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

196 posts / 0 new
Last post
Jerommel
Jerommel's picture
Offline
Last seen: 1 year 2 months ago
Joined: 01/04/2014 - 13:18
Posts: 6380
Location: the Hague, Netherlands

Strange…
You would expect them to first get the efficiency up before reducing the heat path.. :~

But maybe i missed something, like some jokes earlier… Silly

comfychair
comfychair's picture
Offline
Last seen: 5 years 2 months ago
Joined: 01/12/2013 - 05:39
Posts: 6198

The differences shown in that chart are basically the same as the steps between flux bins of the same LED family, nothing revolutionary there that I see. If XP-L was 200L/W instead that would be a different story.

Helios-
Helios-'s picture
Offline
Last seen: 1 year 8 months ago
Joined: 01/18/2012 - 21:12
Posts: 2099

Jerommel wrote:
Strange…
You would expect them to first get the efficiency up before reducing the heat path.. :~

The smaller package allows more LEDs to be tightly packed together. Remember flashlight use only accounts for a small amount of LED sales.
According to cree’s specs, thermal resistance has not been reduced. 2.5 °C/W same as is listed for XM-L2


Counterfeit 18650s, 2,<a href=“http://

leaftye
Offline
Last seen: 7 months 2 weeks ago
Joined: 07/25/2012 - 17:43
Posts: 4278
Location: San Diego, CA

comfychair wrote:

The differences shown in that chart are basically the same as the steps between flux bins of the same LED family, nothing revolutionary there that I see. If XP-L was 200L/W instead that would be a different story.

What?  Look at the chart again.  200 Lm/W is right there at the top.  If has an efficiency and output lead throughout.  Maybe that will change at higher currents.  In my near time mods, that's not going to be an issue.  

Bins don't matter here since the XP-L V6 bin is currently available, and the highest bin XM-L2 is the U3 and is virtually impossible to acquire.  The V5 is almost the same as the U3, so there's still a gain even if the availability of the V6 goes away

If it can't be overdriven to 2000 lumens due to the smaller footprint, hopefully Cree comes out with higher bin XM-L2 or XM-L3.  The problem with XM-L3 is that there aren't any triple or quad direct bonded copper mcpcb's, or Carclo's to put over them.

 

Here's another chart with the XM-L2 U3 bin.

The low mode should be lower.

Jerommel
Jerommel's picture
Offline
Last seen: 1 year 2 months ago
Joined: 01/04/2014 - 13:18
Posts: 6380
Location: the Hague, Netherlands

I thought the typical well known (to us) X-lamps were actually designed for things like flashlights.
Up to now i haven’t encountered them anywhere else.

djozz
djozz's picture
Offline
Last seen: 58 min 10 sec ago
Joined: 09/07/2012 - 17:04
Posts: 17864
Location: Amsterdam

There is at the moment (no date added) an announcement on the Cree homepage that they have improved the XP-G2 (7% brighter), XM-L2 (18% !! brighter) (and MK-R and CXA leds as well). We'll see when those become available.

djozz
djozz's picture
Offline
Last seen: 58 min 10 sec ago
Joined: 09/07/2012 - 17:04
Posts: 17864
Location: Amsterdam

Jerommel wrote:
I thought the typical well known (to us) X-lamps were actually designed for things like flashlights. Up to now i haven't encountered them anywhere else.

I have a GU10 light bulb with four XP-E's in it.

comfychair
comfychair's picture
Offline
Last seen: 5 years 2 months ago
Joined: 01/12/2013 - 05:39
Posts: 6198

leaftye wrote:

comfychair wrote:

The differences shown in that chart are basically the same as the steps between flux bins of the same LED family, nothing revolutionary there that I see. If XP-L was 200L/W instead that would be a different story.

What?  Look at the chart again.  200 Lm/W is right there at the top.  If has an efficiency and output lead throughout.  Maybe that will change at higher currents.  In my near time mods, that's not going to be an issue.  

