LED test / review – Luminus SST-40-W Specialty White N4 BA (+ dedomed) – Here it is, the next gen XM-L2!

LED-Test

EN



Luminus SST-40-W Specialty White LED


SST-40-CW-A120

UPDATE --- 06/23/18 --- raw data added! (link)

This is the english version of my first test (August 2017). Because of more accurate results and some minor changes in the text the german version is no longer valid anymore!

I bought this emitter at Kaidomain several months ago.


Technical data



Tj 85 °C, If 700 mA

Order code: SST-40-CW-A120 (unless otherwise specified)


Type: single die (lateral)
Flux bin: N4 (min. 300 lm)
rated voltage: VJ (typ. 2,8 V, min. 2,7 – max. 2,9 V)
Forward current: max. 3000 mA
Viewing angle: typ. 120 °
Thermal resistance: typ. 2,5 °C/W
Junction temp. Tj: max. 150 °C



Official datasheet is available here (official homepage, pdf)

Caution: datasheet is still in preliminary state (March 2018)!

First appearance



The SST-40-W is very similar to the well known XM-L2. The sharply demarcated die sitting on a silver grey ceramic base. A silicon dome covers the LES which enlarges it’s surface significantly and increases also the total light flux.

The emitter is 4.95 x 4.95 mm (0.195 x 0.195 in) in size and equals to the XM format (also known as 5050).





Thanks to the symmetric package the use of lathe spun center rings is possible. Moreover, because of the standard footprint the whole range of already available accessoires for XM / 5050 can be used.

LED chip and die



The LES is sharply demarcated, like the other well known Cree emitters (XP-L, XM-L/2). There are no discolored areas around the die so the beam pattern and color over angle should be very high.

The LED chip is built in classic lateral design, which is uncommon for actual LED technology and products. The most of new released emitters are designed in Flip Chip or CSP technology which improves the thermal characteristics and more efficient use of the LES.

Typical for lateral LEDs are small areas for bonding contacts which cannot emit any light.




The LES equals with dome 8.02 mm² (0.0124 sq in).

Without dome the LES is smaller by almost exactly the half, but 3.99 mm² (0.00618 sq in).

Power and overcurrent capabilities


Tsp 25 °C, unless otherwise noted

raw data here

Domed

Within official parameters

  • At 5,000 mA (max rated current): 1,843 lm @ 3.31 V
  • Power at rated maximum: 16.6 W
  • Efficiacy at 5,000 mA: 111.0 lm/W
  • At 700 mA (binning current, but 25 °C Tsp): 357.0 lm @ 2.84 V
  • At 700 mA (binning current, calculated to 85 °C Tj, for reference only): 327.5 lm

Overcurrent

  • Maximum at 9,200 mA, 2669 lm @ 3.84 V
  • Power at Maximum 35.3 W
  • Sweet Spot at 7,400 mA, 2388 lm @ 3.56 V
  • Power at Sweet spot 26.3 W
  • Efficiacy at max 75.6 lm/W, in sweet spot 90.8 lm/W

dedomed


Hint: I powered the LED only to max 8,400 mA not to destroy the emitter.

  • Maximum reached at 8,400 mA, 2116 lm @ 3.69 V
  • Power at maximum 31.0 W
  • Sweet spot at approx. 7,400 mA, 1984 lm @ 3.56 V
  • Power at Sweet spot 26.3 W
  • Efficiacy at 8,400 mA 68.2 lm/W, in sweet spot 75.4 lm/W

Interesting facts

  • The limiting factor of this emitter are the bonding wires which burnt at 9,400 mA. In the diagram it can be clearly seen in Vf curve that of about 7,500 mA the voltage increases significantly which is a clearly sign that the wires are reaching their maximum load limit.
  • I cannot recommend a current of more than 7,400 mA because of shorter expected lifespan of the bonding wires.
  • Till 4,500 mA the output of the SST-40-W is nearly the same of a XP-L HD V5. Over 5,000 mA it is brighter because of the better heat dissipation in bigger package and reaches also a higher maximum. At 7,800 mA the XP-L died (burnt wires) at 2,264 lm and 4.39 V.
  • In dedomed state the SST-40-W is less efficient than a XP-L HI V3. In general there is the same behavior as against the XP-L HD: above 5 to 6 Amps higher maximum possible and therefore a higher flux at highest currents. A replacement of XP-L HI V3 not really worth it unless the lower Vf is superficial, like for better regulation or better total efficiency.
  • The XM-L2 tested here is very efficient, due to its reached highest flux bin U4. It is also more efficient than XP-L and SST-40-W. But also the Vf is very high, especially at the highest possible power, and ist bond wires blown at 5,200 mA.
  • The old XP-L HD V5 features a higher flux as the XP-L2! Basically it has only a lower Vf and therefore a higher efficiency, due to a not fulfilled flux bin (see test of XP-L2 here).

