The New Luxeon III Generation Issues

[tvodrd]

No good deed goes unpunished, and the beatings will continue until morale improves!

Larry (I got some of those K2 thingies coming too.)

 

[IsaacHayes]

It's official, the round slug luxeons are now starting to appear with the original die height back. You can no longer look at the slug to tell if it will be a joker or not.

Lumileds has used up all their joker slugs it appears and are back to the normal height.
Now we must inspect /compare visually the die's vs looking at the shape of the slug.

 

[sgt253]

Gents, at the risk of sounding ignorant, here it goes... The current batch of TWOJ 3 watt HD45's are...non jokers? The more I read the less I seem to comprehend. Any help greatly appreciated.

Don

 

[McGizmo]

sgt253,
I think you would need to get that information from Wayne or Cindy. The McR-45 was designed before the joker hit the scene and I have not designed or had a McR-45-J made. The reflector is large and deep enough that it is the most forgiving in terms of beam if it is coupled with a joker. I actually prefer it with a joker but I don't know what the shoppe is using with these. :shrug:

 

[sgt253]

Sounds great. I will definitely ask the Shoppe then. Thank you for the reply. Look forward to acquiring a few pieces of your work. Regards.

Don

 

[McGizmo]

Hi guys,
I have had a reel of the K2's on order all year. I ordered the PW12 with S flux as I see merit in this part even if the Vf is less than optimal. I read where folks were getting K2's of the PW14 part and felt that I wasn't mising out on something I could use.

I just received notice from my wife that Future had shipped me some LED's! Cool! Well, that was my first thought but of course I should have considered the source! :green: It seems that someone at future decided to substitute on my order and what I received was a reel of 250 LED's of the PW14 part number and further, I had my wife determine the bin and what I got was UYAN. Prior to identifying just what I got, I had my wife cut 5 LED's off the reel and add them to a needed care package she was sending me (rush job here). As a result, I am the proud owner of some yawns (phonetic yet appropriate description). Yeah, I really need these LED's! Not! Do I welcome an alternative to the Future/ Lumiled program? No more than I would welcome a suny day after a nasty storm.

Now the K2 is not a new Luxeon III generation issue; it is its own generation but certainly not without issue, IMHO!!!!

Now I have added a question to a note I sent my Future rep regarding this latest shipment and it is one that I would love an answer to but somehow doubt a quick and concise response. In the K2 specs, they caution about a possible .5V drop in Vf within hours of reflow of the K2. Well a .5V drop in Vf of a 1.5 amp LED represents 3/4 of a watt of heat that will go away in a couple hours. Does this desireable break in of the K2 require the heat of reflow?!?!?!?!?! If we plan to use thermal epoxy for mounting K2's do we need to season them first with a good thermal zap? Do we wait for a few hundred hours of operation while the LED slowly reforms into a more reasonably efficient device? WTF? :thinking:

 

[MSI]

Maybe we have to start to make constant temperature controllers instead of constant current controllers to run those K2's properly. :thinking:
Another nice side effect of a constant temperature controller would be a more intelligent light that would give you more light on a cold winter night.

 

[NewBie]

My long term Vf drop tests:

The Vf drop may be more substantial with less heatsinking and occur quicker with less heatsinking.

This is why constant current sources are so important, as it would make a huge difference in how much heat was generated with a given voltage, since the current would go rapidly up.

I have one of these that ended up pulling over 10 Watts, when it was ran off a constant voltage supply of 4.3V, over time. It started off around 6W. Starting current was 1.5A. It was up over 2.3 Amps after the period of time. The device is mounted on a thick copper plate that is over 1/4" thick, 5" wide, and 12" long, with cooling air blowing over it.

These puppies sure drift alot!

 

[IsaacHayes]

Don: that really sucks that you didn't get the Sbins. I'd be mad if they substituted my order! What if you were banking on the 12 p/n for use at 350ma, them sending you the 14's defeats the purpose of having the option to order the 12 for 350ma drive level..

K2 seems pretty wonky to me right now.