Bins don't matter here since the XP-L V6 bin is currently available, and the highest bin XM-L2 is the U3 and is virtually impossible to acquire.  The V5 is almost the same as the U3, so there's still a gain even if the availability of the V6 goes away

If it can't be overdriven to 2000 lumens due to the smaller footprint, hopefully Cree comes out with higher bin XM-L2 or XM-L3.  The problem with XM-L3 is that there aren't any triple or quad direct bonded copper mcpcb's, or Carclo's to put over them.

 

Here's another chart with the XM-L2 U3 bin.

I think you're deliberately making a false comparison. By your own account, where the XP-L is able to do 200lm/W, the XM-L2 is doing 194lm/W. You're still acting like XM-L2 is only 100lm/W and then acting amazed at how the XP-L is such a great leap forward. Apples and oranges. They're also different flux bins even though there's absolutely zero evidence they use a different die - except for the bump up in flux bin. An XM-L2 V6 bin would be virtually indistinguishable from a XP-L V6.

comfychair
comfychair's picture
Offline
Last seen: 5 years 2 months ago
Joined: 01/12/2013 - 05:39
Posts: 6198

193.9 vs 200.8
165.4 vs 171.2
149.5 vs 154.2
136.5 vs 140

What I said still stands, it's basically just a bump in flux bin. An XM-L2 die stuck on a XP substrate.

Helios-
Helios-'s picture
Offline
Last seen: 1 year 8 months ago
Joined: 01/18/2012 - 21:12
Posts: 2099

djozz wrote:
There is at the moment (no date added) an announcement on the Cree homepage that they have improved the XP-G2 (7% brighter), XM-L2 (18% !! brighter) (and MK-R and CXA leds as well). We’ll see when those become available.

What?! :bigsmile:


Counterfeit 18650s, 2,<a href=“http://

comfychair
comfychair's picture
Offline
Last seen: 5 years 2 months ago
Joined: 01/12/2013 - 05:39
Posts: 6198

http://75.65.123.78/screenshot.17-06-2014%2009.48.59.jpg

Difference between XM-L2 T6 & XM-L2 U3 is even more dramatic than between XM-L2 & XP-L.

DB Custom
DB Custom's picture
Offline
Last seen: 3 days 3 hours ago
Joined: 01/13/2013 - 22:28
Posts: 20718
Location: Heart of Texas

Is it just me or does it appear that the tiny fragile bond wires are missing in this new version? If it indeed has a post for the bond wire, might it also handle current better? Higher?

djozz
djozz's picture
Offline
Last seen: 58 min 10 sec ago
Joined: 09/07/2012 - 17:04
Posts: 17864
Location: Amsterdam

I think I see a classic bond wire in this picture:

comfychair
comfychair's picture
Offline
Last seen: 5 years 2 months ago
Joined: 01/12/2013 - 05:39
Posts: 6198

DBCstm wrote:
Is it just me or does it appear that the tiny fragile bond wires are missing in this new version? If it indeed has a post for the bond wire, might it also handle current better? Higher?

They're still there. Maybe a bit shorter because of the more cramped package, or that could just be a distortion caused by the edge of the dome.

Tom E
Tom E's picture
Online
Last seen: 2 min 49 sec ago
Joined: 08/19/2012 - 08:23
Posts: 14453
Location: LI NY

Looks like the Vf is right between XM-L2's and XP-G2's -- that's a disappointment. I think comfy is right - not much to get excited about. Only nice jump would be lumens if it must be done on a XP-G/XP-G2 footprint, and the lower Vf compared to an XP-G2. If you had to de-dome to make it fit optics, you'll lose lumens of course.

Actually I would love to drop one in my T10... I have the stock XP-G in there now.

leaftye
Offline
Last seen: 7 months 2 weeks ago
Joined: 07/25/2012 - 17:43
Posts: 4278
Location: San Diego, CA

comfychair wrote:

leaftye wrote:

comfychair wrote:

The differences shown in that chart are basically the same as the steps between flux bins of the same LED family, nothing revolutionary there that I see. If XP-L was 200L/W instead that would be a different story.

What?  Look at the chart again.  200 Lm/W is right there at the top.  If has an efficiency and output lead throughout.  Maybe that will change at higher currents.  In my near time mods, that's not going to be an issue.  

