K2 and Lux 1, 3, 5 are discontinued?

[Blindasabat]

Looks like it according to future.
http://www.futurelightingsolutions.com/products/luxeon/

From that page:
"LUXEON K2, LUXEON K2 with TFFC, LUXEON I, LUXEON III, LUXEON V, and LUXEON Rebel low flux, are no longer available for new designs. Customers should contact their account representatives to discuss the availability of these products and for information regarding tools available to assist you in migrating to LUXEON Rebel."

 

[lolzertank]

There's already a thread on this.

 

[Blindasabat]

Ah, it's a few days old, that 's why I didn't see it. I don't frequent this sub-forum. My bad.

 

[jtr1962]

There's already a thread on this.

Actually the other thread only mentions the K2. While I'm a little surprised that the K2 TFFC is being discontinued, I'm really more amazed that the Luxeon I, III, and V survived as long as they did given that Lumileds stopped improving their output long ago. Seriously, given the other LEDs out there which not only cost the same or less, but also output far more lumens, I couldn't imagine anyone designing a Luxeon I, III, or V into a new product. I really couldn't even imagine continuing to use them in an existing product when Lumiled's own K2 offers similar spatial distribution, higher output, and similar cost. I do remain surprised at the K2 TFFC, however. That LED almost held its own with the best in my tests. If Lumileds improved output by about 15-20% it would hold its own among most of the offerings from SSC or Cree (with the exception of the new XP-G).

 

[Blindasabat]

Discontinuing K2 TFFC's only surprises me a little. I posted a thread a while ago that TFFC were not as efficient as Rebels, but I since put a TV0D (275L at 1.5A, Vf = 3.1V) in a 1st gen SF L1 head and it is one of my favorite lights. The output, tint, and power handling is awesome. I run it at a stable 2.2 Amps of an IMR123 putting out 300+ emitter lumens and it takes the heat with little droop compared to XR-E &SSC P4. It's not as efficient, but I think I will buy a few more to hold onto for special applications. If they did make a big Rebel to replace it with 100L/Watt, it would be the big torch LED of choice for me.

I agree Lux 1, 3, 5 should have been let go a long time ago.

I wonder if Lumileds is closing a plant.

 

[R@ndom]

Damn, no more side emitters.

 

[lolzertank]

Actually the other thread only mentions the K2. While I'm a little surprised that the K2 TFFC is being discontinued, I'm really more amazed that the Luxeon I, III, and V survived as long as they did given that Lumileds stopped improving their output long ago. Seriously, given the other LEDs out there which not only cost the same or less, but also output far more lumens, I couldn't imagine anyone designing a Luxeon I, III, or V into a new product. I really couldn't even imagine continuing to use them in an existing product when Lumiled's own K2 offers similar spatial distribution, higher output, and similar cost. I do remain surprised at the K2 TFFC, however. That LED almost held its own with the best in my tests. If Lumileds improved output by about 15-20% it would hold its own among most of the offerings from SSC or Cree (with the exception of the new XP-G).

My bad. I thought Lumileds discontinued the Lux 1, 3 and 5 ages ago. The XP-G has me all giddy too.

 

[HarryN]

Discontinuing K2 TFFC's only surprises me a little. I posted a thread a while ago that TFFC were not as efficient as Rebels, but I since put a TV0D (275L at 1.5A, Vf = 3.1V) in a 1st gen SF L1 head and it is one of my favorite lights. The output, tint, and power handling is awesome. I run it at a stable 2.2 Amps of an IMR123 putting out 300+ emitter lumens and it takes the heat with little droop compared to XR-E &SSC P4. It's not as efficient, but I think I will buy a few more to hold onto for special applications. If they did make a big Rebel to replace it with 100L/Watt, it would be the big torch LED of choice for me.

I agree Lux 1, 3, 5 should have been let go a long time ago.

I wonder if Lumileds is closing a plant.

Each LED package has operating points that make it ideal for specific applications. If the LuxV had been upgraded to a K2 package, I would have gladlly used it in some projects.

