Cree's New Green XLamp XR-E LEDs are 70 Percent Brighter
- Thread starter saabluster
- Start date
Help Support Candle Power Flashlight Forum
Daekar
Enlightened
That's one nicely-sized step towards maximum theoretical efficiency... now all they need to do is shrink it down, put with a red and blue emitter, and Hello RGB Arrays! :twothumbs
jtr1962
Flashaholic
If we assume a Vf of 3.2V, then these are up to 78 lm/W. At the emitted wavelengths of ~525 nm that would be roughly 15% efficient. Better than before, but still less than 1/3 of what the blue are. Also, for RGB lighting a green emitter of around 545 nm would be ideal. Still, these new emitters should find their way into traffic signals. I'd guess that one emitter running at less than 3 watts should be able to do what a 108 watt incandescent traffic lamp does now. Or it can replace the typical 8 to 11 watt LED signals currently in use using only one-third of the power.
Things should get very interesting once they can match the blues. If nothing else, a 50% efficient green emitter would be around 275 lm/W, easily besting LPS lamps as the most efficient light source.
Things should get very interesting once they can match the blues. If nothing else, a 50% efficient green emitter would be around 275 lm/W, easily besting LPS lamps as the most efficient light source.
PhantomPhoton
Flashlight Enthusiast
It seems they are hard at work for RGBs :devil:.
I don't think they need to make greens as efficient as current blues to make an effective RGB, but more efficient is always better. Because of our eyes sensitivity we need less green lumens than we need blue or red.
I don't think they need to make greens as efficient as current blues to make an effective RGB, but more efficient is always better. Because of our eyes sensitivity we need less green lumens than we need blue or red.
Gryloc
Enlightened
Sorry for the brief post (I usually cannot help to discuss further), but didn't Lumileds have a Green Rebel 0080 out for quite a while. This means that this green Rebel should produce a minimum of 80lm @ 350mA. I probably wouldn't have noticed if I had not bought the green Rebel 0080 recently. This new "70% brighter" green by Cree produces 87.5lm. Cree says 70% brighter, but isn't that because their old green XLamp7090XR emitter only emitted 52lm @ 350mA or the green XR-E emitted 67.2lm @ 350mA? 87.5lm is about 70% better than the 52lm green. Compared to the best green out there, Cree's new green isn't 70% brighter. This makes Cree sound like they are doing well, I suppose.
-Tony
-Tony
frenzee
Enlightened
- Joined
- Sep 4, 2006
- Messages
- 216
I think you are correct Tony. Cree's new green is only marginally more efficient than the rebel 80 green. Still, it is a welcomed improvement. It shows that they are spending R&D dollars on colored LEDs.
PhantomPhoton
Flashlight Enthusiast
Good to know. I haven't been paying any attention to colored LEDs in the past couple years, and even then it was not at all technical attention. I had no idea there was a green rebel 80. But then again I haven't paid much attention to LumiLEDs in the past year either.
Canuke
Enlightened
How do you reach that conclusion?
As I figure it, the ideal RGB triplet is one whose triangle on the CIE diagram covers the greatest area and/or most closely fits to the edges of the diagram. The frequency at the top which minimizes the leftovers on each side is around 520nm. 545nm is down the right side and would leave a large lobe of out-of-gamut colors in the pure green through cyan area. Perhaps you meant 525nm, which is pretty close and a common color?
My current choices would be 650nm-520nm-455nm for RGB.
As the thread is about white light, we are not concerned with Color Gamut, but efficiency, hence the choice of wavelength.
Semiman
jtr1962
Flashaholic
The tri-color optimizations I've looked at give the best balance of color rendering and emitted efficiency by choosing a green emitter in the 545 nm area. Also, the red emitter would be in the 605 to 615 nm area. Granted, these results are not exactly intuitive, but they are what they are. You can get a CRI in the low 80s and an emitted efficiency around 400 lm/W. For don't gain that much higher color rendering, if any, by going with a 525 nm emitter, but you loose quite a bit in efficiency. For CRI much greater than about 80 you're better off with four or five emitters anyway. See here.
SteveDavis
Newly Enlightened
- Joined
- Nov 20, 2007
- Messages
- 66
PhantomPhoton,
The metric of Lumens is a measure of the radiant output, the Watts of light, multiplied by the sensitivity of the human eye at that wavelength. That means that one lumen of red appears as bright as one lumen of green. Its just that the lumen of green requires less radiant wattage. Think of it like density. It's like wattage is volume and lumens is weight. Which weighs more, a pound of feathers or a pound of bricks?
The metric of Lumens is a measure of the radiant output, the Watts of light, multiplied by the sensitivity of the human eye at that wavelength. That means that one lumen of red appears as bright as one lumen of green. Its just that the lumen of green requires less radiant wattage. Think of it like density. It's like wattage is volume and lumens is weight. Which weighs more, a pound of feathers or a pound of bricks?
Last edited:
?? I think what Steve is trying to say is that at 525nm you get 536 lumens per watt (radiant), but at 545nm, you get 665 lumens per watt.
SteveDavis
Newly Enlightened
- Joined
- Nov 20, 2007
- Messages
- 66
PhantomPhoton made the claim in his post that we don't need as many green lumens as red or blue because the human eye is more sensitive to green. I pointed out that the lumens metric already takes this into account. One lumen is as bright as one lumen, at 555 or 380. It simply takes fewer radiant mW at 555 to produce one lumen.
I apologize that my density simile wasn't as clear as it could have been.
I apologize that my density simile wasn't as clear as it could have been.
His post may have actually been in error. White light made of RGB is predominantly Green (from a lumens stand point). We are at a point where Green is almost the limitation is a white RGB source in terms of current LED efficiency. The lack of progress on RED is starting to become an issue, especially when temperature effects come into play.
Semiman
Semiman
Canuke
Enlightened
Ah, I see now
It is true that for general lighting, color saturation isn't what you are looking for, unlike display devices or party lighting. I even made that point in another CPF thread somewhere, then completely forgot it in here. 
