Cree XP-G S3-bin

That is just a spec sheet. Where does it say the S3 bin is available for sale? We still have not seen any production S2 bins yet.

Manufacturers like to lists bins above what they can make so they do not have to do a new spec sheet every time they make some improvements.

Cree released this road map for the XRE many years ago. The R4 & R3 bin is unobtanium. The R2 bin is in limited supply. The Q5 bin is more commonly used. Many flashlights are still using as low as P4 bin.
Creeroadmap.jpg


SSC ran into a lot of problems with binning the P7. They originally listed the C bin as 740-900 lumens. The flashlight manufactures promptly silkscreened 900 lumens on the sides of the torches. Testing showed OTF lumens at 350-500 for those torches. After much controversy about accuracy of SSC numbers, SSC did actual testing of production LEDs and re-binned the C bin at 700-800 lumens. OOPS.

Take numbers of non-production LEDs with a grain of salt.

S3 bin will be available soon. But it won't be an XP-G. It will be an Xlamp XM.
http://www.candlepowerforums.com/vb/showthread.php?t=270517
 
Last edited:
Are you sure? They listed the S3 in the XP family spec sheet, so I guess it will still be XPG.
 
Cree released this road map for the XRE many years ago. The R4 & R3 bin is unobtanium. The R2 bin is in limited supply. The Q5 bin is more commonly used. Many flashlights are still using as low as P4 bin.

... but XP-E (with the same chip size) is avalaible in R3 bin now ...
 
The XRE development has ended and is replaced by the XPE.
So while some XRE R2 bin had been available in the past, the new R2 and R3 are only available in XPE. While the die size is the same the dome above is not giving a totally different beam.

I expect the XM to come out with an S3 before Cree makes a S3 XPG.
To get 164 lumen with an XM, Cree needs only 41 lumen per mm^2. 2mm*2mm die.
To get 164 lumen with an XPG, Cree needs 82 lumen per mm^2. 1.4mm*1.4mm die.
 
To get 164 lumen with an XM, Cree needs only 41 lumen per mm^2. 2mm*2mm die.
To get 164 lumen with an XPG, Cree needs 82 lumen per mm^2. 1.4mm*1.4mm die.

From this prospective XM looks as a stepping back in the progress as CREE already reached 71 lumen per mm^2 in their XP-G R5 (139lm/w).

I think the main reason why they put S2 on hold is marketing plans for XM, not technology difficulties. In many respects XM is a competitor for XP. For me there would be no reason to migrate to XM if they were released XP-G S2 or S3. XM is damn big, not compatible with rebel, with no optics available etc.
 
The main markets for LEDs are going to be backlighting LCD displays and fixed lighting. Not flashlights. Outside CPF most homes probably have just 1 plastic 2D flashlight. But most homes have 20+ fixed lighting bulbs. And a TV. And a computer monitor or 2.

In fixed lighting the typical 60W incan bulb is ~840 lumens. The XM at 2A is 750 lumens. Because LED bulbs tend to project more light forward than the olde Edison incan 750 lumens LED will feel brighter than a 60W incan. I suspect that is the target market for the XM LED.

Lower lumens per watt is not a bad thing. It means the LED is underdriven and will last longer. The Luxeon IV was only rated 500 hours but I have not seen a single complaint of a Surefire L4/KL4's dead LED. But if you try to sell an LED household bulb that lasts only 500 hours you won't be in business for long.

To recap for a flashlight highest lumen per mm^2 is needed for throw. XRE-R2. XPG-S3 if Cree gets around to making one. 1000 hours LED life is plenty.
For fixed lighting lumens per watt is important. Also a life of 50,000+ hours. So 160 lumens per watt of the XM is prized over the 139 lumens per watt of the XPG-R5 in this application.
Note Cree is recommending max 2A for the XM. The total die area is the same as the MC-E/SSC-P7 which CPF members are running at 2.8A+ for a few years now. Maximizing LED life seems to be the goal of this limitation.

The XM is not going backwards. It is going forward in a totally new direction.
 