The K2 has better thermal properties than the rebel, and the XP-G for that matter. Rebels and XP-Gs are fine for 1 amp applications, but only a K2 can take that kind of abuse. Even an MC-E, really a powerhouse - has very specific warnings in its data sheets about protections that I have never seen in a Lumileds data sheet.

PL is not closing plants - it is producing at full tilt and adding capacity. That shows you how much of the their market has shifted to rebels and the Altilon.

 

[Curt R]

The Lux I, III and V survived as long as they did was to fulfill multiple year contracts. When a large company designs a product they must recover their design cost over a period of years that the product will be in the marketplace. The contract will contain penalties for non fulfillment. That protects the buyer and also gets the lowest cost for the buyer and reverse penalties to protect the manufacturer. Most major flashlight brands require at least a three year run on a design plus the design time to bring the light to market. The average consumer at Wallyworld does not know the light that he just purchased is at least two years obsolete in regards to technology.

Curt

 

[Curt R]

An industry source that I deal with said that many of his customers are upset about the announced death of the Luxeon LEDs. They are mostly non flashlight manufacturers and are scrambling to find an alternative. He also mentioned that the Cree XR series may not be around much longer. I do not know if that is true or not, however if it is true that would leave the Seoul P4 as the remaining LED in that size form factor. Using the new chemical formula phosphors for the white LED that Seoul is starting to use, they should be able to match the current drive and output level of the discontinued K2 TFFC LED. And with an even better output if Cree will sell Seoul the die used in the XP-G LED, and if it can be used in the P4. A lot of ifs.

Curt

 

[HarryN]

Hi Curt - thanks for keeping your ear to the ground on LED package direction. The loss of the Lux V in the past, and the K2 now has really caught me flat footed as well. There are some designs that just are not possible with low profile surface mount LED packages.

For the first time, I am seriously studying Cree's offerings - I just hope the MC-E doesn't go by the wayside.

 

[blasterman]

They are mostly non flashlight manufacturers and are scrambling to find an alternative.

Which baffles me because aside from flashlight applications I can't see what makes a Lux III unique. K2s and Rebels would seem to offer a better solution on either side.

Also, I don't think anybody has mentioned that bulk Rebels when purchased in reels and free-flowed yourself would seem to be a cheaper option than you'd get from Cree.

 

[Curt R]

The K2 death has caused me to redesign 3 flashlights that I was planning to bring out this year. I have about 60 K2-0220 LEDs that I now have no usage for and that I was going to put into those lights. They will be replaced by the new XP-G. The only problem now is finding the right optics to mate with the XP-G and the mechanical design of the flashlight head. The XP-G is limited to less current input than the K2, but the XP-G will produce more Lumens than the K2 at 50% less power draw. That is a big plus as to battery life. The sticking point is still the optics. Raw horsepower is great, but if you can't get it to the ground you come in second.

Because of the mechanical limitations of the MC-E there is no room for improvement. The larger Seoul P7 could use the XP-G die times four. I would expect to see the MC-E to have a short life also and be replaced with a SST-50 type package from Cree. There may also be a XN-I Cree in the future also. X being the single die form. Backing down two letters from R and P to N for package type and up two letters for die size as E and G to I. Just speculation on my part. That could be a XP-G type form factor with a 2 mm size die mounted on a 4.5 mm ceramic platform. That would give an output of about 75% greater than the XP-G LED and a thermal package of about 4 degrees per C.

Curt

 

[HarryN]

Which baffles me because aside from flashlight applications I can't see what makes a Lux III unique. K2s and Rebels would seem to offer a better solution on either side.

Also, I don't think anybody has mentioned that bulk Rebels when purchased in reels and free-flowed yourself would seem to be a cheaper option than you'd get from Cree.

It isn't just a matter of cost of the part, there is considerable engineering cost and time to rework a design, plus setup a new production process, optics testing, supply line, etc. I had a design specifically designed around the K2 which cannot be done in rebels due to the way I built it with the driver - so I am back to the drawing board.

