How much current could the white Luxeon III take before going poof?

I ask this, because I just transplanted a Luxeon III from a 3 AAA 3W flashlight into a 3D Dorcy 1W Luxeon.

The 3AAA 3W obvious have horrible battery life, because AAA batteries do not handle 0.8A of current so well. In a Brinkmann 3AAA 3W LED flashlight, the LED is direct driven from three series connected batteries and uses the battery bank as the ballast. The voltage across LED is around 3.3v and the current is 0.82A, which comes out to around 2.7W. Current is lower than the short circuit current of the batteries, because LED adds resistance and current is smaller than regulated 4.5v across the LED, because internal resistance of battery brings the voltage down enough to lower the current to 0.82A. The design basically exploits the internal resistance of AAA batteries to act as a ballast.

So, the same LED transplanted into a 3D Dorcy light and direct driven draws about 1.6A with fresh alkaline batteries. I used thermal compound on every thermal path and ensured the LED isn't getting too hot to touch, but am I still hurting the emitter?

White, blue and cyan are rated for maximum 1,000mA and red, amber, etc are rated at 1,400mA per spec.

I measured my LEDBeam on fresh NIMH- came up with something like 4.7W as I recall... Haven't had any problems yet- but this thing is definetally overdriven. So long as the thermal path is solid, it should be good for awhile, but I would expect it to have a shorter life than it would have.

Well, the ratings are conservative because the engineers don't know what kind of heatsinking will be used, the duty cycle, etc. They're also trying to maintain lumen output for long periods.

I'd think you'll break down the phosphor faster at 1.6A, and maybe only have a 500 hour lifespan, rather than a 10000 hour one at rated lumens. Still, though, 500 hours is many battery changes, and by the time light output has dropped significantly, there should be even better LEDs available.

You might try NiMH cells, assuming the Vf of the luxeon is low enough, or add a small series resistor. The luxeons are tough little buggers, though, and with enough thermal mass, can tolerate quite a bit of abuse, as you've discovered.

Jon

For those that may not have seen this.


-DF

You shouldn't have any problems as long as you have good thermal dissipation. I'm running a TWOH DD off a 3.6V R123A. Many state not to do that and that concerned me a bit (tho I did it for a while and didn't have any probs). I then saw the chart DF linked to above and saw that an H-bin draws about 900mA which the Lux3 is rated to handle. I've also since added my own heatsink made of a couple of filed copper pennies (must be pre-80s to be Cu). I'm fairly confident the LED will last until I decide I don't want the light any longer or I break it another way (I've dropped this light a few times already, tho, no probs whatsoever...not even finish marring).

I added the 1.2 ohm resistor I removed into the circuit at the tail cap to test the current. With the Luxeon III in Dorcy 3D 1W light, 1.2ohm resistor about halves the current. Direct drive makes the light noticeably brighter, but diminishing return is apparent since ceiling bounce doesn't make it nowhere near as twice as powerful

I've pumped a TWAK to 2A regulated on a hotlips, but I don't recommend it. It still lives after about a total of <30 mins run-time though.

Diminishing returns on efficiency and heatsinking is not all you have to be concerned about. After re-soldering blown die bonding wires in a few LS's over time, just to test my soldering skills, I am amazed the bonding wire can take 2A. It's very fine when seen non-magnified by the domed lens.

I wouldn't recommend any more than 50% overdrive for the LIII if you plan to rely on the flashlight to actually work when that button is pushed, even if you are willing to accept less efficiency and have efficient heat transfer from LED to atmosphere (via flashlight). At 2A I'm waiting for the bonding wires to blow every time I turn it on.

Chris