Last edited:
I would expect that Cree will raise the maximum recommended current for XM-L after more testing (3A?). They did the same thing with XR-E, XP-E, XP-G...

I would even bet that XM-L @ 3A will be as bright or brighter than SST-50 @ 5A.
 
The maximum power driven to the XM-L will be limited to the thermal transfer from the LED to the heat sink.
The transfer area on the XM is much smaller than the MC or the P7. That means a more concentrated point of
heat that must be dispersed. We are able to do that with the SST-50 and the SST-90 by soldering an 8-32 brass
screw to the LED, and then using thermal paste when screwing the LED assembly into the the heat sink. Getting
rid of the heat from the XM will be a challenge at high drive currents because of its small size.

Curt
 
The maximum power driven to the XM-L will be limited to the thermal transfer from the LED to the heat sink.
The transfer area on the XM is much smaller than the MC or the P7. That means a more concentrated point of
heat that must be dispersed. We are able to do that with the SST-50 and the SST-90 by soldering an 8-32 brass
screw to the LED, and then using thermal paste when screwing the LED assembly into the the heat sink. Getting
rid of the heat from the XM will be a challenge at high drive currents because of its small size.
Curt

I can't help being slightly pleased if this means that the XM-L will not necessarily be the next "in-vogue" greatest thing when it comes to flashlights.

The market value of SSC-P7 and Cree-MCE based flashlights seems to have dropped because SST-50/90 are regarded as superior, although in my opinion this is certainly not always the case. It depends largely on the application and current drive-level.

I don't want the same thing to happen to SST-50/90 flashlights so quickly.
 
I acquired 5 of the Cree XRE R2 emitters back in the day. I waited quite a while for the R2 to become available. I have found the same to be true with waiting ( and still waiting ) for the XPG S2. I have had the thought that it might be better to build a mod with a XPG Q5 than wait for the mythical XPG S2 to become available. Oh well who knows maybe Elvis ridding a pegasus will fly down on moon beams and get me a S2 one day.
 
Lower lumens per watt is not a bad thing. It means the LED is underdriven and will last longer. The Luxeon IV was only rated 500 hours but I have not seen a single complaint of a Surefire L4/KL4's dead LED. But if you try to sell an LED household bulb that lasts only 500 hours you won't be in business for long.
I think you meant lower lumens per mm², at least judging by the content of the rest of your post. Other than that, good post and I agree completely. Fixed lighting and backlighting is really where it's at for LED these days, and in those applications the longer the life the better.
 
I would expect that Cree will raise the maximum recommended current for XM-L after more testing (3A?). They did the same thing with XR-E, XP-E, XP-G...

I would even bet that XM-L @ 3A will be as bright or brighter than SST-50 @ 5A.
Remember that the XP-G was originally rated at 1 amp by Cree but in my tests it had no problem handling in excess of 2.5 amps. The XM-L will likely be able to handle 5+ amps with no problem, although of course it will be pointless to drive it past where the output stops increasing. As usual, once I get my hands on one I'll be happy to test it to its limits and report my findings. Cree's ratings have more to do with prolonging life than with the limits of what the LED can physically handle.
 
Always looking forward for your testing!

I remember that back in 2006, the XR-E was rated only for 0.7A (1A now). Than XP-E was rated 0.7A (1A now), XP-G was 1A (1.5A now), recently maximum driving current of MX-6 was doubled from 0.5A to 1A... So it is safe to expect 3A for XM-L :devil:
 
Hi I'm a little new and am hoping I can get a little assistance.

I'm currently using white XPG-R5's (driven at 850ma) in a lamp but I seem to need a little more efficiency (needs to be a bit brighter).

From what I understand, the R5 is a 1.4mm chip, and the S2/S3 is a 2mm chip. From the documentation, it seems as if the S2 is maybe 7% more efficient/brighter than the R5. The size increase will fit in my application, will it be roughly 7% brighter at the same 850ma current?

Also, are the S2 or better, the S3's on sale?
 
Top