LED test / review – Luminus SST-12-WxS Specialty White (≈ 5000 K, typ. 65 CRI)

Flashlight Modding and DIY Parts
1

LED test / review

EN

Luminus SST-12-WxS Specialty White


SST-12-W50S-A120-H3502


EDIT 2023-02-07 — I dedomed the SST-12, you can find the results for flux and luminance incl. color measurement in this post!


LEDs with round LED chips seem to become more and more interesting. After “Yinding” launched an LED with a round luminous surface in the XM footprint (test here), Luminus now also offers an LED with a round luminous surface, although it is smaller than from the “Yinding” one. The round luminous surface should significantly improve the behavior in optics and increase the luminance.

This test will clarify how well this LED performs compared to competitors.

Technical data


Tj 85 °C / 700 mA

Type: single die, domed
Binning: min. H3 (min. 267 / max. 285 lm)
Color kit: 502 (DA, DB, DC, DD ≈ 5000 K)
Rated voltage: 2.5 to 3.1 V
Max. forward current: 1,800 mA
Max. peak current: 2,400 mA (130 °C Tj)
Viewing angle: typ. 120°
Thermal resistance: typ. 4.9 °C/W
Max. Temperatur Tj: 150 °C

Official datasheet here (Luminus, PDF)

First appearance


The round luminous surface, which sits on a gray substrate with a white silicone coating, is particularly striking at first glance. The dome is made of clear silicone, the white coating around the luminous surface seems to be an extra layer that is cast in an extra work step before the dome is applied.

It is interesting that the LED chip used here is connected with bonding wires, which is no longer implemented for many LEDs today. In this case, this was probably chosen because of the round light-emitting surface and the fact that the light-emitting surface is not shaded anyway.


This LED is otherwise similar in design to the XP-E2. It also becomes apparent relatively quickly here that the XP-E2 is the direct competitor, as the operating parameters and electrical properties are very similar.

Due to the symmetrical design, centering rings made by lathe can be used without any problems. In general, accessories for XP LEDs (3535) are fully compatible.


The footprint corresponds to the well-known XP format. The thermal pad is electrically isolated so that DTP boards of this format can be used without any problems.

LES



A special feature is the round luminous surface. This additionally sets this LED apart from existing models. Round luminous surfaces are known so far rather from large COB LEDs, in which several small luminous surfaces are arranged in a circular area. In the SST-12-W, the round luminous surface shines absolutely evenly. Remarkable here is the very sharp delineation of luminous area, which indicates the use of a chip of the classic lateral design - today’s LED chips usually also emit light at the sides to increase the overall efficiency, but this often causes problems with color fringes in reflectors (see XHP70.2).

The round light emitting area is 2.64 mm² in size. Due to the round shape, no light is lost due to shading at the wires despite the use of two bonding wires.

Power and overcurrent capabilities



Official parameters:

  • at 1,800 / 2,400 mA (official maximum current 150 / 130 °C Tj): 601 lm @ 3.20 V / 709 lm @ 3.33 V
  • Power at official maximum: 5.76 / 7.99 W
  • Efficiency at 1,800 / 2,400 mA: 104.3 / 88.7 lm/W

The official maximum currents are already very high. In contrast to many other LEDs, where the maximum current is specified rather conservatively, the performance increase due to overcurrent is unusually low. Possibly this is for marketing reasons, but I don’t know exactly.

The Vf is in the normal range, it is not extremely low as with some Lumileds emitters, but also not as high as with XP-E2 or XP-G2 (old version).

Overcurrent:

  • Maximum reached at 3.20 A, at this point 771 lm @ 3.49 V
  • Power at maximum 11.17 W
  • Sweet spot at about 2 A (642 lm @ 3.24 V)
  • Power in the sweet spot 6.48 W
  • Efficiency at maximum 69.0 lm/W
  • Efficiency in sweet spot 99.1 lm/W

I do not recommend overcurrenting. The power increase is very small. Interestingly, even operating at 2.4 A (official maximum current at 130 °C Tj) is not really recommended, since there are hardly any safety reserves left and at higher ambient temperatures the Tj of 130-150 °C could be significantly exceeded.

At 2-2.1 A, operation with a 6*7135 driver is possible without any problems.