Bins don't matter here since the XP-L V6 bin is currently available, and the highest bin XM-L2 is the U3 and is virtually impossible to acquire.  The V5 is almost the same as the U3, so there's still a gain even if the availability of the V6 goes away

If it can't be overdriven to 2000 lumens due to the smaller footprint, hopefully Cree comes out with higher bin XM-L2 or XM-L3.  The problem with XM-L3 is that there aren't any triple or quad direct bonded copper mcpcb's, or Carclo's to put over them.

 

Here's another chart with the XM-L2 U3 bin.

I think you're deliberately making a false comparison. By your own account, where the XP-L is able to do 200lm/W, the XM-L2 is doing 194lm/W. You're still acting like XM-L2 is only 100lm/W and then acting amazed at how the XP-L is such a great leap forward. Apples and oranges. They're also different flux bins even though there's absolutely zero evidence they use a different die - except for the bump up in flux bin. An XM-L2 V6 bin would be virtually indistinguishable from a XP-L V6.

No, I'm making a mathematically accurate comparison.  It's like you're selectively reading to prove your point.  Like I said, the XM-L2 U3 is virtually impossible to acquire, and even if it were, the XP-L V6 is still brighter and more efficient everywhere on the data sheet.  That's why I didn't put the U3 on the first chart, and apparently I shouldn't have put it on the second chart because you've misinterpreted why I put it there.  Sure, there's a data sheet for the XM-L2 U3, but good luck acquiring that unicorn.  There's no such thing as a XM-L2 V6 at this time, although the news djozz pointed out may change that very shortly.  I'll stick with comparing what I can get my hands on.  That is at best a XM-L2 U2 and a XP-L V6, and the XP-L V6 is undeniably brighter and more efficient.  Whether or not it's perceivable by eye is hardly the point.  I'm talking hard numbers here.

You're wildly misinterpreting how you think I'm acting.  Better is better.  If you are waiting for revolutionary improvements, improvements that are readily determined by eye, you wouldn't even be using XM-L's yet, much less XM-L2's.  Maybe that's all you care about, and if so, good for you.  I can understand that aspect, especially if the older emitters are much less expensive.  As for me, if I can get the brightest and most efficient emitters for an application, then that's what I want even if I can only tell the difference with a lux meter. 

The low mode should be lower.

Jerommel
Jerommel's picture
Offline
Last seen: 1 year 2 months ago
Joined: 01/04/2014 - 13:18
Posts: 6380
Location: the Hague, Netherlands

djozz wrote:

I think I see a classic bond wire in this picture:


Hmm… with those cut offs it’s maybe not suited for use with a refector..
Messy corona perhaps…
comfychair
comfychair's picture
Offline
Last seen: 5 years 2 months ago
Joined: 01/12/2013 - 05:39
Posts: 6198

leaftye wrote:

No, I'm making a mathematically accurate comparison.  It's like you're selectively reading to prove your point.  Like I said, the XM-L2 U3 is virtually impossible to acquire, and even if it were, the XP-L V6 is still brighter and more efficient everywhere on the data sheet.  That's why I didn't put the U3 on the first chart, and apparently I shouldn't have put it on the second chart because you've misinterpreted why I put it there.  Sure, there's a data sheet for the XM-L2 U3, but good luck acquiring that unicorn.  There's no such thing as a XM-L2 V6 at this time, although the news djozz pointed out may change that very shortly.  I'll stick with comparing what I can get my hands on.  That is at best a XM-L2 U2 and a XP-L V6, and the XP-L V6 is undeniably brighter and more efficient.  Whether or not it's perceivable by eye is hardly the point.  I'm talking hard numbers here.

You're wildly misinterpreting how you think I'm acting.  Better is better.  If you are waiting for revolutionary improvements, improvements that are readily determined by eye, you wouldn't even be using XM-L's yet, much less XM-L2's.  Maybe that's all you care about, and if so, good for you.  I can understand that aspect, especially if the older emitters are much less expensive.  As for me, if I can get the brightest and most efficient emitters for an application, then that's what I want even if I can only tell the difference with a lux meter. 

