XPE2 build question

This XPE2 is on a lux board. see here: Cree XLamp XP-E2 White LED
It is also AMBER, what max amp can i hit this with using the infamous Qlite driver found here: QLITE REV.A 7135*8 3.04A LED DRIVER - 17mm ?

At 3 amps, its pretty much guaranteed to blow.

Here is some emitter testing, and you can see that at 1.6A, its already starting to flatten out.

Hi

I had a xpe2 in sinkpad and copper below pushing strong at 3A, when lowering the current to 2.1A this increase performance :slight_smile:

djozz just did a test on XP-E2s with DTP here

I don’t think you want to push much more than about 1.5 amps. Maybe I’m wrong but 3 amps seems quite high for that LED…

+1 on that. On all but the highest binned emitter, you are still well into diminishing return.

How about i just go with this: http://www.mtnelectronics.com/index.php?route=product/product&path=67&product_id=240
Nanjg AK-47A 1.05amps ?

I’d say that would work. You could add one or two 7135s and still be in the safe zone, with more light.

Found this one: Nanjg 101-AK-A1 which is 1.4Amps. and its only a few cents more.
Whats the differences between these Nanjg drivers VS the Qlites ?

I wouldn't push that little XP-E2 on an aluminum MCPCB too hard, past 1A or so.

spread those electrons across three emitters?

I guess it’s time to start making these things extensively programmable, come to think:

… a one-cubic-millimeter computer …

Disclaimer: When explaining all of this I don't intend to give lecture at any specific member, especially not the member I quoted here. I just have seen this concept misused on this forum before, and believe it's commonly misunderstood.

Let's keep in mind the true meaning of "diminishing returns".

If I'm starving, I have no food and manage to buy one sandwich, it's very useful. If I'm starving but have 5 sandwiches I'm about to eat, it's not anywhere near that important to buy one more. That's the principle of diminishing returns. If trying to buy one more sandwich causes the loss of 2 out of my 5 plus my money, it's not diminishing returns. It's just evil and it's worse than diminishing returns. This is known as negative returns.

We have diminishing returns as soon as the lumen/watt graph is not linear. That means we sort of always have some of that when driving any LED in a practical application. We have a lot of that any time we overdrive an LED. Typically the mfg's rate their components so that the amount of diminishing return is small or moderate when approaching specified maximum values under specified conditions.

With XPE2 we already have a significant amount of diminishing returns @ 1 A. This actually depends on the definition. We sorta have a bit of diminishing returns even at .6 A. We have a lot when we reach 1.5 A. We don't need anything close to 3 A for that.

When we are past the point where lumen output starts to decrease(or even when it remains the same) when increasing current, it's not diminishing returns anymore. It's worse than that. That's what we are talking about with a XPE2 @ 3A. It's not "diminishing returns". We are way past that, as we no longer have any returns. Increased effort at that point not only not benefits but also hurts. In other words, we have negative returns.

The way I interpreted the term, and the way I think most people here did as well, is like this, using your sandwich reference.

Firstly, LEDs do not starve per say, but you could say that they are hungry.

If I am hungry, and I want a sandwich, I will buy and eat one sandwich, maybe two. In return am no longer hungry.
If I was to buy and eat 5 sandwiches, I am simply wasting my money, because the problem was already solved, and the return I had gotten has diminished, slightly. If I was to buy 15 sandwiches, now the return has seriously diminished.
Now I’m just fat and broke….
And I really don’t want another sandwich.

If the emitter functions as you need at ~1.5 amps, why give it 3?
Its just being wasteful at that point.
Technically you are correct, the moment the graph strays from linear, the return starts to diminish, but only slightly. Eating 5 sandwiches is going to give you a nasty stomachache later, but it won’t kill you. This is similar to lightly overdriving the LED. Eating 15 could potentially kill you, or at the very least make you very sick, like severely overdriving the LED.

That's what I thought and meant when saying I'm not trying to lecture any specific BLF member or you. But it's commonly misused.

Yes, that would be negative return. One would just gain more weight and feel worse.

Correct. I'm just saying that per match's test results, or at least when intuitively extrapolating them, XPE2 is going to negative return way before 3 A. At 1.5 the point of diminishing return would have already been reached by any definition. Reading your previous post again it becomes even more obvious to me that it's what you meant as well.

Match’s test was done on an aluminum board. If it was reflowed to a noctigon, the ouput would begin to plateau around 2.5A.

+1

Got my parts in today
Used an old Ultrafire C8 host with an alu pill. boy this little XPE2 in AMBER sure is a thrower!
I took pictures, however my new Galaxy S5 GOLD seems to be corrupting every new photo I take.
I am going to try and taker new pics in a little bit. stand by…