The LED how it looks after its death at 9,400 mA:


The comparison of SST-40-W and XP-L2 is also interesting. It is important to know that the XP-L2 shown here does not fulfill its claimed V5 bin so that these statements are not mandatory for other batches of this type!

  • Against the new generation XP-L2 the SST-40-W features the higher efficiency, the lower voltage and – if the wires wouldn’t burnt – a higher maximum light flux at almost same current.
  • Moreover, the SST-40-W achieves a better beam pattern and more homogeneous colors in corona and spill, not like XP-L2 or XP-G3.
  • In general, the SST-40-W performs in almost all disciplines better than the XP-L2. The only real drawback could be the earlier death at 9.4 Amps but considering that the maximum of the XP-L2 is around 11.5 Amps and the light flux increases only at approx 160 lumens this problem is in real world applications negligible.

At low Tj (10 °C and below) the light flux decreases surprisingly instead of rising, like at most other LEDs at these temperatures. I've been able to reproduce this behavior several times. Also the Vf rises significantly higher at low temperatures than stated in the official datasheet.

From this point of view I cannot recommend this LED for applications, which are used at low temperatures, like outdoor lighting or even flashlight. The lower the temperature is so lower the light flux seems. Unfortunately currently I cannot go lower than -5 °C Tj... :(

Dedoming?




The dedoming is not as easy as we know from earlier lateral LEDs like XM-L2 or XP-G2. With chemical dedoming the dome breaks into smaller pieces, but some parts remains still on the LES and ceramic base. Maybe the results can be optimized by changing the time of exposure, but I haven’t tested this yet.

Hot dedoming can work, but unfortunately in not all cases. The risk to damage the phosphor on the LES is still there. In general the phosphor layer of the SST-40-W is more crumbly and sensitive against mechanical treats than of earlier LEDs (XM-L2).

Important hint: Absolutely avoid touching the LES after dedoming, there’s a high risk of damaging the phosphor layer! To stay on the safe side, I recommend the shaving method in which the dome is sliced over the LES to increase also the luminance.

Luminance

I established a new method for determining luminance, especially to ensure more realistic values for 'real-life' conditions (flashlight use). The measurements are taken with a new original Convoy C8 reflector, but with same heatsink / setup as previously used in flux measurement.

Reflectivity for reflector 85 %

Transmission UCL glass 97,8 %

Values at 25 °C Tsp, for 85 °C Tsp values are 6 to 14 percent lower, depending on LED

LEDs marked with Warning sign uses old values are still taken with previous method (determining die size) due to problematic light distribution (donut holes etc).



Hint: I powered the sample used for luminance measurement only up to 8,400 mA not to destroy the LED so the values at Absolute max current are a little bit higher!


In case of improving the range / brightness compared with XP-L HI V2 / V3, the gain in throw is negligible. For sure, the luminance is a little bit higher as of the XP-L HI V3 because the SST-40-W profits from the bigger footprint, but I would not recommend to drive the SST-40-W at such a high current to ensure a higher lifetime (bonding wires).

Because of all my tested XM-L2 samples died at only 5.6 Amps I cannot provide any luminance measurements of it. Despite that the SST-40-W is way better than the XM-L2 because of higher flux in dedomed state, much lower Vf and – depending on prevously used LED – higher efficiency.

Light quality and use in optics



Thanks to the classic lateral design the beam pattern and color behavior corresponds to the XM-L2 and (in some points) XP-L. There are no other colored areas around the spot or corona, neither in use of reflectors nor TIR optics / lenses.

Because of this I will not post any beamshots of SST-40-W in optics, because they are already known of (modded or original) flashlights also in this forum.

