Nichia introduces violet pump based ultra high cri mid power leds

anuragwap

Newly Enlightened
Joined
Sep 29, 2011
Messages
115
Location
India
They finally made the switch! (http://www.nichia.co.jp/en/product/led_sp_optisolis.html) Maybe Prof. Nakamura at Soraa was right to say that violet pump leds will eventually take over blue pump ones. Their new series is named Optisolis. From the datasheet, it has a peak at 420nm, does not suffer from cyan gap, comes in neutral and warm, has respectable 120+lm/W efficacy. I guess exciting times are ahead for high CRI snobs, very soon we may see high power Nichia leds that have better color rendering than R9080 219Bs and yet higher efficacy than Yuji leds.
 

iamlucky13

Flashlight Enthusiast
Joined
Oct 11, 2016
Messages
1,139
Bah! Look at that measly R9 score of 94! :thumbsup:

I'm glad to see Nichia taking a venture in this direction.

Keep expectations reasonable though. Blue-pumped LED's already can achieve excellent performance in the red range, which is where white LED's have traditionally been most lacking. The violet-pumped LED's main place to offer an advantage is in the small range between the ~450nm peak of blue LED's, and the roughly 400nm peak of the violet LED's they're using.

I think the subjects where these blue-violet shades are critical are less common than the red tones that are needed to make the human complexion look healthy, see the undertones of woodwork, etc. But they're not absent, of course - artwork, paints, and flowers are subjects that should benefit from better blue-violet rendering.

Also, the way the color scores are computed, I expect the newer, but less commonly used CQS scores to show larger improvements for violet-pumped LED's than CRI scores.
 

ssanasisredna

Enlightened
Joined
Oct 19, 2016
Messages
457
Bah! Look at that measly R9 score of 94! :thumbsup:

I'm glad to see Nichia taking a venture in this direction.

Keep expectations reasonable though. Blue-pumped LED's already can achieve excellent performance in the red range, which is where white LED's have traditionally been most lacking. The violet-pumped LED's main place to offer an advantage is in the small range between the ~450nm peak of blue LED's, and the roughly 400nm peak of the violet LED's they're using.

I think the subjects where these blue-violet shades are critical are less common than the red tones that are needed to make the human complexion look healthy, see the undertones of woodwork, etc. But they're not absent, of course - artwork, paints, and flowers are subjects that should benefit from better blue-violet rendering.

Also, the way the color scores are computed, I expect the newer, but less commonly used CQS scores to show larger improvements for violet-pumped LED's than CRI scores.

Why do you expect CQS to show a larger improvement? Compared to traditional sources, good LEDs can exceed traditional lighting scores. I would expect only marginal benefits for violent pumped emitters.

I don't see this ever displacing blue pumped, 455 versus 400nm is always going to have an efficiency advantage due to Stokes losses in the phosphor conversion. Perhaps in the home it will achieve high penetration, at least in N.A. and Europe where low CCT is popular in the home. We have the ability to have very high CRI fluorescent but the vast vast majority isn't due to lower efficiency.
 

iamlucky13

Flashlight Enthusiast
Joined
Oct 11, 2016
Messages
1,139
Why do you expect CQS to show a larger improvement? Compared to traditional sources, good LEDs can exceed traditional lighting scores. I would expect only marginal benefits for violent pumped emitters.

I don't see this ever displacing blue pumped, 455 versus 400nm is always going to have an efficiency advantage due to Stokes losses in the phosphor conversion. Perhaps in the home it will achieve high penetration, at least in N.A. and Europe where low CCT is popular in the home. We have the ability to have very high CRI fluorescent but the vast vast majority isn't due to lower efficiency.

Due to the way CQS is measuered vs. CRI - it is more critical of the blue and violet range, as I understand it. I should actually go check Maukka's data to confirm this. I'd expect his measurements will generally show slightly lower CQS numbers than CRI.

I agree about the blue pumped advantage.
 

JoakimFlorence

Newly Enlightened
Joined
Jun 4, 2016
Messages
137
Bah! Look at that measly R9 score of 94!

