maximum current

[degarb]

I blew out a rebel today by cranking up over 2 amps over 6 volts. Wasn't looking at meter, just quickly testing brightness. What suprised me was how quickly it blew in a second-flash of blue and gone. I was expecting that heat was culprit for all led blowouts so would be something that would take several seconds, but now see merely cranking up to some number can pop them out.

I have one home made light that uses an xr-c with variable pot. Xr-c are rated 500 milliamp max. I always run at least a pinch down, somewhere under the .5 amp. Question: how long would it take at 580 milliamps to burn out the xr-c? What grace milliamps do the manufacturers build in?

 

[R33E8]

It depends on the operating temperature.. It wouldn't last long if it ran at 150C compared to 20C at 500mA..

 

[DaFABRICATA]

where did you get the emitters?

 

[saabluster]

I have one home made light that uses an xr-c with variable pot. Xr-c are rated 500 milliamp max. I always run at least a pinch down, somewhere under the .5 amp. Question: how long would it take at 580 milliamps to burn out the xr-c? What grace milliamps do the manufacturers build in?

:poof:You have learned a valuable lesson. I have done some extreme testing with the XR-C taking it up to 2A. It still did not blow even at 2A(Cree is just awesome) but it stopped gaining in lumens before that. After pushing it past 1700mA instead of gaining in output it started to go down. The main thing with LEDs is to make sure your thermal path is good. I soldered the XR-C to a solid piece of copper. As long as you have a good way for the heat to excape from the light that XR-C would probably outlive you even at 80mA over spec.

 

[2xTrinity]

I blew out a rebel today by cranking up over 2 amps over 6 volts. Wasn't looking at meter, just quickly testing brightness. What suprised me was how quickly it blew in a second-flash of blue and gone. I was expecting that heat was culprit for all led blowouts so would be something that would take several seconds, but now see merely cranking up to some number can pop them out.

Current through an LED will increase exponentially as you incerase voltage. That means a small increase in voltage can mean the difference between optimal drive current, and instant death. For the sake of simply testing your LEDs, I'd recommend using a power supply with current limiter dial, or if you don't have that option, at least adding some resistance in series before testing.

 

[TigerhawkT3]

6V should be WAY over 2A. I'm not surprised it :poof:ed.

BTW, if you actually had been testing it at 2A, you'd need to have it mounted to a good heatsink anyway, or it would :poof: as well.

 

[Oznog]

Yeah if you put it across 6v with no ballast resistor you pulled way more than 2A. 2A should be in the 4v-ish range. And I'm sure it can sustain 2A for hours if heatsunk very well. Might have detectable fading within tens of hours or even within an hour.

Rebel is rated 350mA nominal, 700mA max continuous, I think the spec sheet goes up to 1A.

If you put 350mA on it with no heatsink, it'll overheat in seconds.

 

[sailor]

I have one home made light that uses an xr-c with variable pot. Xr-c are rated 500 milliamp max. I always run at least a pinch down, somewhere under the .5 amp. Question: how long would it take at 580 milliamps to burn out the xr-c? What grace milliamps do the manufacturers build in?

I don't know, but I'm about to make an overvolting experiment.

I "inherited" a 3 cree led GU10 fixture, similar to one sold by DealExtreme.
I plan to use it as a bicycle headlight bypassing AC circuitry.
The Crees inside look like xr-c. The original AC driver limited LEDs to about 320mA.
I was in my "we don't need no stinking electronics there" mood yesterday, so was curious how could I direct drive these three (connected in series).
So with freshly charged 7.2V (nominal) NiCd battery pack it has puny light output ( ~ 50mA current).
Things looked better with 8.4V - 170 mA, tapering to 140 after a while.
So decided to go for 9.6V.
890mA on a freshly charged pack is clearly outside of specs for these, but I decided to give it the a try. Heat sink temp was probably in 60 Celcius range (140F).
Since it was a small battery pack ( 600mAh NiCd AA) the current quickly tapered down reaching roughly 480 mA after 10 minutes.

I'm certainly not looking for 50k hours from these LEDs, I would be happy with few hundred hours (I use the bike lights slightly less than 100 hours per season).
Time will tell.