It is possible to misinterpret the significance of a mathematically accurate comparison, you know. The difference between any one XM-L2 flux bin and its one-step-higher or -lower flux bin shows about the same difference in total lumens and efficiency and everything else as the chart shows between XM-L2 U2 and XP-L V6. It's nothing but a single-bin bump XM-L2 die stuck on a smaller substrate. That's not a bad thing, of course, but it's nothing particularly remarkable or revolutionary.

So since you're all about the math, how does a dedomed XM-L2 compare to a domed XP-G2, since that's what started this whole thing? Of course dedoming loses some output, but it will still be more than the XP-G2 with very similar beam characteristics. You seem to be saying that a dedomed XP-L would be a step backwards from a domed XP-G2 and that's just not supported by facts.

Tom E
Tom E's picture
Online
Last seen: 2 min 49 sec ago
Joined: 08/19/2012 - 08:23
Posts: 14453
Location: LI NY

Qty 2 of the V6's ordered from IS... Gotta try em out. Your both right - it's a bump, not a big bump, but could be the best available for what it is... The Vf is closer to XM-L2 than XP-G2, so should make a nice replacement of XP-G2's. I'm not planning to de-dome them - I'll let some other sucka, I mean modder, try that Smile.

 Come on, get your feet wet!

Helios-
Helios-'s picture
Offline
Last seen: 1 year 8 months ago
Joined: 01/18/2012 - 21:12
Posts: 2099

djozz wrote:
There is at the moment (no date added) an announcement on the Cree homepage that they have improved the XP-G2 (7% brighter), XM-L2 (18% !! brighter) (and MK-R and CXA leds as well). We’ll see when those become available.

Where did you see this?


Counterfeit 18650s, 2,<a href=“http://

Boaz
Boaz's picture
Offline
Last seen: 38 min 33 sec ago
Joined: 11/07/2010 - 09:31
Posts: 7606
Location: Birthplace of Aviation

Tom E wrote:

Looks like the Vf is right between XM-L2's and XP-G2's -- that's a disappointment. I think comfy is right - not much to get excited about. Only nice jump would be lumens if it must be done on a XP-G/XP-G2 footprint, and the lower Vf compared to an XP-G2. If you had to de-dome to make it fit optics, you'll lose lumens of course.

Actually I would love to drop one in my T10... I have the stock XP-G in there now.

Put one in my t-10 too...!!!   it's all torn apart on my desk waiting for an emitter.  I need to make a Match mod clip for it too.

       καὶ τὸ φῶς ἐν τῇ σκοτίᾳ φαίνει καὶ ἡ σκοτία αὐτὸ οὐ κατέλαβεν

                            

       Dc-fix diffuser film  >…  http://budgetlightforum.com/node/42208

djozz
djozz's picture
Offline
Last seen: 58 min 10 sec ago
Joined: 09/07/2012 - 17:04
Posts: 17864
Location: Amsterdam

Helios- wrote:
djozz wrote:
There is at the moment (no date added) an announcement on the Cree homepage that they have improved the XP-G2 (7% brighter), XM-L2 (18% !! brighter) (and MK-R and CXA leds as well). We'll see when those become available.
Where did you see this?

Right on the homepage. The top-banner changes news automatically, it is not there, but below that there are three more items, one of them is 'Xlamp Leds Are Even Brighter'.

leaftye
Offline
Last seen: 7 months 2 weeks ago
Joined: 07/25/2012 - 17:43
Posts: 4278
Location: San Diego, CA

comfychair wrote:

leaftye wrote:

No, I'm making a mathematically accurate comparison.  It's like you're selectively reading to prove your point.  Like I said, the XM-L2 U3 is virtually impossible to acquire, and even if it were, the XP-L V6 is still brighter and more efficient everywhere on the data sheet.  That's why I didn't put the U3 on the first chart, and apparently I shouldn't have put it on the second chart because you've misinterpreted why I put it there.  Sure, there's a data sheet for the XM-L2 U3, but good luck acquiring that unicorn.  There's no such thing as a XM-L2 V6 at this time, although the news djozz pointed out may change that very shortly.  I'll stick with comparing what I can get my hands on.  That is at best a XM-L2 U2 and a XP-L V6, and the XP-L V6 is undeniably brighter and more efficient.  Whether or not it's perceivable by eye is hardly the point.  I'm talking hard numbers here.

