which is what has been throwing me off all along. This is actually very intriguing now. :twothumbs

 

[Th232]

Yes, you've got it!:twothumbs

 

[saabluster]

Just curious, what do you think the drop in output is if you compare a bare LED to one with the collar? Personally I'd have to say that the collar gives us that efficiency hit, but I obviously haven't played with it yet so I don't know how large that drop is.

I have not done those tests myself but the supplied information seems reasonable. It says the bare LED puts out 2136 lm and adding the collar takes that down to 1090 lm.

Now when I say you can get neutral white output without the efficiency hit you have to understand the context which maybe was not clear enough. The context, and really the intent of this technology, is throw enhancement through increasing of surface brightness or intensity.

To get a neutral output LED it normally takes a thicker layer of phosphor which reduces the output and therefore the surface brightness.

Now we have a way to warm the light and increase surface brightness without needing to use that intensity sapping thick layer of phosphor.

If I wanted to take the neutral white LED and have it match the cool white version's intensity I would have to run the neutral white harder just to match the cool white.

Let's put this in more practical terms. I could use the collar on the DEFT and simultaneously reduce the current consumption and lower the CCT. So longer runtimes with better color while having equivalent or greater lux in the existing product.

 

[Th232]

Thanks for the clarification, got it now.:thumbsup:

 

[TorchBoy]

Am I the only person to have started wondering what the aluminium ring on the XR-E really does?

 

[Th232]

My understanding is that it pretty much converts the emission pattern from lambertian to the tighter one shown in the XR-E datasheet. Cree's LED luminaire design guide shows on page 7 that the output pattern is pretty similar to that of a CFL, so maybe that was its original purpose?

 

[saabluster]

My understanding is that it pretty much converts the emission pattern from lambertian to the tighter one shown in the XR-E datasheet. Cree's LED luminaire design guide shows on page 7 that the output pattern is pretty similar to that of a CFL, so maybe that was its original purpose?

The ring is not directly responsible for the tighter beam. That would be the job of the lens. However the fact that the ring sits under and raises the lens does narrow the beam more than it otherwise would be. The purpose of the ring is to allow a mounting point for the lens. It also keeps the optical gel in between the lens and die from going everywhere. They did however make the inside of that ring look like a teeny tiny reflector. All that does is boost the lumen output of the package though. No beam changing effects to be seen from it.

 

[saabluster]

Back on topic. I took some pictures of the LED with and without the collar.

The camera settings are frozen and the current to the LED is unchanged from shot to shot.

First up is the LED without the collar and then with the collar.

As you can clearly see the collar makes a marked improvement in surface brightness and a warming of the color. Interestingly you can see more of a concentration towards the center.

 

[kito109654]

Well doesn't the center concentration make sense because of the shape of the collar? Very nice pictures, the color change is sweet.

 

[MrGman]

can't you crop in those photos somewhat so we can see the die magnified a bit more and less of the empty black space? Good work in finding a way to make a clear visual presentation though.

 

[saabluster]

can't you crop in those photos somewhat so we can see the die magnified a bit more and less of the empty black space? Good work in finding a way to make a clear visual presentation though.

Ask and you shall receive.

Notice how you can now see the light reflecting off the bond wires from the collar.

If anyone wants to take these pictures and make a rotating gif be my guest. That would be the best way to see the change with and without the collar.

 

[MrBenchmark]

It is absolutely amazing that the same amount of electricity is used to generate both of those pictures. People who are still talking about efficiency need to contemplate that for a minute and understand a couple of fundamental points:
1. Not all work is created equal - some work is more useful than other work.
2. Maximizing the useful work is the soul of efficiency.

 

[Painful Chafe]

That's kind of my point, though admittedly ill explained... He stated something about increasing output by putting his light recycler thing on it, but didn't say anything about whether it increased lux or lumens...

Perhaps he is saying that the light collected and redirected doubles the output by using the waisted light. Or he may be saying they are using the side reflected light and redirecting it to the hotspot thus increasing the lux. He isn't doing a very good job describing it though. Not sure why he is being so vague.

 

[Machete God]

(images from saabluster's post #171)

 

[MrBenchmark]

Folks who don't understand how this can work need to look at that picture - clearly the chip is significantly brighter, and yet is consuming no more current. In general, brighter = better. (9 out of 10 threads on this forum agree on this! 😗)

The tradeoff is you lose the sidespill. If the spill is important, this isn't a great idea. If you want throw, or the spill is actively undesirable, say for a projector (you don't want the spill - it washes out whatever is on the screen where you are projecting your image) then this is simply amazing.