Keep expectations reasonable though. Blue-pumped LED's already can achieve excellent performance in the red range, which is where white LED's have traditionally been most lacking.
Just my opinion on this, I think in general the violet-emitter white LEDs put out by major brand name companies are going to have only moderately high R9 red values, when it comes to CRI. The reason is that violet-emitter already achieves exceptionally high R6 and R12 (blue) values, therefore when it comes to R9 (deep saturated red) it's not seen as so absolutely critical to achieving overall high CRI values. By making sacrifices in the R9 value it's possible to achieve substantial increase in overall luminous efficiency, more than enough to offset the slight decrease in efficiency from using a violet-emitter.


I think the subjects where these blue-violet shades are critical are less common than the red tones that are needed to make the human complexion look healthy, see the undertones of woodwork, etc.

Actually human skin tones can look just fine with only moderately high R9 values, 80-90. Skin tones actually have a more orange-red hue, so as long as the light strongly illuminates orange-red colors, it doesn't matter so much whether the red lacks deep saturation. We're talking about typically 92-94 CRI. The only downside is that deep ruby colors may appear somewhat coral colored. They're brightly illuminated and lively but just not the most perfect deep red color hue saturation. This isn't going to be the most flattering to crimson red colors.

You are right, in most applications there is going to be no need for violet-emitter white LEDs.


The violet-pumped LED's main place to offer an advantage is in the small range between the ~450nm peak of blue LED's, and the roughly 400nm peak of the violet LED's they're using.
You know, it is theoretically possible to achieve a more even distribution of blue wavelengths without using the violet-emitter approach. (Like say with better blue-green phosphors to make sure there's less of a dip in the 470-500nm territory, or designing in separate emitters to fill in the gap) If violet emitter LED lighting started popping up everywhere I'm thinking all that violet light might not be the best for us. Light starts getting more reactive and can induce more chemical changes the shorter the wavelength.

I've played around with some theoretical CRI software and what I've found is that if you want to maintain excellent R12 values with a dominant blue spike, it's important that a gap be maintained between the 450nm and around 480nm or so (no precise values here) but it doesn't matter much after that. What I mean is that adding in a 500nm (cyan-green) spike is barely going to have any effect on the R12. If, on the other hand, you add even a small 470nm spike, you better add some shorter wavelengths on the other side of the 450nm (normal blue LED spike) to balance it out, otherwise the R12 is going to get thrown off. Even adding a small amount of 430nm will have a big positive effect on the R12. What I mean is that if you're not using the right blue wavelengths, even adding in a very small amount of the ones you don't have will have very substantial effects.

I'm imagining that in commercial retail applications they might choose to go 420nm instead of 430nm because of the significantly better effect on optical brighteners, even though it will have no difference on the R12. If we're talking about trying to minimize the shorter wavelengths, there's a bunch of trade-offs. (But nothing specifically wrong with violet LEDs if you're okay having metal discharge lamps everywhere)


I don't see this ever displacing blue pumped, 455 versus 400nm is always going to have an efficiency advantage due to Stokes losses in the phosphor conversion.
You know, it is potentially possible for a hybrid approach. Where a blue emitter would activate red and green phosphors while the violet emitter would activate just blue. I'm not sure how that would be combined into one LED dye though, might cause some manufacturing complications.


Maybe Prof. Nakamura at Soraa was right to say that violet pump leds will eventually take over blue pump ones.
Prof. Nakamura is working really hard trying to develop violet emitters that exceed the efficiency of blue emitters, so that his new violet-emitter white LEDs will be able to slightly exceed the efficiency of current generation blue-emitter white LEDs and overcome the inherent lower efficiency of converting violet into white light. Of course, any advances in the efficiency of violet LEDs would very likely carry over into blue LEDs also (since they are both based on the same GaN chemistry), so ultimately blue-emitter white LEDs will still be the most efficient in the end.

 
Last edited:

degarb

Flashlight Enthusiast
Joined
Oct 27, 2007
Messages
2,036
Location
Akron, Ohio
Due to the way CQS is measuered vs. CRI - it is more critical of the blue and violet range, as I understand it. I should actually go check Maukka's data to confirm this. I'd expect his measurements will generally show slightly lower CQS numbers than CRI.

I agree about the blue pumped advantage.