The old XP-G2 S4 2B with the smaller light area as well as the Black Flat offer a significantly higher luminous flux and pull away mercilessly thanks to their considerably higher overcurrent capability. The Black Flat offers about the same Vf as the SST-12, but reaches a significantly higher maximum current of 5.8 A.

Luminance



The luminance is not very high. Interestingly, the luminance of the SST-12 is only minimally higher than that of the XP-E2 R4 despite a significantly higher luminous flux, and that only when the maximum possible current is used. The luminance stated here was confirmed both with the reflector measurement method and by means of direct measurement at 1m. The only minimally increased is mainly due to the larger luminous area than the XP-E2.

The SST-12 is not recommended for extreme throwers. Here I consider the XP-E2 (if it depends on higher luminance at low current and the higher Vf is little relevant) more useful. Alternatively, a CSLNM1/PM1.TG or Black Flat HWQP can be used, whereby the latter has no electrically neutral thermal pad and the conversion with this LED is therefore much more difficult.

Tint and light quality



The SST-12 tested here produces a greenish light, especially at lower operating currents.

A duv of 0.0062 @ 350 mA is relatively high, at 50 mA this is 0.0098. The color rendering is also not usable, an Ra of 67 and an R9 of -40 is very low, whereby this LED is basically only available in this CCT in the range of typ. 65-70 CRI.
There are no other surprises in the spectrum; it is typical for a neutral white LED with low color rendering.

However, at this point I would like to say a few words about Luminus newer products in general.

I have already tested several LEDs from this manufacturer, even though I have not published a separate test for each emitter. All these LEDs are visibly green, which is why I will not recommend a single emitter from Luminus when it comes to color and light quality. The source of supply doesn’t matter in this regard, as LEDs from Aliexpress, Kaidomain, as well as Mouser all have the same green tint, making this independent from seller or date code.

I don’t like this development. Low color rendering is one thing, but greenish tinted LEDs were and are basically not very popular, especially when a somewhat higher demand for light quality is essential. Few years ago, Cree showed with the MT-G2 how low CRI still looks good enough for flashlight use, even for today’s standards. The green tint also occurs in the variants with high color rendering (4000 K SST-20-H, for example), whereby the color saturation also suffers here - the Ra is nominally very high, but the color rendering felt is still not very good.

The Nichia 519A provide a minimally reddish light (we know this as ‘being rosy’), but this increases color saturation and subjectively looks very good. Why Luminus does not pay more attention to the tint and at least offers special binnings with negative duv is beyond me. Technically the emitters are without question very good, only the light color is not. Of course, there is the possibility that special color binnings are available, but as long as these do not reach the broad mass market and are available at wholesalers, I wonder why this area of good light color remains reserved almost exclusively for Nichia.

Light quality and use in optics


In reflectors, the light image is very good. Depending on the quality and calculation of the reflector, rings can appear around the spot, but the spot is still sharply defined. Spill is only slightly present, most of the light is concentrated in the spot. There is no color distortion in spot or spill.


The statements from the Yinding 5050 test also basically apply to the SST-12; due to the circular luminous surface, this LED is also well suited for use in zoom optics (lenses).

Conclusion


The SST-12-W is one of the first LEDs from a well-known manufacturer with a dome and circular luminous surface. The circular luminous surface provides an excellent light pattern in secondary optics, and the Vf is relatively low.

I like less the greenish light color and the low luminance, which makes LEDs like CSLNM1/PM1.TG or Black Flat HWQP and even the old XP-E2 seem more suitable for thrower applications at lower operating current. In addition, overcurrenting makes little sense, and the manufacturer’s specifications are already set very high here.

Pro

  • XP footprint
  • circular light area, therefore very good light image in secondary optics
  • reasonable price

Neutral

  • Luminance on the level of the XP-E2 R4
  • overcurrent is hardly worth

Contra

  • high thermal resistance
  • Poor color rendering and green light color

Thanks for reading the test! :slight_smile:

Many greetings, Dominik

11 Thanks
2

Thank you for the analysis.

I have several of these in different CCTs on the way. I have to say I’m a little less excited now.

Kaidomain provides the tint bin. Their 4000k is FD2, the 5000k is DA. I also ordered the new SFT40 5000k in the DA bin. Per the datasheets these should have a good chance of being on or below the BBL. Have you tried similar tint bins, or just the much broader color kit bins like the 502 here?