The K2 is capable of extreme abuse - far more than just about any LED out there. Sure - you can say - design in this feature - but even then, having an LED with real abuse resistance built in is really handy. Overdriven , the K2 is like an over clocked pc - you can go far beyond spec with it.

The place where the Lux III is particularly strong technically was in red / orange. It has a much larger die than the Lux I, K2 or rebel, and at 1 amp - I think is still at or near the top for a "moderate" size die. (XP-G like ) IIRC, just one r/o Lux III can drive a rear tail light application. I have one of these in a custom light from Photon Fanatic, and it is just amazing. It also operates over a wide range of current - much wider than the Rebel equivalent - making it great for poorly regulated uses.

I don't doubt PL had good reasons to move in this direction, I just wish I had not been so close to a product when they decided to pull the trigger.

I agree with you on the Rebel and its applications though - it is quite inexpensive in volume applications. Given that each LED needs an optic also plays a part in any design selection.

 

[Curt R]

HarryN has it correct with the Lux III red LED, we were able to drive that LED direct with a 3.7 volt RCR123A battery in our IV medical light. The closest replacement is from LedEngine at five times the cost. Both the Seoul P4 and the Cree XP-E are slightly more efficient but are limited to drive current.

Curt

 

[Black Rose]

I have about 60 K2-0220 LEDs that I now have no usage for

Regular or TFFC?

 

[Curt R]

TFFC

Curt

 

[R@ndom]

The K2 death has caused me to redesign 3 flashlights that I was planning to bring out this year. I have about 60 K2-0220 LEDs that I now have no usage for and that I was going to put into those lights. They will be replaced by the new XP-G. The only problem now is finding the right optics to mate with the XP-G and the mechanical design of the flashlight head. The XP-G is limited to less current input than the K2, but the XP-G will produce more Lumens than the K2 at 50% less power draw. That is a big plus as to battery life. The sticking point is still the optics. Raw horsepower is great, but if you can't get it to the ground you come in second.

Because of the mechanical limitations of the MC-E there is no room for improvement. The larger Seoul P7 could use the XP-G die times four. I would expect to see the MC-E to have a short life also and be replaced with a SST-50 type package from Cree. There may also be a XN-I Cree in the future also. X being the single die form. Backing down two letters from R and P to N for package type and up two letters for die size as E and G to I. Just speculation on my part. That could be a XP-G type form factor with a 2 mm size die mounted on a 4.5 mm ceramic platform. That would give an output of about 75% greater than the XP-G LED and a thermal package of about 4 degrees per C.

Curt

Interesting. I recall 4sevens teasing, as he does, about a mc-g.

 

[Blindasabat]

I am still finding use for the K2 due to its toughness. I just put a TV0D on a not-so-well heat sinked Terralux Mag drop in with great results. I run a K2 TV0D and a K2 (I think TSNF) neutral direct drive off IMR123, 16340, 18500, and 14500 with no problems. I blew one XR-E that way a few months ago. The TVOD especially responds to this due to low Vf even at high amps.

I think 2x K2's running in series off a 2.8Amp P7 or MC-E driver would be a hot (yes hot) setup. Hmmm, I have a new bike headlight idea...

 

[HarryN]

Hi - yes, the K2s will be around for a while - just trickier to have a known supply base for a high volume product. For general hobby use- they are quite good.

For a project I am working on, I designed it around 2 Series K2s, so now I need to transition it to something else. My options seem to be:
a) 2 XP-Gs - and the risk of product shortage / limited optics
b) 4 Rebels - 2S x 2P - and try to force fit in optics (but at least they are around)
c) 1 MC-E - and try to buy tight optics in a small package, and possibly the part going away + it is crazy expensive.

No really good options, but I have decided to try both (b) and (c)

In light of this, I have opened a thread on Rebel Optics / reflectors / beam shots here in case anyone wants to advise / comment.

http://www.candlepowerforums.com/vb/showthread.php?t=247081

Thanks

HarryN