Tech Q re LED Life

[DonK]

I notice all the manufacturers quote 50,000 hrs as the life of a LED. I just wondered how that varied wth different types of use

Is that 50,000 hrs running at peak turbo output ?

Is life reduced if you use strobe a lot ?

Is life reduced if you do a lot of momentary on operations ?

Just thought that the high temps of turbo mode may reduce life, and the thermal cycling of strobe and momentary on may reduce life due to thermal cycling.

Anyone seen any research on this ?

Anyone had an LED fail ?

 

[DFiorentino]

Read each manufacturers datasheet for each given LED. They typically state lumen maintenace projections or test conditions for their ratings. Every LED is different.

-DFy

 

[DonK]

thanks, looks like there is no set standard on this then

 

[wrf]

I notice all the manufacturers quote 50,000 hrs as the life of a LED. [...]

Someone posted quite candidly regarding this, but I can't find it. Basically an estimate of some LED lifetime (to some % of output) at a given temperature and drive current gets quoted by flashlight manufacturers without regard to actual operating conditions.

[...]Is that 50,000 hrs running at peak turbo output ?[...]

Unlikely.

[...]Anyone had an LED fail ?

Some threads are out there. Here is one.

 

[idleprocess]

thanks, looks like there is no set standard on this then

No, no real standard, but "power LED" tests tend to be far more real-world than commodity low-power LED's in age-old 3/5/10mm through-hole packages which often claim 100k hours life. I seem to recall seeing the old Lumileds standard perpetuate itself to other manufacturers about 5-6 years ago - I want to say it was something like heatsink temps of 50C, continuous design current (then 350mA for the Luxeon I), standard "room temperature", and 70% rated lumens at end-of-test.

While performed within a lab under conditions that simulate normal wear and tear under accelerated conditions (usually done with >100% maximum drive current under set heat/humidity/duty-cycle/thermal-design conditions), this testing is of some value since it's used to determine the manufacturer's warranty and/or liability should their product prematurely fail under otherwise good conditions.

 

[pavithra_uk]

Is that 50,000 hrs running at peak turbo output ?


Actually 50,000 hrs "word" is defined for 70% brightness reduce. LED will last much long time.
I think its @ standard current/temp condition. Anyone never actually run LED 50,000 hrs.

Is life reduced if you use strobe a lot ?

Will be.

Is life reduced if you do a lot of momentary on operations ?

No.

Just thought that the high temps of turbo mode may reduce life, and the thermal cycling of strobe and momentary on may reduce life due to thermal cycling.

High temps reduce life/reduce light output. There is no idea about thermal cycling (LED chip level)

Anyone seen any research on this ?

Just google it.

Anyone had an LED fail ?

yes I had failed hundrads of 5mm LED, few 1W~5W LEDs

 

[idleprocess]

Is life reduced if you use strobe a lot ?

Will be.

I have a recollection of reading a number of manufacturer's data sheets claiming that strobing <=200% maximum current for <=100ms at a duty cycle of <=10% would not effect operating lifespan. Exceed those parameters and lifespan suffers. Perhaps you have some hands-on experience in these matters; I've neither the EE knowledge to build the circuits nor desire to overdrive costly LED's.

Strobing at <=100% of maximum current should have no particular effect on rated life since there's no "activation penalty" for LED's.

 

[DonK]

thanks for the input - looks like some issues are out there