I have this Luminus SFT70 cool white led for ages, bought some from Simon, but did not come round to testing it. I had some time this afternoon for testing, but not enough to make it a slick post, so here it is, a bit rough. For methods see my other led tests (just 1 led tested, on a DTP coppper board on a chunk of fan-cooled copper, light was measured in an integrating sphere, djozz-lumens multiplied with 0.93 get you close to maukka-lumens which are absolutely calibrated).
I did not do a tint test (yet) but like the SFT40 I bet the tint is between 5000K and 6000K and under 70CRI.
The die size seems almost exact the size of the Cree XHP35 Hi, here a vague pic for comparison (also with a XP-L Hi)
The SFT70 is a tiny bit rectangular which seems to make the die size about 9% larger than the exactly square XHP35 Hi.
Here are my current to voltage/output numbers, and an analog graph.
I did not add other graphs of earlier tested leds, but the most relevant would be the XHP35 Hi and I never tested it (Koef3 did test a E2 bin one in 2017 though) so I can not add those numbers anyway. If Koef3’s calibration is similar to mine (I believe it is), the SFT70 is an upgrade to the XHP35 Hi with the added advantage that it is a 6V led which is a bit easier to drive in a flashlight than the 12V XHP35.
(At 12A the tint is not so friendly anymore , but the led survived fine: when going back to 3A both voltage and output were close to the first measurement)
If my numbers are correct (probably not) the SFT70 performs worse than the XHP50.3HI in overall output lumens (you measured the XHP50.2 up to 5k lumens so its probably expected). I measured the XHP50.3HI at just over 4k lumens in my L6 and at 3 amps in a C8 it does about the same as SFT70 at 2650 lumens.
It would be interesting to see the candela numbers.
I did not compare this led to all other relevant leds (even less so than in previous led tests), so I can not answer that now, but at least keep in mind that using a led in a flashlight will reduce the output by easily 20% compared to measurement of the bare led as I do.
Thanks Djozz. Unfortunately, the 6V SFT70 is not terribly exciting anymore given the variety of new domeless Chinese Leds which are 3V. If the Chinese Leds can improve their efficiency and CRI, they will be great.
I remember that long before the pandemic, someone here contacted Luminus directly and there was information that they were planning sst40 in high cri. Unfortunately, I don’t remember who it was and which thread.
Predictably though, I don’t have access to these contacts, and even if I did, I don’t think they would be willing to answer any new questions
Still, the fact that they actually did make their larger LEDs in high CRI variants does mean client demand for such a SKU is rather high.
I’m surprised it took this long to find them in the market however, which is rather strange considering that the last time they talked about it was in 2019.
Maybe they are sub-contracted to a test factory? And what is sold is only a test run which is under constant improvement?
We know CREE before sale started subcontracting factories to produce some of their LEDs, there have been also rumors that before the sale of the company they even gave the rights/tech to some factories to produce some LEDs in exchange for cheaper manufacturing costs.
Which is not that far fetched considering that not longe before CREE was announced to be sold (usually such process take months of even years), all of the sudden all kind of “fake” CREE LEDs showed up.
Recently put the sft70 in a convoy m21d, as a thrower; the beam is quiet powerful, yes,bit greenish at lower amp, but at full blast the tint turn creamy white.
Noticed some donut hole, same as sst70 in some flashlight.
Tested before in a convoy S11, the beam is solid. This led throw for sure
, but the led survived fine: when going back to 3A both voltage and output were close to the first measurement)