3

Thank you Koef for the great test on this emitter.

Would it be possible to slice or dedome one? I’m curious how it would perform (in tint and intensity) without the dome, the SST20-WxS had a marked improvement in intensity and tint when dedomed (5000k became more intense than SFT40).

1 Thank
4

Since I buy my emitters from Mouser or other wholesalers (if possible) to avoid of getting any counterfeit LED I bought the wider color binnings. I don’t think that KD will sell counterfeit emitters but I’m still a little skeptical tbh…

Yes I was able to dedome one without destroying the bonding wires. (not the best looking dedoming, but for luminance measurement it should be enough nonetheless)

I’ll post some results soon :slight_smile:

2 Thanks
5

I dedomed the SST-12 and measured luminous flux and luminance.




(Yes, my LED boards have a very good life indeed when used in my tests. :smiley: )

I pried off the dome in cold condition with a sharp knife. This is possible in principle, but there is a risk of damaging the bonding wires, and residues of the silicone and the white potting compound around the LED chip remain, which also simply doesn’t look as nice as it could be. Slicing seems to be the far better method here.

I don’t know to what extent chemical dedoming makes more sense, since I haven’t tested this for lack of suitable chemicals. It could imagine that this could be somewhat difficult since the white potting resin around the LED chip seems to be made of another material as the clear silicone dome.

Interestingly, the maximum possible operating current has slightly decreased after dedoming. Instead of 3.2 amps, this is now only 2.8 amps max., with 610 lm @ 3.41 V (63.9 lm/W). The Vf has not changed.





Luminance is about 40% higher, despite reduced luminous flux. Nevertheless, the XP-E2 offers a slightly higher luminance, with at the same time usually much more appealing tint. With the XP-E2 Torch with even higher binning which is available at wholesalers, even higher luminance could be possible with the good old XP-E2.





This is because the tint of the dedomed SST-12 is anything but beautiful, which was unfortunately already to be expected after the first color measurement. A pure green awaits the user of the dedomed SST-12, as similar as we already saw with the last tested “Yinding 5050” LED. This is not nice and not even tolerable for use in throwers. Although the color location moves more and more towards BBL with increasing operating current, the color quality is no longer usable at lower operating currents.

I recommend LEDs like the XP-E2 (dedomed), Black Flat HWQP or CSLNM/PM1.TG for throwers in the low/medium power range. Even some special emitters like the Osram SYNIOS DMLN31.SG could be suitable for some special applications and usecases. In addition to the higher luminance, these also offer the advantage of a relatively color cast-free light, albeit at colder CCTs.

3 Thanks
6

duv 0.0181

:face_vomiting:

Although some SST-20s 5000/6500 have equally vomit tint, the high CRI 2700 to 4000K will probably be much better.

7

Thanks so much @koef3 for sharing another one of your fantastically comprehensive tests!

1 Thank
8

I have a MF01 mini in copper, with SST-20-H 4000 K (typ 95 CRI), which are also ‘green’ (duv around 0,005). This might be not high, but on these warmer CCT this looks not so nice like the mixed reddish light from my S21F (519A 5700 + 2700 K with duv -0,0032), which is just perfect. The interesting thing is that especially at 4000 K and below the tint becomes even more important, since only small greenish tints could be annoying. For these CCTs a ‘rosy tint’ is mandatory, in my opinion, with a duv below 0.

But this is a subjective thing, since everyone rate the light quality different. I know some people which are happy with the ‘Luminus green’ and low CRI in newer Fenix lights…

It is good that some special color bins are available at KD, but as long as these more special color binnings do not arrive in the wholesale (where for mass production of flashlights the large numbers are sold), I think these offers are not very relevant, because here only modders and enthusiasts will buy these LEDs.

9

Green :nauseated_face: , looks like it suffers from the same as the newer XP-G2.

What’s the light emitting area like? It might hopefully be less green in the WxH high-cri version.

10

What do you mean? The LES is a classic LED chip without side radiation, and is really a round one, like you can see in the pic above. There is no glass or other layer of protection on the phosphor, so you have to be careful when handling the dedomed emitter.