TM-30-15 was chosen as the next on over the cqs. https://www.energy.gov/eere/ssl/tm-30-frequently-asked-questions

Color discrimination, not color preference, is my life. However, some colors and artifacts are easier to discriminate at all kinds of crazy color temperatures, light levels, angles, and cri. I still haven't been convinced there is any one perfect light source, there are only optimal conditions with inherent blind spots. Tunable lighting... now, that interests me (once base levels of light are first met-limited by lpw, output, intensity, area covered, battery, and cost).

My issue hUUGE paranoia is: not being able to see certain bands of light.

So, this is very interesting for me. But, I need the massive cob and the highpower flashlight focusable led. I don't have much hope for Nichia fulfilling my needs based on their past offerings, unless I missed something. I think the 2008 1 watt purple tinted white Jupiter, was the most massive thing I have seen from them.
 

JoakimFlorence

Newly Enlightened
Joined
Jun 4, 2016
Messages
137
I think high CRI blue-emitter white LEDs are still going to have a lot of applications. They can go all the way up to 97 CRI and some of them aren't even that bad when it comes to rendering differences between different types of blue color hues.

It might be more than just merely about color rendering, however. Some picky and really exacting lighting experts have said they feel the light from violet-emitter white appears a little more natural. Though I have to say that the light from the very highest CRI blue-emitter white LEDs is not bad. ( 95 ?)

Violet-emitter white will always be a little bit inherently less efficient than blue-emitter white however.
 

PhotonWrangler

Flashaholic
Joined
Oct 19, 2003
Messages
14,432
Location
In a handbasket
I work in an office that is lit entirely with LED fixtures. I suspect that they are blue-pumped LEDs as I can see a slight blue spike when I look at them through the edge of my glasses. Skin tones seem subjectively ok to me, although I would like to compare this with a violet-pumped source.
 

JoakimFlorence

Newly Enlightened
Joined
Jun 4, 2016
Messages
137
I work in an office that is lit entirely with LED fixtures. I suspect that they are blue-pumped LEDs as I can see a slight blue spike when I look at them through the edge of my glasses. Skin tones seem subjectively ok to me, although I would like to compare this with a violet-pumped source.
Obviously the appearance of skin tones isn't going to depend whether it's pumped with blue or violet.

And if LED lights are being used in an office setting they are almost certainly going to be blue pumped.
 

JoakimFlorence

Newly Enlightened
Joined
Jun 4, 2016
Messages
137
Color discrimination, not color preference, is my life. However, some colors and artifacts are easier to discriminate at all kinds of crazy color temperatures, light levels, angles, and cri. I still haven't been convinced there is any one perfect light source, there are only optimal conditions with inherent blind spots. Tunable lighting... now, that interests me (once base levels of light are first met-limited by lpw, output, intensity, area covered, battery, and cost).

My issue hUUGE paranoia is: not being able to see certain bands of light.
I have a 91 CRI garage fixture I got because it was advertised as "high CRI". Although it makes skin tones look okay and warm colors don't appear dead, the problem with 90-92 CRI is it can make red colors look more of an orange-red. I had an orange colored extension cord and a red (scarlet red) tool cabinet, and under the light they both kind of looked like the same red orange color.
 

PhotonWrangler

Flashaholic
Joined
Oct 19, 2003
Messages
14,432
Location
In a handbasket
Obviously the appearance of skin tones isn't going to depend whether it's pumped with blue or violet.

I agree, however the pump wavelength could have an impact on the choice of phosphor blend, which could have an impact on skin tones. I seem to recall some early G-E literature that stated that shorter wavelengths offered a wider choice of downconverting phosphors, which in turn could offer more options for the red end of the spectrum.
 

ssanasisredna

Enlightened
Joined
Oct 19, 2016
Messages
457
I agree, however the pump wavelength could have an impact on the choice of phosphor blend, which could have an impact on skin tones. I seem to recall some early G-E literature that stated that shorter wavelengths offered a wider choice of downconverting phosphors, which in turn could offer more options for the red end of the spectrum.

I think it comes down to cost, and whether UV pumped can offer top-end (not general market) light sources at a better cost than current LED technology.

Xicato: http://www.xicato.com/technology/color-rendering is as good and arguably better than any halogen source given the improved gamut area and Bridgelux Decor Ultra is right up there at competitive prices.
 

degarb

Flashlight Enthusiast
Joined
Oct 27, 2007
Messages
2,036
Location
Akron, Ohio
I think it comes down to cost, and whether UV pumped can offer top-end (not general market) light sources at a better cost than current LED technology.

