Cree XP-L, XM-L performance in an XP size package

This is from about 15’, camera settings adjusted to show the hot spot.

And the other one de-domed clean in 1 1/2 hours, at 90º F.



I put the de-domed XP-L in a SupFire F9, very close to being the same as a C8. This one, with a fresh 35A at 4.21V pulls 5.78A for 1363 OTF at start, 1301 at 30 seconds.

Here’s a side by side photo of the hot spot. Camera got the white balance odd, the de-domed looks about right but the domed one shows much more blue in this pic than the eye sees. Go figure.

That's disappointing, but it's still putting out more lumens than XP-G2, so it's still a win if you're okay with more flood.

Thanks for the pictures. That die does look very familiar. So the story sofar is that indeed it seems like an xml2 die on a smaller substrate, leading to a somewhat worse heatpath and a bit lower output?

I’m sure you’ll be able to get more accurate and clear readings on em.

This is good news for the small triple crowd though. Considerably more lumens and higher current capability, doing a fair job of maintaining tint. If Carclo gets those TIR Triple Optics done I’d like to see these in the V5 2A bin. I have a couple of 2C as well, not sure I’ve got stars for em. Had to disassemble working lights to put these in, lol.

To my eyes the domed 2A looks nice in tint. The de-domed is considerably warmer, orangy yellow by comparison, but not green at all so that’s a big plus. There may be that bit of a blue or light lavendar to the 2A, it’s still pleasing in actual use. The de-domed tint is pretty nice, especially in the lower modes.

So, it’s not difficult to get em focused in a reflector, but you would do well to have some XP sized centering rings that I do not have.

Thanks Dale!

Not as impressed as I hoped to be.

I get the distinct impression that Cree designed these for arrays, making them easily capable of butting up against each other with the square sides. As such, they’d be awesome in that context. For us, there’s only a few instances they’d be of major gain…excellent in those places but not pre-empting the XM-L2 emitter for general use.

With more die area than substrate, imagine a single row of them a few feet long... 8)

Get these and we can start fighting the sun with photon wars.

How about if they were arranged like the mini-dies in an MT-G2…72 of em in a solid pattern…90,000+ lumens of awesomeness! The “die” base would be about 4” in circumference, oh yeah, that would be impressive. Build one Comfy! lol

In a single row, that's 87 LEDs per foot... 296 volts @ 3A, 888 watts...

The actual die measures .15” square.

It's bigger than XPG2's substrate?

All I know is that my digital calipers say .15” measuring the square base.

I measure XPG2 as .1365" side to side, .190" corner to corner. So, is this one using a bigger substrate just with the same XP pad layout on the bottom?

I took a bit more care measuring and got .13” or 3.5mm. My caliper doesn’t go to the 4th digit.

Don’t know how much these can be trusted. El cheapo plastic ones from Harbor Freight.

I like the de-domed XP-L V5 2A enough to put it in my Sinner Ti, where it draws 5.71A from a Sony C5 at 4.21V for 1345 OTF.

Actually, I'm still very hopeful. Sure, it doesn't do well when over driven, but that's not so easy to do with a triple or quad mcpcb. If it can still pull 7.5A in RMM's S3 like it does with XP-G2's, that will be around 3000 lumens. I don't see why hope for those results should be diminished yet.

Let's just hope that Cree does whatever magic they did with this thing to bump up the efficiency of the XM-L soon while allowing it to be driven hard without burning up those little wires. XM-L3?

I’m probably misunderstanding, but I don’t think it’s possible to get 3000 lumens from 7.5A, especially in a quad or triple.

Slewflash, in theory 2.5A to an XM-L2 is 1000 lumens. The triple XM-L2 will do that, I’m pretty sure, and maybe then some depending on cell used.

That said, we ARE talking theory…I have a triple XM-L in a chopped mini mag pulling 7.47A from an AW IMR14500 that puts 1356 OTF with Carclo 10507 optics. These are older XM-L emitters and they’re de-domed, so there are losses.

Yeah… that’s just it. That’s in theory with adequate cooling. Even my Shocker running each XML at 3.8A each doesn’t touch 3000 lumens. I get closer to 2800 OTF. With quads and triples, there’s way less material to pull away the heat and less surface area so in theory (and in practice) triples and quads get heatsoaked way faster and have less ability to cool than singles (which is how an XML2 is reaching 1000lm @ 2.5A in theory).

Hence, there would have to be some kind of wizardry to reach 3000lm from 7.5A.

I’m not trying to diminish anything, but let’s be real - the release of the XP-L is not the day we reach 3000lm @ 7.5A.