You're wildly misinterpreting how you think I'm acting.  Better is better.  If you are waiting for revolutionary improvements, improvements that are readily determined by eye, you wouldn't even be using XM-L's yet, much less XM-L2's.  Maybe that's all you care about, and if so, good for you.  I can understand that aspect, especially if the older emitters are much less expensive.  As for me, if I can get the brightest and most efficient emitters for an application, then that's what I want even if I can only tell the difference with a lux meter. 

It is possible to misinterpret the significance of a mathematically accurate comparison, you know. The difference between any one XM-L2 flux bin and its one-step-higher or -lower flux bin shows about the same difference in total lumens and efficiency and everything else as the chart shows between XM-L2 U2 and XP-L V6. It's nothing but a single-bin bump XM-L2 die stuck on a smaller substrate. That's not a bad thing, of course, but it's nothing particularly remarkable or revolutionary.

So since you're all about the math, how does a dedomed XM-L2 compare to a domed XP-G2, since that's what started this whole thing? Of course dedoming loses some output, but it will still be more than the XP-G2 with very similar beam characteristics. You seem to be saying that a dedomed XP-L would be a step backwards from a domed XP-G2 and that's just not supported by facts.

It looks like we're getting closer to being on the same page, except for the second part.  Let's look back at what I actually said.

Quote:
That kills most of the efficiency gains though.

I don't see how that could be any clearer. 

I am truly interested in those facts though.  I've seen many claims of lumens being lost to dedoming, but I haven't seen a formula for it.  There's some data here and there, but I haven't seen it tabulated.  I'd really like to have a better look at those numbers so I can come up with a decent prediction of how much of the gains would be lost.  Is there enough information to estimate a data sheet for a dedomed XP-L?  That's academic though.  As I said, I don't have plans in the near future to put it under a Carclo, and if someone doesn't try it by then and there's still no XP-L Carclo, then I'll go ahead and take that gamble myself.

Actually, I may have been mistaken when I listed the XP-G2 S3.  The S3 is pretty hard to find too.  The S2 would have been a better fit in that chart.  Anyhow, I don't feel like pulling up the S2 data, so I'll stick with the S3.  The XP-L V6 has 26% more lumens at 1.5A than the XP-G2 S3.  How many of those 152 lumens would be lost when dedomed?  I predicted that the loss would be <152, but >76.  Note that I used ">" instead of ">=".

The low mode should be lower.

degarb
Offline
Last seen: 9 months 1 week ago
Joined: 07/02/2012 - 14:28
Posts: 109
Location: Just south of the future ice sheet

I think the next thing I will be interested is, when kaidomain or fasttech or such gets these in stock. Cutter.com.au is a bit pricey for hobby builds or any series of builds.

1. Cree claims the smaller size means they are cheaper to make than the xml2. I want to see this in action at that checkout.

2. Pretty much a small, but important, improvement on the xml (in efficiency and beam discipline), not xp-g2. (The xpg2 is still what I would want in anything sub 2×18650, like 1×18650 or 4 aa).

3. I got my willbuying xml2 u3 in like 7 days from order to usa. Really would like to see the xpl against this led. I haven’t done anything yet with my u3. I am thinking of building a tripple goose neck (3x hydra neck of about 3ish inches long each, pointed seperately—not too long and bouncy, but long enough), maybe xpg2 in center, mounted to a 4× 18650 pack with ledsupply 700 ma cc dimable controller, and velcro straps for mounting on a paint roller pole &/or bike.