One of the things this seems to indicate is that current LEDs leave a *lot* on the table - if the phosphor of the LED were completely saturated by the output from the internal blue LED, I am guessing this idea probably wouldn't be of much benefit.

 

[get-lit]

I haven't seen anyone say it doesn't work for a while now. A lense the size of the collet could utilize all of that size spill too, but it's about how that side spill is utilized differently.

I was one of the last to be convinced because I made a mistake in my inital calculations.

This really has unlimited potential with increases in collect and emitter reflectance, as well as increases in collet and emitter surface precision!

 

[TorchBoy]

Re: Anything out there brighter at a distance than our flashlight?

Folks who don't understand how this can work need to look at that picture - clearly the chip is significantly brighter, and yet is consuming no more current.

But knowing that it does work won't help those people understand how it works. :nana: I think if LED phosphors were close to being overloaded at the rated maximums we'd see more blue-shift when getting close. That would be undesirable.

Thanks for the pics saabluster.

 

[RedForest UK]

Folks who don't understand how this can work need to look at that picture - clearly the chip is significantly brighter, and yet is consuming no more current. In general, brighter = better. (9 out of 10 threads on this forum agree on this! 😗)

Okay, so if I'm understanding this correctly then it's only brighter from the angle the picture was taken from, not overall. That is the point, it takes a chunk of light emitted from the sides of the led, reflects it back onto the led, where 40% or so is lost, and the other 60% reflects back out the aperture. Luckily a higher percentage of blue shift spectum light is absorbed than higher cct spectrum so the light appears 'warmer'.

A reflector would be more efficient overall, as none of the sidespill light would be taking a round trip back off the led and so you would get that extra 40% of sidespill light back if it was a well fitted reflector, however reflectors for some reason can not focus light as well as an aspheric anyway. These dont use the side spill light from an led at all so that is effectively wasted. So with this, in an aspheric set-up you are recouping almost 60% of that wasted light and throwing it out the front via the asperic, thus giving after some calculation an extra 12% of throw?

One more perhaps stupid question; could you not use a well fitted reflector behind an aspheric lense to focus more of the light out of the front directly without having to first reflect it back onto the led and take the efficiency hit, or is this just not a plausible optical design?

 

[saabluster]

It is absolutely amazing that the same amount of electricity is used to generate both of those pictures. People who are still talking about efficiency need to contemplate that for a minute and understand a couple of fundamental points:
1. Not all work is created equal - some work is more useful than other work.
2. Maximizing the useful work is the soul of efficiency.

So true. So true.

(images from saabluster's post #171)

Thank you so much for doing that for me. I've never had much success making gifs. That turned out perfect.:thumbsup:

thus giving after some calculation an extra 12% of throw?

Well first off throw is not the word I would use here as we still have not come to a conclusion on how to quantify throw just yet. (this subject is handled in another thread). Let's use the word lux. This collar will translate into a 58% higher lux reading.

One more perhaps stupid question; could you not use a well fitted reflector behind an aspheric lense to focus more of the light out of the front directly without having to first reflect it back onto the led and take the efficiency hit, or is this just not a plausible optical design?

I have just such a design and while it may be optically more efficient it will not create the increased throw that this collar can.

 

[admacdo]

Ah, this thread has been awesome.
Many times, I've seen industry and technical people wander into forums the world over and with a few misplaced terms, get totally smoked. I've been in three industries where the one term used had different understandings of meaning. It caused some grief each time with some re-education required.

What is gratifying to see on cpf is the level of knowledge and discovery that is shared here, along with the standard level of scepticism.
Look at other forums that you may be on and think about what you can see in this one. This thread has all the traits of a good teledrama.

Event, Surprise, disbelief, support, counter claim, calculations, negotiation, peer review, support, argument, explanation ( close male bonding through calculation ) turning point for one of the protagonists and a successful conclusion which sees potential for the future.

Everyone seems genuinely happy and have learned stuff.

I could almost map the characters here to "Lord of the Rings" (saabluster, you're Gandalf..)

This type of situation is rare in the outside world. Thanks to the participants. It has been so far quite educational to the technical level of competence of participants of cpf, not to mention the general decorum of the members here.

This doesn't happen very often in other places on the 'Net.

Well done, gentlemen.