Luxeon 1 died on me, too much heat or defect? Hmm.

[IsaacHayes]

I had a Q2K 1W in my dome light for a while. I had this along with a LuxeonIII RB. I figured the switch broke on me going to the white one because if I jiggled it, it would light up. Well I looked at it today and seems the switch is fine, but the luxeon isn't. Now the Royal Blue lux is doing just fine, but the white isn't.

I saw a small black "dot" of dirt or something on the phosphur on the luxeon before I ever used it. I'm wondering if this was defect or something, perhaps the die was cracked and ruptured upwards?? Or maybe it was just dirt in the silicone gel and un-related. The white 1w was driven at around 800 ma, yes an over drive but I wasn't expecting much extra lumens from it, that's just what I ran the RB at via a constant current setup.

I know people have ran 1W's at 750ma with no problem, but does that hurt them so much more because they aren't "beefed" up like the LuxIII is? This white was used very rarely, the RB had a TON of time on, getting way hotter than the white did. The white was used only a few times, and when that was it was for maybe 10seconds at the most!!

The bond wires look OK under a pocket 2aa microscope (minolta), but what else could have failed? The chip? Woulnd't it just get dim and not just die? Anything interesting I can do besides take it apart, or microwave it (j/k). /ubbthreads/images/graemlins/smile.gif

 

[IsaacHayes]

Re: Luxeon 1 died on me, too much heat or defect?

BTW this is the first Luxeon I have "killed"!!! I've done a ton of mods, but never killed one. Well this one didn't die directly from me handling it. It worked fine for a while. So not sure if I really killed it.. /ubbthreads/images/graemlins/wink.gif

 

[evan9162]

Re: Luxeon 1 died on me, too much heat or defect?

Can you apply power directly to it, bypassing all the circuitry/wiring?

 

[IsaacHayes]

Re: Luxeon 1 died on me, too much heat or defect?

evan, I haven't tried to yet other than using my DMM set to check resitance, as that has enough juice to light up any luxeon at a nice glow of the die. I've heard of people putting reverse voltage to them for a short burst and they fix them selfs?? Tommorow I'll play with it more, perhaps send it some power while in a totally dark room to see if there is any faint glow at all..

 

[evan9162]

Re: Luxeon 1 died on me, too much heat or defect?

Use the diode test function on your DMM - that should work too.

 

[The_LED_Museum]

Re: Luxeon 1 died on me, too much heat or defect?

If you use the diode test function of your DMM and nothing happens, reverse the leads and try again.

 

[evan9162]

Re: Luxeon 1 died on me, too much heat or defect?

Isaac,

I've destroyed a few luxeons in my time. The main reason for all failures was bond wire fusing. I was able to "repair" a few by popping off the dome cap, and soldering a jumper wire from the power lead directly to the remaining gold bond wire.

In one of them that I repaired (Lux I white), it did not produce any light with less than 300mA of current, and it was much dimmer than it should have been, so there was definiely some damage to the die as well. I wouldn't discount that possibility in your case.

 

[IsaacHayes]

Re: Luxeon 1 died on me, too much heat or defect?

Evan & Craig: The diode function on my DMM doesn't cause it to light up. But I'll try reversing it and see what happens.

Evan: So why did it fail? Just got too hot or too much current for a 1W? I wonder what they did different to a 3W to make them run happy at high current. Just more solder blobs on the back of the die?

 

[Bullzeyebill]

Re: Luxeon 1 died on me, too much heat or defect?

I have run well over 1 amp to lux 1's (Q bins) with no ill effects, though best current for great output is around 700mA's. Lux III's handle heat better, and can be very efficient at 1 amp. Not sure why this is so, though I do notice that the lux III stars have a thicker heat sink, just like Lux V's.

Bill

 

[evan9162]

Re: Luxeon 1 died on me, too much heat or defect?

Isaac,

How often is your white luxeon switched on? Ever have it on when you started the car? Switch to the white while the dome light was on?

I think I've learned that it's better to not attach a luxeon to a powered open circuit regulator, since the regulator is turning its pass transistor on full, and will give the load a full blast for a few microseconds while the regulation kicks in.

When I attach a luxeon to my test current supply, it's set to its lowest current setting ( a few mA through a resistor) so initially, the output is at just a few volts, rather than the full volage from the power supply. Then I can switch between any current that I need.

It's possible that when switching to your white luxeon, it was hit with multiple-amp pulses. Because your switch was likely break-before-make, it would be open circuit for that little bit between contacts, and the output would go full to near 12V, then when contact is made, would produce a tiny but high current surge (a few microseconds tops) until the regulator ramped down the current.

Looking at the LM317 datasheet, it looks like it takes 5us for the regulation to stabilize. During this time, your luxeon could be hit with a couple amps easily.

However, I don't know if this would be enough to kill the thing or not...

 

[IsaacHayes]

Re: Luxeon 1 died on me, too much heat or defect?

Hmm, I can't recall if I did or not. The white wasn't on unless I needed to see inside the car. The RB was left in the ON positin while starting the car many times, once you turn the key to start the dome light will slowly dim out.

Someone else on this board has 4 luxeons wired in series as a dome light. So at ~14.4v they'd get ~ 350ma. Would that be any safer because it wouldn't rely on a regulator? Would adding a diode to the luxes be a pretty bullet proof design?

I might go that route, as I'm thinking of taking off the RB emitter (carefully!) and putting it on an o-sink to charge my GP..

 

[evan9162]

Re: Luxeon 1 died on me, too much heat or defect?

Just putting them in series without a regulator isn't a good idea. The power system in a car can spike and surge quite a bit. Without a regulator, it'd be easy for you to pop one or more of those luxeons if the dome light were turned on while the car is running.

A diode in series is a good idea to protect against reverse surges that can happen.

 

[IsaacHayes]

Re: Luxeon 1 died on me, too much heat or defect?

So would it still be unsafe with a diode in the mix and 4 luxeons? As forward surges could occour correct?

 

[evan9162]

Re: Luxeon 1 died on me, too much heat or defect?

yep. I've been looking at a switching regulator designed for auto applications. It states that it can take up to 60V - and said in the datasheet that the 60V capability was so it would survive in an auto environment.

 

[TORCH_BOY]

Re: Luxeon 1 died on me, too much heat or defect?

The same thing happened to me a while back, I was running
a 1 watt at 500 ma with minimal heat sinking, mine just went
out dull with a violet glow

 

[IsaacHayes]

Re: Luxeon 1 died on me, too much heat or defect?

Looking at it with an 8x maginfier and ~200 lumens from my tri-mag, I can see that the phosphur looks like it was scraped off the die in an area. Sort of like an eruption. This looks like a black spec on the die with normal vision.

This was always there though so I'm not sure if it caused it to fail, or doesn't matter... Hmm....