Greenish SST20

Is the SST20 with high CRI supposed to look greenish?

It definitely looks greenish in the Moonlight mode in a Gray Convoy S2+ with luminus SST20,DTP copper plate,ar-coated glass lens,7135 biscotti firmware

On the left Sofirn C01 High 94 CRI Yuji LED - 5600K
On the right Convoy S2+ SST20 - 5000K - with a Moonlight mode

As shown in maukka’s tests, yes. The emitter is green at low power levels. It doesn’t get white or rosy until it’s running pretty hot:

This is why I’ve mostly avoided SST-20. I almost always use low modes, and I don’t like green beams, so it’s not really suitable for me. I use emitters which are white or rosy at low power instead.

The 4000k is indeed a bit green. 3000k is far better if you don’t want green and use low modes mostly, but can be to warm for some.

A Lee minus green filter might help you.

@sochi111, only the SST-20 4000k and below are 95CRI.

The SST-20 5000k and up only have 65CRI.

That means the green tint will be exaggerated by a lot compared to the 95CRI SST-20.

Any recommendations on high CRI LEDs that have good tint in low modes?

Pretty much this. With current industry practices expect very little red in a 65 CRI LED. Without the R9, it’s a imbalanced and distorted light. One could also argue that insufficient R12, saturated blue, is another failing point of low CRI emitters. Currently the 70 CRI emitters that are tested typically have R9 values in the –20 to –33 range. R9 is how well a light source reproduces red, and a light source with values that low renders reds more like pale shades of brown/gray. 80 CRI emitters seem to typically be a bit better with demonstrated R9 values usually in the teens with competent 90 CRI emitters pushing the 50’s and finally the 95 CRI tint queens pushing into the 90’s for R9. Or at least that the general pattern I have observed with some execptions.

If you want good color rendering and good tint go for the 2700K-3000K 95 CRI range of the SST-20. My current experience with these emitters limits my current recommendations to these comparatively warm CCT, but they are quality emitters, and I eventually plan on getting a 3500K SST-20 for tests. We shall see how that goes.

Mixed Nichia E21A’s have a good tint on lower modes. Maukka’s measurements.

The only rosy WW emitters on low mode are mixed tint E21A, Oslon Square 2700k emitters if you have the right bins, and the XP-L2 V4 7D emitter from KD. Mixed tint E21A are the rosiest though and the best looking.

Do you think mixing 3000K and 4000K SST20’s in equal ratio’s will yield a noticeably lower DUV value than using either on their own?

The E21A is an excellent emitter, by itself. Looks like R9 values are actually worse with the tint mixing. I think this fascination for a below the black body tint is a bizzare perculairity, even at the expense of objective light quality, a preference of certain flashlight enthusiasts. R9 is worse and light color saturation, R (g), is not significantly higher then just the a single emitter. Give me single CCT light source as close to or on the BBL.

In my case, its not a fascination of below BBL, although I do prefer below instead of above. I just really dislike the green tint when its above.
I have a Nichia Tiara from Clemence and tint is perfect. I’m looking to put some xp-g2 replacements in an existing light and want to avoid green tint at low outputs. E21a unfortunately is not an option for this light. Has to be XP-G2 / XP-L footprint.

Thank you for explaining it. Now I understand the difference.

I should’ve gotten a 2700K, or FA1 FA4 bins
Should Simon make it more clear that the CRI95 doesn’t apply to the 5000k ?

But I agree, as you guys mentioned the high modes are not as bad.

I will start using it outdoors, maybe hook it up to the head strap, I actually like the narrow beam when biking at night.

That would certainly be helpful! I don’t really know how many people actually pull up the spec. sheets for individual emitters when flashlight shopping.

From what I’ve been hearing,so far, with the 2700K range other have, is that they are happy with the light qualities. I have that emitter myself in a quad, and the light is indeed good quality.

For what it’s worth, Kaidomain has the 3000K SST-20 with a below the black body tint. I have that emitter and its 3000K brethren from Mouser in another light, and I don’t see either one being noticeably greener or rosier.

Our perception of light is subjective, so the objective measurements might be treated as ideal from a theoretical standpoint, but still not be what naturally appeals most to a given person. And that subjective preference certainly does differ between users, so it’s completely fair of you to prefer on the BBL while another person prefers below BBL.

As for R9 values, if you had a 20 point difference in R9 values that are otherwise very good (above 70, for example), or a difference in DUV of 0.003, I’m pretty sure the DUV difference is a lot more noticeable.

Speaking for myself, I find CCT the most noticeable difference of all, so in most cases, my choice will probably be prioritized as CCT, then DUV, then R9, assuming typical values of these parameters. So if I want a 4000K light for a given application, but it’s DUV +0.003, and the R9 is 60, I’d take that versus a 3000K light where the DUV is 0 and the R9 is 80. I don’t approach that as a strtict, rule, however.

Different people have different preferences. For me, I’d rather have a rosy 70CRI emitter than a green 95CRI emitter. I think XP-L HI 5A looks much better than SST-20 4000K 95CRI.

Nichia 219B is also nice. I really like my 219B lights.

I’d like to try E21A too, but haven’t yet.

Same, I like hi CRI, but I really hate green in my lights. I just did my first mod with E21A and it’s really nice.

I recall your favourite CCT is somewhere between 4500K-5000K. So you could either go for non mixed 4500K/5000K, or mixed (2+2) 4750K or even 3+1 mixes with 4625K and 4875K, the possibilities are endless.

Btw the brighter D240 flux bin 4000K is above the BBL, the regular D220 bin should not, SKV98 described it as similar to the SW40. 1000K mixes are guaranteed below BBL and noticeably rosy. Clemence also managed to “copy” the SW45 but I can’t remember in what led configuration.

Toykeeper, would you describe the tint of the SW45 as rosy or rather as violet?

The E21A is a astonishing emitter in quad setup, but it’s pricey, it seems to work only with wide TIRs and the number of host is a bit limited due to ~9A max current. If only one could persuade Armytek to sell hosts or at least to use blue locklite instead of the red one.

Not violet. The camera thinks my 219B emitters are rosy, but to me it just looks pure white.

Here is a FF ROT66 219B (on the left) next to a LH351D 5000K (on the right), with the camera’s white balance set to its built-in 5000K “daylight” setting:

Here’s that same photo, but processed to show tint more clearly. Basically, blur the photo and then normalize the brightness of each pixel individually as far as possible without distorting the hue:

The only time I’ve seen emitters look violet was with XP-G2 3D. It’s a weird effect, especially since XP-L HI
3D doesn’t do that.

Funny how everybody sees it different. For me negative duv at 4500K looks violet and the light itself as a shade of grey, that’s why I stick to 3000-4000K. But if you like noticeable negative duv a 500K or 1000K difference mix could work out. Tintshots including mixes: [Reference] Nichia E17A/E21A (2000K - 6500K, R9050/R9080, color) CCT and tint shots

Is the SST40 green as well at lower levels?