Stephenj37826
Well-Known Member
Where is the ceiling to umol/j in phosphor leds? If I'm understanding the way phosphor conversion happens then we are going to hit a ceiling somewhere in the range of 3.72 umol/j.......... Am I missing something?
Phosphor conversion based LED limit......
- Thread starter Stephenj37826
- Start date
Stephenj37826
Well-Known Member
I base this assumption on the the QER if the photon source. Royal blue QER is ~3.72. If we are only starting with that number then it seems impossible to achieve a higher photon count. The 300 lm/watt is more to do with the lumen sensitivity curve ie the closer to 555nm the higher the LER of each wavelength.
Stephenj37826
Well-Known Member
This is of course if there is no recombination of lost energy in conversion into more photons......
guod
Well-Known Member
3.72umol/J from the blue source and after the phosphor conversion we have around 4.7umol/J....damned physics
Bill Lidgate
Active Member
As to phosphors there are things called quantum dots (QDs) which are tiny semiconductor beads made of layers of (usually) cadmium and selenium that at least one company QD Vision / Color IQ is trying to add to display screens. Traditionally white LEDs were YAG:Ce phosphors combined with nitride-based blue LEDs, but its not usually the (blue light -> Phosphor ->white light) reaction that's limiting, its the initial (electricity -> LED-> blue light) process that's currently ~15-20% efficient, I believe.
Check this hybrid QD/LED (device B) emission spectra:
the current QD/LEDs are being sold for having a much broader color gamut, lots of IR and a higher CRI, not really whats needed in Ag stuff.
Check this hybrid QD/LED (device B) emission spectra:
the current QD/LEDs are being sold for having a much broader color gamut, lots of IR and a higher CRI, not really whats needed in Ag stuff.
Last edited:
guod
Well-Known Member
"The use of cadmium is generally decreasing due to its toxicity (it is specifically listed in the European Restriction of Hazardous Substances[3]) and the replacement of nickel-cadmium batteries with nickel-metal hydride and lithium-ion batteries."
https://en.wikipedia.org/wiki/Cadmium
Cadmium-free quantum dots
http://www.nanocotechnologies.com/what-we-do/products/cfqd-quantum-dot-films
Bill Lidgate
Active Member
Nanoco (with DOW) is a pioneer in this area, but Quantum Materials and Nanosys are still making tons of Cd containing QDs. On the bright side the Cd is sequestered in a durable shell, but the toxicity once released is well documented.
Stephenj37826
Well-Known Member
So the 4.7 umol/j is qer of roughly 3000k. This is the photon count of 3000k spectrum per joule of light output. One of two things will/is occur/ing (A we will hit a ceiling of 3.72 umol) (B the phosphorus is not simply "peeling" energy off each respective photon but rather redistributing the energy)
Can someone please clarify this for me if I'm wrong.
Stephenj37826
Well-Known Member
As to phosphors there are things called quantum dots (QDs) which are tiny semiconductor beads made of layers of (usually) cadmium and selenium that at least one company QD Vision / Color IQ is trying to add to display screens. Traditionally white LEDs were YAG:Ce phosphors combined with nitride-based blue LEDs, but its not usually the (blue light -> Phosphor ->white light) reaction that's limiting, its the initial (electricity -> LED-> blue light) process that's currently ~15-20% efficient, I believe.
Check this hybrid QD/LED (device B) emission spectra:
the current QD/LEDs are being sold for having a much broader color gamut, lots of IR and a higher CRI, not really whats needed in Ag stuff.
I've known about these for quite a while. Surprisingly there is a grow light that actually uses quantum dot tech now. We have been in a side by side with them for about 6 months now. So far the cxb3590 3500k @ 1.4a has about the same yield as 80 watts of this quantum dot grow light. Also to consider is will this tech still suffer from my proposed "bottleneck"
Its true this tech will get better over the next few years as Samsung's newest tech uses quantum dot. Also I think Sony and some others are going in that direction as well.
Stephenj37826
Well-Known Member
I meant white leds based on a uv diode instead of a royal blue diode.
Stephenj37826
Well-Known Member
How efficient is UV currently.
Stephenj37826
Well-Known Member
Cree reports that the LED efficacy was measured at 303 lumens per watt, at a correlated color temperature of 5150 K and 350 mA. Standard room temperature was used to achieve the results.
I wonder at what CRI. I'd love to see the SPD of this test light.......
@PurpleBuz I feel the trade off of an even bigger bottleneck would negate any possible gains made using UV/Phosphor
As a matter of fact as the higher nm wavelength technologies advance we will see the ability to make more photons to "convert" with phosphorous per watt.
Bill Lidgate
Active Member
The durability of the QDs (where I've seen data) hasn't been up to the std phosphors, they quote some 2000h lifetime at 110C but at reg temps they have been inferior. Maybe this has changed but it was of note when I saw it. QDs were "the next big thing" about year 2000 but many of the companies folded, they seem to have followed the "Gartner Hype Curve" but may be out of the trough
FranJan
Well-Known Member
Where you been bro? Actually IDK if these would be worth it for our thing but you could talk to $ORRA about something custom. Gotta love the beam angles on these. GaN on GaN in da house! Actually West NY in da house!*
http://www.soraa.com/products/23-PAR38
https://res.cloudinary.com/soraa/raw/upload/v1460478841/resources/specifications/par38-120v.zip
*Now stop whining about no jobs, stop trying to kill our water and pay some fucking taxes!
Last edited:
Bill Lidgate
Active Member
I haven't seen Cree's data but there is an argument that
"the most perfectly engineered light that we would perceive as “white” cannot achieve much more than about 250 lm/W"
see: http://physics.ucsd.edu/~tmurphy/papers/lumens-per-watt.pdf
Since the silly lumen scale is centered in the green, this means we could get 683 lm/W using monochromatic light at 555 nm, but that's a real exercise in futility for Ag applications. Historically LED Mfg have not been above padding their lm/W numbers using lots of tricks like peak current pulsed emissions at very narrow bandwidths (sort of like the digital power amps which claim to put out more watts that they take in!) And they also use the best "bins" possible, meaning near perfectly matched individual LEDs on COBs that you can never actually get without picking through a few hundred thousand units...
"the most perfectly engineered light that we would perceive as “white” cannot achieve much more than about 250 lm/W"
see: http://physics.ucsd.edu/~tmurphy/papers/lumens-per-watt.pdf
Since the silly lumen scale is centered in the green, this means we could get 683 lm/W using monochromatic light at 555 nm, but that's a real exercise in futility for Ag applications. Historically LED Mfg have not been above padding their lm/W numbers using lots of tricks like peak current pulsed emissions at very narrow bandwidths (sort of like the digital power amps which claim to put out more watts that they take in!) And they also use the best "bins" possible, meaning near perfectly matched individual LEDs on COBs that you can never actually get without picking through a few hundred thousand units...
Last edited:
