I fail to see how your light recycler accomplishes a task more effectively than a TIROS, or how it decreases the color temperature of the light emitted. It seems as though a TIROs would also be more efficient than your light recycler design, as it would surround the LED "dome(as it were)", gathering most of the light emitted, and concentrating it...
Anything out ther brighter at a distance than our flashlight?
- Thread starter enginyr
- Start date
Help Support Candle Power Flashlight Forum
I fail to see how your light recycler accomplishes a task more effectively than a TIROS, or how it decreases the color temperature of the light emitted. It seems as though a TIROs would also be more efficient than your light recycler design, as it would surround the LED "dome(as it were)", gathering most of the light emitted, and concentrating it...
tolkaze
Enlightened
Reflecting light onto any non-black surface will re-radiate light, so yes... you could increase the light this way to some extent, but I don't think that you would be doubling it. This principle is sort of trying to cancel out some of the light emitted sideways from the LED dome, and put it back in the middle. This isn't doubling the light, it is replacing some of the lost light in an inefficient manner. I assume they will then use some kind of aspheric to project the image of the die. My question is this, instead of trying to overcome the lost light from the plastic LED dome... simply removing the dome all together would mean more light going forward, and less going sideways in the first place.
The dome is just an optic and protector, this allows the use of a reflector to focus the light. I would simply remove the dome, or at least alter it with a cylinder (flat on top) to let more light out the front and less out the side, then use an optic or aspheric.
BTW, LED's seem to work fine with no domes, and people have already indicated their thoughts on whether or not LUX is increased. As for Lumens, personally I think lumens decrease a bit with a removed dome, but thats just a matter of opinion.
alpg88
Flashlight Enthusiast
- Joined
- Apr 19, 2005
- Messages
- 5,382
as old 1.99 beeper commercials said, we could tell you how we do it, but do you really care?
I'd like to see the light using this tech. tested head to head with other high power lights, and see how it behaves.
I'd like to see the light using this tech. tested head to head with other high power lights, and see how it behaves.
kramer5150
Flashaholic
What is your measurement method? Are you measuring maximum Lux at a specific distance or overall lumen output in a sphere?
The measurement method needs to be understood here before any intelligible discussion can be of any value.... lets talk apples-apples and oranges-oranges.
**EDIT** if you are measuring lux at a specific distance... ANY LED will more than double its lux value, wide open emitter versus emitter with reflector/aspheric lens.
Last edited:
Been watching this thread and laughing. That was a rather botched introduction. FYI my light, the DEFT, is the current king of LED throw and has been so for a long time now. I don't think your proto light can beat mine. You have had your fair share of skeptics here and for good reason. I won't go into all those right now but I will say I think your product may have merit. I don't for one second believe your numbers but I think it may help some. Anything we can do to increase surface brightness is good in my book since that is one of the cornerstones to throw. I am skeptical but willing to see if your technology has merit. I will give you a call tomorrow about getting a sample.
bullettproof
Enlightened
- Joined
- Feb 23, 2009
- Messages
- 461
kramer5150
Flashaholic
Popsiclestix
Newly Enlightened
- Joined
- Feb 14, 2010
- Messages
- 136
So what you're really doing is taking light that is emitted out to the sides of the LED and redirecting it back into the center. Then, some of this redirected light is absorbed by the phosphor and reemitted. This sounds like you're doing nothing but trading the side-emitting lumens for more surface brightness.
One can easily see that this arrangement will not yield more lumens.
Assume 100 lumens are emitted to the side and reflected. If one gives yellow phospher a generous 50% efficiency, only 50 are emitted. Thus, you have lost 100 lumens and gained 50 lumens.
But you are right, it seems that the surface brightness may be higher than our current king, the XR-E R2. Whether or not one can package that into a light and collimate it with an aspheric lens to beat the R2 remains to be seen
I haven't compared it with a TIROS. That might be a good idea.
As for the color temp, the phosphor is the key. Any high frequency light reflected back onto the phosphor re-emits at a lower frequency on average. We have a striking demo of this phenomenon that we will bring to Infocomm 2010.
When we measure total output, we use an integrating sphere. But my claim was that if I added the reflector to the flashlight, all other things being equal, I will roughly double the output (in lumens if you like) of that flashlight. We typically use an integrating sphere for this measurement too but a LUX meter gives us the same results.