Xicato: http://www.xicato.com/technology/color-rendering is as good and arguably better than any halogen source given the improved gamut area and Bridgelux Decor Ultra is right up there at competitive prices.

RaChart_Updated%20August%202015.jpg


Now, that chart is the first meaningful discussion on color I have seen. With specific numbers like these, no opinion is needed. I know immediately what light will work for the application and what will not.
 

degarb

Flashlight Enthusiast
Joined
Oct 27, 2007
Messages
2,036
Location
Akron, Ohio
The xicato v95 looks so impressive, I may need one for experimental reasons. If battery operated flashlight with wide beam, good color matching tool, possibly.

The halogen, to be fair, is an ir coated bulb. Unless I am mistaken, this is the hir 39lpw technology ; I own one or 2 Lowell Tota hir, and will concede it's light is a hair green. This downside is made up for its more light, less heat and electric. Still, 39 lpw at 650W, pales compared with a 600W mh at 80,000 lumens of output. Probably, the higher gai, as well as as luminance, is reason why I would choose mh over incandescent any day, or night. Anyway, my point is that perhaps they are not being fair to halogen. (Though for halogen to be taken seriously, it needs to get its lpw number to a respectable number, when most light sources today are >100 lpw. Also, hir bulbs, I think, have short lifespans. I know Tota bulbs do not last long. Also, grew tired of popping in halogen replacement bulbs like candy on large jobs. Never liked incan household bulb lifespan either. I find deadbulbs emotionally draining, apparently.)
 
Last edited:

degarb

Flashlight Enthusiast
Joined
Oct 27, 2007
Messages
2,036
Location
Akron, Ohio
Soraa violet pumped bulbs are available https://finance.yahoo.com/news/soraa-healthy-violet-leds-want-002831038.html


My limited attention to the subject says Soraa was founded by Nichia's Nakamura who got the Nobel prize for inventing the high power blue in 92/93 (phosphor conversion of blue to white known well, used in color tv). I wonder, guess, Soraa is a totally separate company, based in California. And Nichia is simply copying Nakamura's violet pump?


I am sure I am probably off, based on my shallow digging. So, correct me, please.
 

degarb

Flashlight Enthusiast
Joined
Oct 27, 2007
Messages
2,036
Location
Akron, Ohio
I wonder which of these high cri emitters would have the best efficiency and focus.


Sure, the 219, but really want to leap frog the 219.

I am thinking a petite lipo tir- poorly heatsinked- headlamp with a narrow 15 to 20 center focus. Small amperage, maybe 150ma. Not as any main light, rather as a supplemental cone boosting light that can be worn under a real 8.5 hour straight 374 lumen light driven at 2 watt 2s18650. If fails to help the high output/intensity/efficient headlamps, still would have made a fair runtime, high cri camping, reading, color matching-doublecheck light. If nothing else, I could admire my healthy hand color of both my palms at one time, rather than just one at a time using my current 219 flashlight.

My other question is whether particular optics can adulterated color on various emitters. I assume the light at differing angles emitted from the led may have different cri values. This seems quite apparent to me in a xte dropin I bought years ago, and the Xp-l hi I put in a 31mm is reflector nearing 2 years ago. The color does not necessarily look uniform. And that is just the amount of color shift which is detectable with the human eye.
 
Last edited:

JoakimFlorence

Newly Enlightened
Joined
Jun 4, 2016
Messages
137
I wonder which of these high cri emitters would have the best efficiency and focus.
The efficiency is usually inversely related to the R9 value, since that's where most of the efficiency losses tend to be, trying to generate those deeper red wavelengths through phosphor.


The xicato v95 looks so impressive,
Looking at the spectral graphs from the specs, it looks like the difference between the Artist series and Vibrant series may be that the Vibrant series is using a slightly shorter wavelength blue emitter (looks like maybe 440-445nm but that's just from eyeballing it on the graphs which isn't very exact). It also looks like the Vibrant series has a tiny 470-475nm blue spike (from a separate emitter I wonder?) to help compensate for the blue saturation color shift of going with a shorter wavelength emitter.

(you can see the specs here and here )
 
Last edited:
Top