 

[McGizmo]

Isaac,
From my admittedly subjective point of reference, it seems that Future/ Lumileds just keeps lowering the bar in regards to customer service, communication and general apathy. :shrug: Like a lame TV soap, it is interesting to watch and see what next! :green:

I was not impressed at all with the efficacy of the initial K2 samples and felt safe that the -12 S would be a better part than those I had seen and I was willing to wait as long as it took! We have good Luxeon III's to work with and others are biting at Lumiled's heels so in no time, we will have alternatives! I have some real impressive Nichia samples of a new LED that is reported to be coming out later this year. At about 1.1 -1.2 watts, it delivers a legit 65 lumens. I am looking forward to Lumiled's answer in this arena, not in a device that you drive at 1.5 amps with 7+ watts (mostly heat)!!

Still, IF the K2 can be subjected to or needs to be subjected to, an elevated temperature cycle to yield up to a .5V reduction in Vf, ceterus peribus (assumed) then it would be nice to friggin know this!!!!!!!



yeah,

 

[IsaacHayes]

Good news on the Nichia, sounds nice. I just hope the nichia and other competitors can make their power leds compatible with reflectors/optics that are in current use. Having a prefocused dome like a 5mm makes it very hard to collimate the beam tighter than what the led's dome can produce it'self.

I get what you're saying about the K2 Vf. Are the typical #'s after reflow and voltage drop? It will have to drop a lot to get to what they list as typical Vf...

I don't understand how they decided to go with the new electrical distribution dots in the K2 if it ends up with a less efficient led (due to the high Vf). Is not the dots there to extract more light due to not blocking it? If that's so then you are going one step forward two steps back in terms of efficiency. I don't see how the few extra microns of exposed surface area makes up for the redicoulous high vf...

Since the R/O/A K2's appear to use the traditional lux1/3 die, I wonder if their vf at 350ma will be the same as Lux1's... It *should* be, unless the interface of the tip of the pyramid to the slug has some resistance due to a new thermal interface. I would assume that it's probably the white/blue/etc chips that have the changes and maybe the r/o/a don't even have any changes...

 

[McGizmo]

OK, update regarding my bad and some crow for lunch. :green: I was informed by my Future rep back in Feb that I could only order the PW12 in R bin and I guess at that time, said to go ahead and give me the PW14 in U bin instead. Bad memory strikes again.

I found this out in conversation on the phone today with rep. He and another fellow there understand my query about the magic of reflow and .5 Vf reduction and they will follow up with Lumileds to find out more about this. It was reported that the Vf drop is rather consistent from part to part (an issue for any matched parts paired in parallel). If these parts need to be nuked to become more reasonable in terms of efficacy, it would be handy to know; especially if one is not intending to mount with reflow!!!

 

[IsaacHayes]

Anyone remeber the thread where someone's KL4 was lost and fell into the toaster and they finally found it there. Then they reported lots better runtime. Heat must have reduced the Vf, and thus with a constant current driver, that equals more runtime.

I'm waiting for the directions on cooking the K2's... how long in the microwave on high??

 

[tvodrd]

A thanks to Don and Newbie for keeping the masses of the unshaven informed! :bow: and

Larry

 

[NewBie]

I just looked it up...Note 4 page 6, http://www.lumileds.com/pdfs/DS51.pdf

"The forward voltage of the Luxeon K2 LED will reduce by up to 0.30V at 350mA during the first few hours of operation after SMT reflow."

Emphasis, mine. Notice that this is after reflow is done.

But even more important, keep reading:

"Due to this effect, Philips Lumileds recommends current source drive for consistent and reliable performance. Cross connected series/parallel arrays or voltage drivers which could result in current hogging or variation in drive
current are not recommended."

Emphasis mine again.

For a T bin K2, see page 7:

"The forward voltage of the Luxeon K2 LED will reduce by up to 0.50V at 1000mA during the first few hours of operation after SMT reflow. Due to this effect, Philips Lumileds recommends current source drive for consistent and reliable performance. Cross connected series/parallel arrays or voltage drivers which could result in current hogging or variation in drive current are not recommended."

