Thanks for the heads up. That XP-P looks very interesting.

Superstocker wrote:

Thanks for the heads up. That XP-P looks very interesting.

Interesting. XP-P implies an XP mount size and from the pic, looks the phosphor is W1/W2 size, smaller than Boost HX.

Could this possibly de throne the w1 or Boost HL? Hmmm. Wonder what kind if vf it will have. Hopefully the xpg4 will be a viable alternative to the sst20?

welight wrote:

comes in 7000k,6500k,6000k, 5700k and 5000k

5K new throw king? Yes please!!! If it’s rated at 700 lumens by Cree I feel there is a good chance it will at least equal the cslnm or culnm in throw.

XP-P in 5700K with a price lower than the osram W1 would be nice. Imagine these in an MF01S…12000lm 300kcd.

I’d gladly welcome Cree back into the mix here. I won’t hold my breath for the xpg4, but the xpp has my attention. 3535 white flat equivalent? Yes, please.

That is exciting. Hope we can drive it past 1000 lumens.

XPP looks interesting. thanks for sharing

You down with XP-P?

Yeah, you know me!

1mm^2 die area is nice.

XP is 3.45×3.45mm size, which is smaller than the CULNM1 (4040 size) which means worse thermals
The 700lm rating is not that great, and the actual performance will depend on how much it can be overdriven to.
The CULNM1 can be driven to over 1000lm so the XP P will need to do more than 150% output to beat it.

Enderman wrote:

1mm^2 die area is nice.

XP is 3.45×3.45mm size, which is smaller than the CULNM1 (4040 size) which means worse thermals
The 700lm rating is not that great, and the actual performance will depend on how much it can be overdriven to.
The CULNM1 can be driven to over 1000lm so the XP P will need to do more than 150% output to beat it.

Hope it takes some juice to serve as a straight replacement in XP-L Hi lights.

why do i feel like it’s more high cct low cri junk?
why can’t they release good leds for throw?

welight wrote:

comes in 7000k,6500k,6000k, 5700k and 5000k

What do they mean by max lumens >700 ?
Is that just a way for them to conservatively estimate a new untested product?

FWIW I’m hoping for something FET-DD capable, and hopefully brighter than the CSLNM1.TG.

kramer5150 wrote:

What do they mean by max lumens >700 ?
Is that just a way for them to conservatively estimate a new untested product?

FWIW I’m hoping for something FET-DD capable, and hopefully brighter than the CSLNM1.TG.

I’d prefer a low enough Vf that we can use regulated drivers.

Agro wrote:

4040 CULNM1 is not much better than 3030 CSLNM1 which means that the improvement in thermals ain’t large.
I expect that the difference between 3535 at this die size would be meaningless in practice (though could be measurable on the bench). So it boils down mostly to who can build a better die.
For several years Osram used to be the throw leader so I wouldn’t bet on Cree. But there is some hope and it’s nice to see Cree trying.

Enderman wrote:

Agro wrote:

4040 CULNM1 is not much better than 3030 CSLNM1 which means that the improvement in thermals ain’t large. I expect that the difference between 3535 at this die size would be meaningless in practice (though could be measurable on the bench). So it boils down mostly to who can build a better die. For several years Osram used to be the throw leader so I wouldn’t bet on Cree. But there is some hope and it’s nice to see Cree trying.

It’s like a 10-15% difference, pretty big…

But small enough as to be overcome by binning. It’s hard to say with what we have but the above would perhaps put the XP-P at about a lower bin Culnm1. So I think Agros point about whose die is better is right. It was very hard to find good bins for the Culnm1. So, if good bins of the XP-P come out, it does not seem unreasonable that it could equal or exceed what we can get from osram. Additionally I would happily take a little loss in performance for a warmer cct.

Thanks welight

chouster wrote:

Thanks welight

+1

If datasheet is correct, it can't touch Osram at higher currents, and will not be able to handle overdrive like Osram.

Two graphs show this:

1) Flux vs temp - Cree (left) drops significantly more at higher temps:

2) Flux vs current - Cree(left) needs 3,0Amps to double flux measured at 1Amp,Osram needs only about 2.35Amps for the same 200% increase; at 3Amps Osram is at ~250%.

Cree at 3Amps (and that's with pulse testing at 25C) looks already close to its peak output, under real life testing peak would be probably at ~4Amps with lumen output marginally better compared to 3Amps.

Vf voltage is also significantly higher compared to Osram.

Agree - I noticed that flux vs. current chart and thought there's no way you 'd get anything out of 5 amps, maybe a little at 4 amps, if the LED can take 4 amps.

Well I guess I’m too slow on this one.. I agree with Tom and L4P that the overdriving isn’t going to be there on this one if the prelim datasheet is accurate. 225% relative output max. At 420lm binning, maybe we see 945 lumens. Might be a more efficient alternative at the 3.5A range. Only one way to know I guess.

Thanks for the data sheet welight and thanks for bringing those charts together Led4power. It is unfortunate that the cslnm1 “handles heat” better than the XP-P. I was hoping the latter would do slightly better than the cslnm1 in that area and be able to be overdriven slightly more. The data sheet has bad news there. The good news is that if we can get top bins then it will still be close with a better color temperature.

The output curve sees to peak at ~225%
Even the best bin only does 400lm at 100%, so max would be 900lm best case scenario.
Still definitely worse than the CULNM1 unfortunately (4040 size), and probably also worse than the CSLNM1 (XP size) when you take into account thermals.

The final version of the datasheet:
https://cree-led.com/media/documents/dsXPP.pdf?WT.mc_id=CRD027

Nice! So we can expect 720 lumen at 3A, and 5000K, now see how they compare to the Osram version