Runtime=Usefulness

Tom E
Tom E's picture
Online
Last seen: 2 min 49 sec ago
Joined: 08/19/2012 - 08:23
Posts: 14453
Location: LI NY

I've seen ~20% in loss of lumens from de-doming with XM-L2's, maybe slightly less, using the gas method. I've heard of lower percentages but I just haven't seen any much lower, maybe 18%, but don't think less than 15%. This is several lights, including Super Shocker triples, but several XM-L2 singles as well, like HD2010's, C8's, etc. Can't guestimate XP-G2's - didn't do enough to see apples to apples. With the de-domed XP-G2's I always seem to lose amps, thinking from a higher Vf - others seen this too.

18sixfifty
18sixfifty's picture
Offline
Last seen: 1 year 9 months ago
Joined: 12/25/2012 - 20:19
Posts: 4161

I bought six of them. Should be interesting to see how they work on the 20mm triple noctigons.

I’m a junky, I mod lights so I can sell lights so I can buy more light to mod so I can sell lights to buy more lights to mod.

wight
Offline
Last seen: 1 year 4 months ago
Joined: 11/27/2013 - 16:40
Posts: 4969
Location: Virginia, USA

I’d also like to have a better feel for efficiency lost by dedoming. Sounds like a lot of work for somebody though, so I/we may have to make do without.

comfychair wrote:

The dome is the same size/shape as XM-L2, it’s just truncated at the sides to fit on the smaller substrate.

I was going to reply that I wasn’t so sure about this. I thought I’d looked into this previously and concluded that even a cut XM dome would not fit. Turns out that after reviewing the facts again, it seems clear that comfychair is correct. I don’t know what went wrong the last time I did this, but here’s the deal: The XM-L dome is approximately 4.4mm. I roughly measured this myself just now. The diagonal of an XP- sized package is ~4.737mm. My math isn’t so good these days so I used a calculator to get the diagonal. As you can see, this leaves plenty of room for the little spaces we’ve got in the corners of the XP-L (>0.15mm for each corner). Clearly comfychair is correct about the dome size.

I think they may have sharpened up the area where the dome joins the plain of the rest of the package. Previously there seemed to be sort of a small curved transition, this seems to be gone for the XP-L.

Still fine, still on a break. One day I’ll catch up with you folks! previous wight catchup Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

Tom E
Tom E's picture
Online
Last seen: 2 min 49 sec ago
Joined: 08/19/2012 - 08:23
Posts: 14453
Location: LI NY

18sixfifty - Are you gonna dedome to fit the optics, or maybe mod the optics? Maybe decide after in hand? I'm goin single for now. Definitely my T10, maybe a zoomie too - might be interesting... Should have ordered more than two.

 

wight
Offline
Last seen: 1 year 4 months ago
Joined: 11/27/2013 - 16:40
Posts: 4969
Location: Virginia, USA

Tom E wrote:

I’ve seen ~20% in loss of lumens from de-doming with XM-L2’s, maybe slightly less, using the gas method. I’ve heard of lower percentages but I just haven’t seen any much lower, maybe 18%, but don’t think less than 15%. This is several lights, including Super Shocker triples, but several XM-L2 singles as well, like HD2010’s, C8’s, etc. Can’t guestimate XP-G2’s – didn’t do enough to see apples to apples. With the de-domed XP-G2’s I always seem to lose amps, thinking from a higher Vf – others seen this too.

Since it came up…. have you noticed any change in Vf for those dedomed XM-L / XM-L2?

Lower output is what many of us care about the most. But lower output + higher Vf is an even worse situation. Just getting the same output at a higher Vf means a drop in efficiency… together that’s extra bad. Wink

Still fine, still on a break. One day I’ll catch up with you folks! previous wight catchup Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

Ejected Filament
Ejected Filament's picture
Offline
Last seen: 5 months 4 days ago
Joined: 12/03/2013 - 04:29
Posts: 2343
Location: Qld, Australia

Didnt someone successfully even if accidentally dedome as a result of heating the LED via high amps? IIRC, they said the colour did not shift and I was wondering if maybe that might translate to maintaining lumens too?

Maybe the ‘solution’/gasolene is the reason for all/much of the loses and colour shifts rather than dedoming perse?

Pages