The setup for comparing the output with the LUX meter requires us to keep the distance equal and measure in the same location.
I'm not sure I understand what you mean about the wide open emitter versus emitter with reflector/aspheric lens.
So what you're really doing is taking light that is emitted out to the sides of the LED and redirecting it back into the center. Then, some of this redirected light is absorbed by the phosphor and reemitted. This sounds like you're doing nothing but trading the side-emitting lumens for more surface brightness.
One can easily see that this arrangement will not yield more lumens.
Assume 100 lumens are emitted to the side and reflected. If one gives yellow phospher a generous 50% efficiency, only 50 are emitted. Thus, you have lost 100 lumens and gained 50 lumens.
But you are right, it seems that the surface brightness may be higher than our current king, the XR-E R2. Whether or not one can package that into a light and collimate it with an aspheric lens to beat the R2 remains to be seen
That's the tradeoff. Side-emitting photons are lost in a system consisting of an LED followed by a lens (or any other limiting aperture). We are putting a fair percentage of that energy back in play.
As for yielding more lumens, it depends on what you mean. Over the entire hemisphere, or through the aperture.
MrGman
Flashlight Enthusiast
- Joined
- Feb 6, 2007
- Messages
- 1,777
so you are saying that the "white" light that is being reflected back into the phosphor is exciting it some more to emit more white light which turns out to be at lower frequencies (more towards the red and away from the blue). Thus the phosphor is giving off more light from double excitation (the blue LED source underneath and the "white" light being recycled on top) and thus the total light spectrum coming out of the central area is a broader spectrum of white light, higher in intensity and thus increasing both lumens readings at a sphere input or lux readings on a meter?
And you are saying you can still overdrive the LED with higher than rated current by some percentage and yet the phosphor isn't totally saturated already and you can still get more light out with the recycler than with a reflector type system just from what makes it through the aperture and not burn the phosphor or ruin the die?
I am impressed with the DEFT however, I remain unconvinced that a redesign which makes use of our recycling retro-reflector would not outperform your current design.
Our prototype "as is" might not be a fair comparison to the DEFT. It was conceived as a practical way to demonstrate the technology. If you plan to attend Infocomm 2010, please drop by and check it out.
The spectrum from the bare LED is that classic blue LED luminescence spike followed by a phosphorescence curve that maxes out in 550nm region. The spectrum from an LED with the retro-reflector has a similar shape but the peak of the blue spike is closer in amplitude to the peak phosphor.
Higher lumen readings through the aperture. The idea is to match the aperture of the retro-reflector to the aperture of the collimating lens. The higher the f/#, the greater the benifit from recycling.
We have never had a phosphor burn out from recycling. Just from experience, I can tell you that recycling has no ill effect on the phosphor. If you overdrive an LED to failure with recycling, it would have failed without. Over-driving is kind of a separate issue.
Part of the problem in you getting you point across has been a lack of proper explanation or the use of the right terms. To say your device doubles the output is very misleading. It would be better to say you are increasing the throughput or increasing surface brightness. These terms will be better understood here.
I am very experienced with overdriving LEDs. Your technology will stress the phosphor more than if it was not there. That is unless you would like to claim zero absorption of the reflected light into the phosphor. We both know that is not possible. Some of that light will die as heat in the phosphor. This therefore means more thermal stress on the phosphor. That said if your device works it would not be necessary to overdrive as much to get the same results so the same end-result could be achieved at greater efficiency.
BTW I talked to Harry and I am still waiting on the form to fill out. I am very curious to see if this technology works. If it does it will help in an area that has been all but abandoned by LED companies. Namely surface brightness. This is very important for projectors and flashlights.
Yes exactly, and after all of this communication time & energy is spent, my guess is that we'll all eventually come to the grounded understanding that lumens aren't being increased, multiplied, or amplified here. I still think the OP and optics guys are referring to increased lux at a distance or what we used to generically refer to as "candlepower."
If lumen increase were possible by this technique it would be perpetual in that you could keep "recycling" the light and attain a mean increase each time it was recycled. It also raises the question of why so many LED manufacturers failed to ever stumble up this phenomenon given the billions of dollars and millions of man hours already invested into LED technology. Certainly if it was possible to double the lumen output of an emitter by redirecting some light back into the die, it would have been explored by these competing manufacturers. In this industry an output gain of 10% with no increase in current is considered a big leap. Remember here that the claim is double and that's not just lux but overall lumens.
