What is really pushing a LuxIII to the limit(mA)?

With a good heatsink and the best bin LED (U? I think?) What is pretty much the limit for a LuxIII as far as mA's go.

i've heard of setups pushing 1.5A, maybe more. life expectancy is reduced but who is really going to get those thousands of hours anyway?

I remember reading somewhere that the bond wires in the led will start to melt if you start pushing more than 1.5amps into the led.

The bond wires will melt and fuse open in the 2.5-3A range - I've experimentally verified this myself.

Going past 1.5A, you are exceeding the maximum junction temperature for the device. High junction temperatures rapidly accelerate the aging process for LEDs, causing them to dim much faster than normal. A Lux III only lasts 20,000 hours to 50% brightness at 1A, at 1.5A that number is probably more like 5000 hours or less. Plus, you have no idea how reliable the thing will be at that drive current. You may be using it one day and *poof* it dies an early death from too much current. Then you're stuck with a dead light. A light at 1/2 the current still produces more light than a dead one.

Some people do use Luxeons for 1000s of hours - I am one of them, using them for general lighting purposes, so sane drive levels are part of my requirements to achieve reasonable reliability.

In some of the early directly driven lights the Lux 111 were pushed as high as 16oo ma, the FT-3C was one of them

evan9162 said:

The bond wires will melt and fuse open in the 2.5-3A range - I've experimentally verified this myself

Click to expand...

What type of heatsinking were you using at the time or it doesn't matter in this case?

I've got an L2P that has a lux III and a different driver in it and if i measure the current from the negative battery terminal to the body i get 1.95A I've checked this on a calibrated multi meter as i work in the electronics industry i have access to all sorts of equipment but don't know alot about electronics(that could be a bit of irony there) The light gets hot very quickly but is really bright.

eebowler said:

What type of heatsinking were you using at the time or it doesn't matter in this case?

Click to expand...

I used a peltier to freeze the luxeon down to -20C. The junction temperature at time of failure was about 140C - over the limit, but not exceedingly so.

only if you cool the thingy with liquid nitrogen for quite long and run it to 2.5amp(while still being cooled by the LN), it won't die but go past 2.5 amp and the bond wire will melt ... becuase the gold wire isn't really being cooled alot...

this is my theory!!! i repeat this is my theory!!!

leeleefocus said:

I've got an L2P that has a lux III and a different driver in it and if i measure the current from the negative battery terminal to the body i get 1.95A I've checked this on a calibrated multi meter as i work in the electronics industry i have access to all sorts of equipment but don't know alot about electronics(that could be a bit of irony there) The light gets hot very quickly but is really bright.

Click to expand...

Since the L2P has a driver circuit in it, you have no idea how much current is getting to the Luxeon - you need to disconnect one of the Luxeon's power leads and insert your meter in series with it in order to measure the current to the LED. And since you're using a driver, you need a meter that can measure true RMS at DC voltages to get the most accurate measurement possible.

evan9162 said:

I used a peltier to freeze the luxeon down to -20C. The junction temperature at time of failure was about 140C - over the limit, but not exceedingly so.

Click to expand...

The reason I asked was because I believe that the failure point of a Lux III LED depends largly on how well heat is removed from the die and not ONLY on how much current you run through it. If you're talking about an emitter only, failure may be at 600mA. A Lux III LED on an 'O'sink or hotlips may fail close to 2A. If evan cooled his LED with dry ice (-70deg C) the LED may fail at a higher current.

LEDs can take a lot but the higher the die temperature is above the tested 25 degC, the shorter the life of the LED. I'm sure a 'well heatsinked' Lux III LED can handle 2A at room temperature but, the more important question is, for how long?

Can't exactly remember the thread/s but I believe in real world practical applications, the life of a Luxeon LED is not the stated 50000 hrs to 70% brightness, but closer to 20000 hrs or less.

This failure point (bond wires melting) is pretty much completely independent of heatsinking. Much like a fuse blows open at the same current regardless if it's -20 C or +60C, the bond wires will fuse open around 3A. I supercooled the luxeon to remove excessive heat in the junction as the failure mode, and to expose bond wire failure as the exclusive failure mode.

So it is pretty much dependent only on how much current there is, and not dependent at all on the heatsinking.

:thinking: Ok. That makes sense. The fuse/bond wires anology was good.