Cree XM-L3 U4 1B from kaidomain (and Luminus SST-40 N5 5000K from convoy) tested

I hope it becomes available in warmer CCTs than the SST-40.

But even then what will be the gain over the currently available XM-L2’s?

Thanks for coming out of retirement to do these tests djozz. As usual your work is invaluable (I know you dont agree) in selecting leds. :+1: :beer:

Thanks Steve, I guess you are right that I’m sort of retired from led testing, the procedure has become a bit boring and I’m not as curious as I was before. And there are not as many really exciting new leds as there were a few years ago it seems, progress is slowing down.
But if a really promising new led appears I will probably test it again, i.e. I’m looking forward to DIY LEP components becoming available :slight_smile:

Like, 200 lumens with FET, looks like. Maybe this one will come in 4000K 90+ CRI…

Oh wait, it’s Cree. Nevermind. Luminus will probably do HCRI SST-40 in warm bins first.

Thanks for the review. At last, proof that the single-die 5050 hasn’t evolved much in 6 years. I wouldn’t exactly call it an upgrade from the sst40. I think the 4040 would be of interest for innovation. Luxeon has three market cornered for that with the V, but Osram’s Boost HX is a very interesting emitter indeed.

I wonder why. The conclusions I’ve seen for 3535 footprint seem to indicate the footprint can only handle so many watts before all the emitters peak. But 5050 I’d expect to see a lot more from. Certainly the XHP50.2 3V has proven how much heat it can handle. But even the SST-40 doesn’t really do anything different than Luxeon V at 4040.

An explanation for that could be that the heat in the led itself does not spread sideways well and does not reach the edges of its thermal pad as easy as the middle. So if there is a good heatsink under the led the thermal handling is more determined by the die size than the package size.

Thanks taking the time to test these LEDs and share the results djozz. Being relatively new to the flashlight hobby just about the first thing I do when I encounter an unfamiliar LED is check your tests - they’re a fantastic resource.

Mouser has some U4 5700K's in stock: https://www.mouser.com/ProductDetail/Cree-Inc/XMLDWT-00-0000-0000U40E2?qs=%2Fha2pyFaduiteuA67J2OdsXxK1prd%252BHc7s8LT7qmCbfZ%252BOvlG%252BRwzoCzfBtgIwnn

U4 5000K's are expected in November: https://www.mouser.com/ProductDetail/Cree-Inc/XMLDWT-00-0000-0000U40E2?qs=%2Fha2pyFaduiteuA67J2OdsXxK1prd%252BHc7s8LT7qmCbfZ%252BOvlG%252BRwzoCzfBtgIwnn

your second link is to 5700k not 5k, dont expect 5k from Cree in XML-3

Ooops, yep. This is the full Mouser listing of XM-L3's: https://www.mouser.com/new/cree/cree-xlamp-xml3-leds/

But looks like the 5700K's which were in stock are out of stock now

Do you have them, or are looking to stock them at Cutter? It's an interesting LED coming from CREE, but compared to the 5000k SST-40, it's not quite there yet.

That Mouser page has part #'s down to 4000K, though not real yet.

Thanks. Good info. Just as suspected.

im advised by cree they are not planning any CCT below 5700k at this time. we are stocking only the 5700 to 6500k
Cheers

Are you going to test the CULPM1.TG?

Sounds DOA to me. Too bad. I’m sure they’ll still make their way into plenty of stock lights though.

I’m still planning to.A few months ago I have already tested a sample that Hank sent me (it tested not revolutionarily better than the 3030 version, seemed a low bin to me), and I have one that I bought from Simon that still needs testing. One of these days…

I’m hoping it can handle a fet driver. There’s a video on youtube of it running on the A17DD-L FET+1

Hank’s sample had 95% of max output at 7A/3.23V, its max at 9A/3.32V/1500lm and was still alive at 12A/3.47V (with 80% of the max output).

So if you run the led with a high drain 18650 on a direct driver while keeping resistances low, the current will almost certainly go over 9A and the led is run over its max, and you might not even notice that because the output is still high.
On the other hand, if you introduce some resistance here and there (a bit less high drain cell, an un-bypassed spring) you might hit the 7A sweetspot while still using a FET driver.

I do not expect the curves of Simon’s Boost HX leds to really differ from this, but I hope that they are a bit better bin, so that you can actually see the advantage of the 4040 footprint over the 3030 footprint.

Thanks djozz.

Simon is making an 8A driver so maybe I’ll stick with this. These usually run .5A less so probably the best option over a fet

Guessing that driver is 22mm so might have to use my fet drivers in a C8. I’ve ordered 10pcs so will test once they arrive