If they're using an I-sphere that's really indicating double the lumen output, it sounds to me as if it's not baffled properly. Is this even possible in a certified, commercial I-sphere? I don't know but Gman probably would.
Very Interesting Concept, by collecting lights to the side back to LED &
re-emit out the front using a reverse reflector called "Recycling Collar".
A few things that I'd like to point out.
1, It will not increase total lumen out put of the LED, lumen is a type of energy measured via the form of light. One can convert energy, but one can not generate energy out of nothing.
The lumens from bounced back beam & regenerated into new light will be less than the original lumens blocked by the "Recycling Collar".
the measured lux out via the hole in the center of the "Recycling Collar" might increase, but the total lumen output of LED is reduced.
Therefore, it is really a collimation system, although an innovative one.
2, OP need to demonstrate efficiency of the recycling by quantitative data.
White LED use yellow die for a reason. The yellow phosphor absorbs blue light (Excitation wave length is blue) & re-emit white light. When the light being bounced back, it's white light, only a portion of the spectrum is able to excite the yellow phosphor again. The efficiency might be less than anticipated vs. if all the light bounced back is blue light.
3, The "saturation effect", a hard driven LED might have a "saturation effect" where the majority yellow phosphor have been fully excited.
Therefore, the ability to get more light emitted by increasing excitation density is greatly reduced. This is especially true with over driven LED, takes a lot more additional "driving" to get a lot less additional light out at near saturation level.
The "saturation effect" is everywhere, it's basically a point where 1+1 no longer equals 2. For example, if one guys "tool" is 7 inches long when fully excited by one women, you add another women, it won't grow into 14 inchs, 7.5 inches maybe. The ability to grow has been met with "saturation effect".
4, The surface brightness might increase a little, but because it's really a collimation technology after all, you lose the light that could have been sent forward to begin with (Those that are blocked by the collar),
there fore this is a trade off. It remains to be seen if it will actually be able to beat out existing collimation systems available.
I would not expect this to be a huge improvement over simple Aspherical.
If used in conjunction with Aspherical, the trade off is that it'll force you
to use smaller lens.
re-emit out the front using a reverse reflector called "Recycling Collar".
A few things that I'd like to point out.
1, It will not increase total lumen out put of the LED, lumen is a type of energy measured via the form of light. One can convert energy, but one can not generate energy out of nothing.
The lumens from bounced back beam & regenerated into new light will be less than the original lumens blocked by the "Recycling Collar".
the measured lux out via the hole in the center of the "Recycling Collar" might increase, but the total lumen output of LED is reduced.
Therefore, it is really a collimation system, although an innovative one.
2, OP need to demonstrate efficiency of the recycling by quantitative data.
White LED use yellow die for a reason. The yellow phosphor absorbs blue light (Excitation wave length is blue) & re-emit white light. When the light being bounced back, it's white light, only a portion of the spectrum is able to excite the yellow phosphor again. The efficiency might be less than anticipated vs. if all the light bounced back is blue light.
3, The "saturation effect", a hard driven LED might have a "saturation effect" where the majority yellow phosphor have been fully excited.
Therefore, the ability to get more light emitted by increasing excitation density is greatly reduced. This is especially true with over driven LED, takes a lot more additional "driving" to get a lot less additional light out at near saturation level.
The "saturation effect" is everywhere, it's basically a point where 1+1 no longer equals 2. For example, if one guys "tool" is 7 inches long when fully excited by one women, you add another women, it won't grow into 14 inchs, 7.5 inches maybe. The ability to grow has been met with "saturation effect".
4, The surface brightness might increase a little, but because it's really a collimation technology after all, you lose the light that could have been sent forward to begin with (Those that are blocked by the collar),
there fore this is a trade off. It remains to be seen if it will actually be able to beat out existing collimation systems available.
I would not expect this to be a huge improvement over simple Aspherical.
If used in conjunction with Aspherical, the trade off is that it'll force you
to use smaller lens.
Last edited:
Fair enough. Throughput is what I mean. I don't want to mislead. We are using an integrating sphere to measure total lumens through the flashlight lens. With this flashlight, using this lens, the reading we get with the retro-reflector in place is almost double the reading we get without it. But just to be clear, the LED is not emitting more photons. We are just getting more through the aperture.