If it is below duv, this could be great, but at least at wholesalers it is likely to get a green Luminus since they have only the wide color groups for sale (as usual for Luminus in wholesale). Tbh I don’t want to use LEDs above duv anymore, the 519A have the perfect tint with the right amount of red and of course great Ra / R9. Of course I could minus green or something like this, but if I have to fix the greenish problem with 3rd party components then I will prefer the use of an LED with great tint out-of-the-box.

2 Thanks
11

Sorry, I meant the actual phosphor area once dedomed. It should help giving an idea of the beam profile if dedomed, even if the tint isn’t great that migjt change.

Any chance to try the Nichia 219CT-V2 (ideally 4000 or 3000k)? It is a domeless 219C, currently available only in 80cri though, but could be promising for warm-cct throwy lights if the SFT40-WxH ends up having a bad usual tint or underperforming (719a :frowning: )

12

So you need a die shot with low current from top and certain viewing angle?

If you have one, of course :slight_smile:
I don’t have one.

13

Quick test of SST-12 E3 FD2 4000K from KD with a 17mm beaded TIR :

0.1A
Ambient = 3048.6 Lux, CCT = 3781K (Duv 0.0042)
Color Rendering Index (Ra) = 96.4 [ R9 = 83.3 ]
R1 = 97.2 R2 = 98.0 R3 = 98.5 R4 = 95.2 R5 = 95.8 R6 = 97.4 R7 = 95.9
R8 = 92.7 R9 = 83.3 R10 = 95.3 R11 = 96.8 R12 = 79.2 R13 = 97.4 R14 = 98.3
IES TM-30-15 Rf = 90.86 Rg = 96.85

0.5A :
Ambient = 10982.3 Lux, CCT = 3841K (Duv 0.0038)
Color Rendering Index (Ra) = 95.3 [ R9 = 77.3 ]
R1 = 95.3 R2 = 96.0 R3 = 96.2 R4 = 96.4 R5 = 94.7 R6 = 94.7 R7 = 97.7
R8 = 91.7 R9 = 77.3 R10 = 90.4 R11 = 96.7 R12 = 80.8 R13 = 95.3 R14 = 97.7
IES TM-30-15 Rf = 93.36 Rg = 98.70

2A :
Ambient = 29052.9 Lux, CCT = 3880K (Duv 0.0013)
Color Rendering Index (Ra) = 93.0 [ R9 = 68.5 ]
R1 = 93.4 R2 = 94.1 R3 = 94.1 R4 = 94.2 R5 = 93.0 R6 = 92.2 R7 = 95.0
R8 = 88.3 R9 = 68.5 R10 = 85.9 R11 = 94.6 R12 = 82.5 R13 = 93.3 R14 = 96.6
IES TM-30-15 Rf = 92.23 Rg = 100.47

FD2 supposed to be between -0.002 and +0.001 integrated at test current (350mA ?)

2 Thanks
14

Thanks for the testing, could be good for some cheap high CRI triples and quads.

15

Thank you for the measurement.
So we have also a (at least much less) green SST-12. The color rendition is okay.

And since flashlights are also operated at lower modes it is not really helpful that duv is getting good at higher currents.

Btw, I have a 719A on my rig now and this light is much more pleasant than everything I saw from Luminus to date (which is a shame since Luminus produces LEDs which are really good from the technical standpoint since they still use the classic LED chip design without side radiation which gives a really good light pattern with optics, but not from light color)

Since the price for these emitters is fairly low, this could be true (maybe with minus green)

1 Thank
16

The advantage of tipple/quad is that CCT can be mixed to bring down the duv, this works relatively well with SST-20 2700+4000K.

17

But only if CCTs are mixed. If mainly 2700 or 4000 K should be used, this advantage is not as relevant.

18

Just would need the diameter of the phosphor to calculate the area.
I’m hoping that better binned SST12 are available.

Clemence might have some 219CT-V2, they were out of stock when I bought optisolis but that was a few months ago. I’d be happy to donate some to get them tested.

19

1.38 mm² without dome

Maybe I will ask him.

Since I (and other users) wait since seven months (!) for delivery of LEDs of a group buy in the german flashlight forum I am not sure if I want to wait so long.

1 Thank
20

Well now shipping from Indonesia is like 10€ so group shippings aren’t needed anymore, not that it’s very quick since he still take quite a bit of time to ship stuff.