I have seen this same Vf drop effect, after SMT reflow, and without SMT reflow, but I wanted to characterize things closer before I talked about it much... The Luxeon I and Luxeon III also show a Vf drop with time. Not all brands of high power LEDs show this Vf drop with time.

There is an older thread where I talked about this with one of the scientific researchers at LumiLEDs (Michael Krames- Manager Advanced Laboratories, Lumileds Lighting), and it was due to several mechanisms.

We also talked a bit about the subtantial increase in LED efficiency when current dimmed, and the loss of efficiency when PWM dimmed (for white, blue, green, cyan, dental blue, royal blue), and why it doesn't occur in their reds.

It is just much worse in the K2, than the Lux I and Lux III. But when you start driving a Lux III up at 1,000mA, they come close to changes one would find in these K2s.

There is an old thread from 2004, where we were talking about this issue, and you can find some of the charts here:
http://candlepowerforums.com/vb/showthread.php?t=60217&page=4&pp=40

evan9162 has some better charts in some other threads, dealing with Vf shift in Luxeon I, Luxeon III, and Luxeon V devices, but I'm not sure where those are hiding right now....

 

[evan9162]

Although this won't be super scientific due to the low samples, I plan on running my 3 K2s that I ordered at 3 different current levels to see the effects that it has on the Vf drop. I'll run them at 700, 1000, and 1500 mA.

I think heat and current have a lot to do with the Vf drop. When I was testing my 5 TY0Ls, I decided to experiment on one. After about 3 days, the Vf hadn't dropped (my meter can read a 10mV difference) for about a day. This one was being tested at 700mA. I cranked the current up to 1000mA, and I got another 50mV over the next 12 hours - this was a permanent drop, not simply caused by an increase in Tj.

I've seen a general trend that higher current (and therefore, higher Tj) result in a larger drop. But I haven't yet separated the high current from the high Tj, as I like to use a fan cooled heat sink for the Vf drop testing.

 

[McGizmo]

If one were to take two LED's that initially seemed to be twins in terms of flux and Vf and subject one to the elevated thermal cycle of re flow and not the other and then mount both with a thermal bonding agent to identical heat sinks and drive both at the same current and monitor Vf I wonder what would be seen?!?! Any of us can season or break in a LED with high current and its associated thermal elevation but the even higher temperatures seen in the reflow are not so easy to duplicate. My concern is that portion of the LED's improvement that may be attributed to the reflow cycle. If, as one Future rep stated, the K2 can be reflowed a number of times and if this elevated thermal condition does drop Vf, then why doesn't Lumileds friggin cook the LED's prior to binning and shipping and give us a more stable and mature part?!?!?!? Let the bloody parts shift on Lumileds shift and not on ours!! As I have stated before, a reduction in Vf of .5 V at 1.5 amp is a reduction in drive power of .75 watts!! If flux remains constant, this is a reduction of .75 watts of wasted heat and energy!! We want this reduction if at all possible and we want it sooner and not later! If one designs a a cool running and low current demand light, one may never see the gain that could have been had if the part had been snapped into a better place by tempering it with heat and current!?!?

I used to have a moped shop and was told by a 2 stroke mechanic that the best 2 stroke engine in terms of efficiency and power was one that was broken in hard and at a level just below seizing the engine. Is this a similar situation here?

 

[CM]

BTT.

 

[NewBie]

I just got looking at the pictures, and combined the two emitter halves into one:

Not only is the die a different height from the slug bottom, it is also optically oriented differently within the dome "lens". It appears to no longer be oriented in the center of the half dome lens. This too could be corrected, but would need a prescription to be applied to the reflector.

Anyone got better, and clearer, and larger more detailed high resolution photos of the two, so we could make a better comparison?

So, what was the final resolution on this, did they go back to the old height, or adopt the new height?

 

[wasBlinded]

They went back to the old height, once all the old joker slugs were used up.