Luminus SST-20-W 4000 K CRI95 color and output test

[maukka]

I got this emitter from Kaidomain on a 20 mm MCPCB.

Mounted on a CPU heatsink with a nice splash of thermal paste and over the top 16 AWG wires.

Purchase link: http://kaidomain.com/Luminus-SST-20-J4-FB4-Neutral-White-4000K-LED-Emitter?search=Luminus%20SST-20

Data sheet.

Exact order and bin code: SST-20-W40H-A120-J4FB4 VH

Where
SST-20=self explanatory
W=white LED
40=4000 K
H=CRI 95
A=solder pad configuration (3.25 x 3.25 mm)
120=viewing angle, 120°
J4=flux bin, 107-114 lm at 350 mA at 85°C, >118 lm at 350 mA at 25°C
FB4=tint bin, 4000 K above BBL
VH=forward voltage bin, 2.5-2.7 V at 350 mA at 85°C

CRI data measured from the hotspot with a

Bare emitter

Yajiamei 10° beaded optics at 1000 mA

Small op reflector from the Reylight Pineapple

Medium sized smooth reflector from the Eagle Eye X7

CRI data with increasing drive current with the Yajiamei optics measured from the hotspot

Tint shift with Yajiamei 10° beaded optics with increasing drive current

Tint shift is quite linear all the way to the max output current.

Output measured in a 50 cm integrating sphere

At 350 mA I measured 121.6 lumens at 126.5 lm/W.

Emitter died shortly after being run at 7 A. I can’t see anything visibly wrong with the dead emitter.

[maukka]

Sadly I wasn’t able to do any real world testing inside a flashlight, since the emitter broke during the output test. What little I observed during testing seemed very good. The tint is at least as good as with most Nichia 219Cs I’ve tested. With a smooth reflector the shift is even a bit less severe in the corona.

[djozz]

Thanks for the test. The led performs nicely up to specifications. I’m sure that many people will complain that the tint is above the BBL but I care a bit less about that.

[contactcr]

so what niche does this LED fulfill that others do not? maybe good on AA sized/keychain lights or high CRI triples? Not really sure

[maukka]

1000 very high CRI lumens from a single small die emitter with small tint shift is pretty good in my book.

[adam7027]

Thank you for this thorough measurement :+1:

There seems to be a noticeable drop in the cyan and cobalt blue area of the spectrum - but the high R12 value is still very impressive - and R9 is nearly at Nichia E21A level.

How this LED compares to Nichia 219C and Samsung LH351D high CRI offerings with the same CCT in regards of tint shift, output, and Vf?

Do all of these Yajiamei optics consistently push CCT into warmer direction, and a bit above BBL (this latter is no problem for me)?

I was very happy for a brief moment, when I read your bare emitter CCT measurement (43…K), so I wonder, if there are TIR optics available, which don’t - or just marginally alter CCT.

[BlueSwordM]

Yep. It’s very efficient. About on par as other small die 4000k 90+CRI LEDs, with a lower forward voltage.

This would make a very nice LED array in a 6A LD-A4 Sofirn C8F build. It would be gorgeous.

[adam7027]

@maukka

1000 very high CRI lumens from a single small die emitter with small tint shift is pretty good in my book.

Back a few months, I ‘snapped’ two C8F hosts for a very good price, but I was hesitating to put Nichia 219C in those hosts because the rather low Vf - I guess, this LED will be a good choice for triplets or quads.

[djozz]

@adam7027

@maukka

1000 very high CRI lumens from a single small die emitter with small tint shift is pretty good in my book.

Back a few months, I ‘snapped’ two C8F hosts for a very good price, but I was hesitating to put Nichia 219C in those hosts because the rather low Vf - I guess, this LED will be a good choice for triplets or quads.

The difference is not great but this SST-20 starts getting a slightly higher voltage than the 219C, when going over 3A, so yes it will draw a bit less current and heat in high current direct drive triples or quads.

[The_Driver1 (The_Driver)]

Thanks for the test!
It’s certainly an interesting LED. The light quality is very high and the efficiency thus low. It’s advantages over other LEDs with high-light quality are the XP solder pad layout, the single-die design (compared to the Luxeon MZ) and that it tolerates high currents (compared to the Nichia E21A and 219B-V1).

[Jerommel]

So…. where to get a 5k low cri version to dedome….?
Could that be a XP-G2 killer?

[djozz]

@Jerommel

So…. where to get a 5k low cri version to dedome….?
Could that be a XP-G2 killer?

http://kaidomain.com/Flashlight-DIY-and-Tools/Luminus-SST-20-L3-DA-Neutral-White-5000K-LED-Emitter?sort=p.date_available&order=DESC

[djozz]

The dedoming will be the challenge, no leds dedome so nicely as the old-school Cree’s

[Jerommel]

Thanks Djozz.
I’ve been away from BLF for some months, so i have some catching up to do.

Yeah, i bought 1 Liter of 99.5% Toluene for those SST’s, still a challenge though, but i’ll get some anyway.
I’ve still got a B158 waiting for a high intensity LED, for example.

I’ll try some of these high CRI ones too.

[X31 (-X3-)]

@djozz

The dedoming will be the challenge, no leds dedome so nicely as the old-school Cree’s

Slicing is the way to go with SST…
Now I’m impatient to get mines, they will go straight in the D4 :-p

[MRsDNF]

Amazing testing maukka. Thanks.

[Sirius9]

Thanks for this test :+1:

[DavidEF]

@maukka

Emitter died shortly after being run at 7 A. I can’t see anything visibly wrong with the dead emitter.

Does nobody else see the burned place under the top bond wire? Is that just a slightly darker connection area? Was it that dark before the test?

[maukka]

There’s the before image at the top of the OP.

[SKV89]

Thanks Maukka, Im always interested to see your reviews. This one would have been the perfect 219b replacement if the tint bin was below tbe bbl.