High CRI leds, comparing R9 values, XP-G3 N219c N218b XM-L2 XP-L2

Wow, that’s like the Hitler of flashlights…

Jon_slider, thanks for those very useful pics. I really like the magnetic rotary brightness control on the Jetbeam RRT-01. If only it has USB recharge, it would be the perfect gift light for muggles (after a HI CRI swap ofcourse).

The 4500k 219B on the right seems a bit too magenta instead of pink. the 4500k 219B in the middle seems fine. Why is that?

I think I’m going to use the E21A 2000k while watching TV. It’ll allow me to see my surrounding while not enough to distract me from the screen.

my solution to USB charging is to buy 16340 cells like Olight and Utorch, and Klarus, are including in their S1 Mini and Mi1c. The USB rechargeable 16340 is also available with a Fenix label:

The RRT-01 is actually also 18350 compatible, though those dont come w USB, I have chargers for that option.

I do find the sw45 are very “violet”, not just pink. At first I did not like them, but when Im adapted to daylight, I dont notice the violet much.

Comparing the 219b SW40 9080 and the 219b SW45 9080, I find the sw40 slightly more yellow than I like… but Im very nit picky at this level of detail. The sw45 is bluer and pinker, but also cooler CCT, and there are times when I like cooler CCT

atm my use of sw40 has decreased, and my use of sw45 has increased.

Since I already have a Utorch S1 Mini w 2000k E21A, and another Utorch S1 Mini w N219b sw45, Im pondering what LED to opt for in the #3 Utorch being modded by Clemence. Im thinking I will not go with sw40, as originally intended, Im becoming more interested in cooler High CRI options. I may go with another sw45, am also considering a 5000k Optisolis, based on the very balanced and high levels of the saturated colors, R9, R10, R11, R12 and R13 on the 3000k optisolis posted above. That decision will depend on Clemence being willing, and arrival of an LED board for the Optisolis. I dont know if the Voltages of the Optisolis is a good match for the Utorch… the quad E21A did not work out because the voltage was too low, and the Utorch driver was increasing the current, and decreasing runtime, so we went to a single E21A.

For your Armytek I dont think there will be an issue with using a quad E21A, I suspect that is where your Clemence mod is headed, and I trust his judgement.

I have a couple other warm lights, 3000k N219c versions. Though I like them in full darkness, and on my nightstand, I dont think I need two more warm lights, Im finding I use the SW45 more often than my 3000k options. Im still really excited to get the E21A 2000k. It color is much more orange, in a way I like, than the 219c 3000k, that I find overly yellow (which I improve with a Lee 1/4 minus green filter).

fun times in the LED selection world, all thanks to Clemence

"And a couple Zebras.. you can see they are very low in R9, which imo is the reason there is so much “tint lottery”. Its not really lottery, its just that the High CRI Zebras have very low R9.. Not a lot of red, so the green and yellow dominate their tint."

I can't agree with you, because new mkIV Fc and Fd Zebralight's with XHP50.2 has much better color rendering,

that old models,

foe example H600Fc mkIV

I has H600Fc IV and Sc600Fc IV, both has good color rendering.

I say about significant difference between old Zebras (H600w, H600w II, H600Fc III) and new mk IV - new are much better.

Tint ?

Both Zebralight's has nice tint (very slight green in Sc600), these Cree XHP50.2 (imho) better, that Nichia 219b (9050 grade) in my 2 old flashlights, better , that Nichia 219c in my old flashlights.

(I can't compare with Nichia 219b (9080 grade), because I dont' have it.)

Compare with my Osram 's flashlights (CQAR.CC) 4000K : OSRAM's has slightly better color rendering and tint.

Hi Gang,

I was just wondering if anyone here has ever tested to see if using optics like the common 3x carclo type deteriorates CRI? The main reason I ask is I’ve been using some of the 90 CRI optisolis emitters in a tripple and noticed a very nasty shift in color when using them. It turns my 5000k color temp into like 6500.

With a wide optic the color is smooth and even just cooler than it should be. With a narrow optic I get three blue dots surrounded by some varous warmer colors. When I mule it the beam us not a great pattern but the color is about what you’d expect. A nice white with a mild tan overlay.

The Optisolis is special and does not work with smooth optics. Most single-die LEDs should work in clear optics if they fit. The current Cree LEDs will produce an ugly green corona or spot, just like with smo reflectors.