The both small non-lit areas of bonding pads does not affect the beam pattern at all. The tint is a cold and bit blueish white, at approx 7500 to 8000 K CCT. The color rendition is low which is normal for cool white light sources in this color (rendering) class. Unfortunately this LED is only available in low CRI and only in cool white although in the (preliminary) datasheet are stated also warmer tints (low to 4500 K CCT) which aren't available still today.

Conclusion


Here we go – the next gen XM-L2!

That’s basically everything you know about this LED. The efficiacy is very high, it can be overpowered very well. Also the Vf is reasonably low, even lower as of the XP-L2.

Luminus delivers a very nice and state-of-the-art LED which is in almost every thing better as the actual Cree emitter generation and the (inofficial) predecessor XM-L2. Only drawback are the burnt bonding wires at 9.4 Amperes which limits the maximum possible flux in a very hard way.

I did not like the difficult dedoming, I can only recommend the shaving method for this.

For lights and applications with XM footprint or still equipped with older XM-L2 especially in lower bins this LED might be the (almost) perfect upgrade, also the beam quality is really high, also known from earlier LEDs.

In short summary as I mentioned at the beginnging: here we go – the next gen XM-L2!

Pro

  • higher maximum current possible as XM-L2
  • lower Vf as XM-L2
  • high efficiacy at highest currents
  • good beam pattern in conjunction with optics
  • high luminance in dedomed state
  • low price

Contra

  • dedoming is somewhat difficulty
  • only available in low CRI and cool white CCTs
  • burnt bonding wires which limits the maximum power possible with its LED chip
  • at low Tj light flux decreases
  • datasheet (as of mid March 2018) still in ‚preliminary‘ state altough this LED is on sale for longer time

Addendum: I think it is not really professional for such a LED manufacturer like Luminus to publish a datasheet still claimed as ‚preliminary‘ (in which crucial subjects still could change without any further notice!) although this LED is already in mass production and has been delivered to suppliers and companies. Also it might not the best way for change the subjects and important maximum (electrical / behavior) characteristics of the datasheet AFTER the LED was released for general sale. In an earlier version of the datasheet the maximum forward current was stated with 3,000 mA, not 5,000 mA like now.


Thanks a lot for reading! :)

Greetings, Dominik (aka BLF member koef3)

Mistakes and suggestions are best sent via pm.

Very nice review. Thank you! What kind of tint shift did you get after dedoming?

Nice test!

Warmer (approx 5000 - 5500 K CCT) tint with a bit green in it. But much less than of Cree LEDs (XP-G2 / XM-L2), so it should be okay. Better color tint you might get if you slice the dome.

Wow, almost 100% improvement of cd by dedoming :open_mouth:

Nice, thanks. :slight_smile:

Thanks :+1:

Thanks for the testing :beer:

Thanks for the test. With the low Vf and high luminance dedomed this LED is the optimal one (IMO) for single cell triple emitter applications. The emitters are driven very well without the risk of blowing the bond wires. With an aspire 4300mAh 26650 my convoy L2 triple SST40 pulls 19A.

I built a triple that pulls just over 24A with these (multiple cells in parallel though). It’s a fun light but I’m not very fond of the tint.

At that brightness you could could use a filter and it would still be very bright.

Orsm testing. Thanks. :+1:

Thanks Koef3

Thanks.

Thanks for the test :slight_smile:

Nice review!
I achieved 160 kcd in an Emisar D1S with dedomed SST40. Nothing else changed just the led.
Same led in an Eagle eye X6 with fet driver: 88,5 kcd
Same led and fet driver in Convoy C8 AR coated lens: 136,3 kcd with dome: 70,5 kcd
Convoy L2 Fet driver with dome on: 172 kcd

Can anyone suggest a good 17mm to 22mm 1s-2s buck drivers to run this LED up to 7-8A, or so?

hey ZozzV6

how did you dedomed SST40?

I’m curious too, I put one with dome in my D1 so I want to try in my D1S too

It was an early led that somebbody told here on blf it was easier to dedome. I putted it in regular 95 octane petrol floating on wires upside down. And the dome falallen off. It leleft a little on the phosphor but I removed it with an Xacto knife under microscope.