The optics don’t reduce CRI. They may have some effect on tint just like AR lenses. Tint shift is another matter altogether and just like The_Driver said, Optisolis needs a diffuse optic to prevent that nasty shift. Otherwise the hotspot is very cool, while the edges turn yellowish green and warm.

I didn’t happen to have a frosted Carclo 10623/10624, but I sanded a Carclo 10622 and while the results were good, the edges of the beam still had visibly warmer temperature to them. After adding d-c-fix the beam turned perfect. Add a mild minus green for that rosy goodness.

With d-c-fix:

Thank you guys for the input.
.
.
Maukka, just to make sure I understand correctly. The test your showing says 4974k temp, that is with the optic in place correct? So I assume this is the 5000k temp rated version of the LED and your saying proper diffusion will eliminate the tint shift all together?

Is this the version of the product d-c-fix that I should order?
https://www.amazon.com/d-c-fix-Self-Adhesive-Glossy-White-346-0011/dp/B002TDOSNG/ref=sr_1_9?s=hi&ie=UTF8&qid=1534940011&sr=1-9&keywords=dc+fix

Exactly. Optic + d-c-fix.

This is the correct d-c-fix:
https://www.amazon.com/d-c-fix-346-0211-Decorative-Self-Adhesive-Window/dp/B007B9PNOQ

Thank you!!!

What gradation of sand paper you used?

He probably went in first with 400 wet grit, then 1000 wet grit, then 2000 wet grit.

P1200. Wet first, dry second.

Not too much. I tried 1000, 1500, 2000, 5000 and 7000 and I think the best is 2000. All wet.

So I finally got to try the Oslon Square 2700k 96CRI 6U tint bin and 3500k 92CRI 5Q tint bin. The 2700k is the nicest warm white tint I have ever seen and R9 is even higher than the E21A and SST-20 95CRI when I shine it on my wood texture table. For reference, I have tried a number of warm white emitters, Cree XM-L2 90CRI, XM-L2 7A 70CRI, XP-E HEW 7A2 90CRI, XP-G2 HE 90CRI S2 7A1, XP-L2 V4 7A 70CRI, SST-20 95CRI HD2, Nichia 219B 9050, Nichia E21A 3000k, Yuji, and a few others. The only one I haven’t tried yet is the 219B 9080 3000k but I already have some that I will try in the future. But until then, the Oslon Square 96CRI 2700k is the BEST warm white I’ve ever seen.

The 3500k 92CRI 5Q tint bin looks great too and possibly the best 3500k tint in existence unless compared with the mixed CCT emitters. My Tiara Pro from Clemence with quad E21A 2x 3000k with 2x 4000k still looks the best for 3500k. But it’s hard to beat mixed CCT especially if the E21A tint was so good to begin with. R9 of the 92CRI Oslon is noticeably not as good as the 96CRI Oslon. It is also worst than the SST-20 95CRI and E21A. Even the 219B 3000k 9050 has better R9 than the Oslon 3500k. But that is not to say it is bad. Color rendering and R9 seems to edge the 219C 4000k 90CRI by a bit.

Are there any Cool white 6000K and up LEDs you guys know of (besides low power optisolis etc) that have a fairly high CRI rating? 80+?

E21A

The Luxeon MZ is available with 5700K and 90CRI.

Cree is claiming to have added 95 CRI value to XPL2 and XHP50.2 (in their newsletter). I would like to see how these stack up.

A clip from the news letter:

New! TLCI 90 & 95 Options Added to XP-L2 & XHP50.2
If you develop sport or entertainment-focused lighting products you no longer need to choose between illumination optimization for eyes or cameras. You can have them both with Cree’s enhancements to our leading high power products. These new 5700K CCT options give you.

Best possible illumination for eyes and cameras
Reduced energy utilization for optimal illumination for broadcast videography
TLCI 90 and 95 ratings ensure minimal video-chain color processing and compensation for broadcast video.

Interesting, but I find myself somewhat hesitating. Until now high CRI Cree led emitters have been found less desirable versus others concerning rendering quality and tint, hope that is to change and get better. I used this opportunity to read about TLCI:

TLCI vs. CRI, CQS, etc.: How do they stack up? @ ProVideo Coalition

It's about how cameras “see” stuff.