Have you ever seen the death of an appropriately driven (<1.5x max current) and appropriately heatsinked LED?

You can see the archived poll results on the Wayback Machine:

https://web.archive.org/web/20221220093741/https://budgetlightforum.com/t/-/72589

The key word being “appropriately driven.” :nerd_face:

We know that LED’s have a limited lifespan, but it’s in the order of hundreds of thousands of hours.

Just wondering if anyone has killed one of “natural causes.”

I’ve seen death of appropriately driven LED, which was inappropriately cooled… so yeah, that’s possible. (:

What is the definition of appropriately driven? Do you mean the stated amperage on the data sheet?
350mA nichia 219, 3A xml, 9A sst-90 etc?
If so, no.

Yup, I guess it has to be within datasheet specs. :stuck_out_tongue:

Wow, I’m surprised to see many have voted “yes.”

Any stories of how that happened?



XM-L on a star was attached to heatsink using just thermal grease while testing a driver (3A one)… thermal grease failed at being sticky enough, so the star detached from heatsink, and few seconds later LED was dead (I didn’t noticed it in time to do anything about it :frowning: )

Raccoon City, you just had to do that. :stuck_out_tongue:

Shadow, what color does the LED turn? I heard it turns blue or something when overheated.

When my SST-90 got detached, it melted the solder right off the star. Amazing how it didn’t damage the LED.

It’s tint changed to deep-blue indeed. No idea what causes it, but it’s fascinating to watch (for those 3-5 seconds that it lasts).

I've been lucky. I haven't seen the death of any LED. [quote=ryansoh3] Raccoon City, you just had to do that. [/quote] Yes, I had to. :evil:

EDIT: Fixed some poor grammar.

Yeah, but it was intentional. I fried a SSC P7 just for the heck of it. I poured some water onto it and it started to boil. :0

I’m not saying it can’t happen, but this poll may not be an accurate reflection of the presumed reliability of an LED. There are other factors such as “appropriate” heatsinking to take into account.

I for one have not experienced the demise of an LED while being driven within spec and with appropriate heatsinking.

Yeah, I forgot to add “appropriately heatsinked” as well.

I guess you can edit your vote if you want to. :slight_smile:

I said yes, but there’s a caveat to my response…

I’ve never seen a modestly driven and well heat-sinked NAME BRAND power led fail. And by fail, I assume we’re talking about being completely non-functional. I have seen them get old enough to experience a significant loss in brightness, but I still have yet to see one FAIL.

I have, however, seen lots of off brand .5w and smaller LED’s fail after only a few hundred hours. The ‘cheap’ no-brand 0805 and 5050 LEDs that are available on every chinese website do not last, even when being far UNDER driven.

PPtk

I have seen my oldest ones start to dim out and yellow slightly but never a properly setup and driven one completely burn out.

Having tortured lots of leds more than most, I must say that no...I've never lost one that was appropriately heatsinked, even at very high current levels. Then again, my idea of appropriate heatsinking isn't the norm.

Actually, only one comes to mind that I purposely killed: It was the star mounted xp-g that I did the emitter test on. Took it to 5amps (I think) and just watched it fade to blue then fizzle. A single tear soon followed. (I've got a strong suspicion that had it been directly mounted to my copper test post that wouldn't have happened.)

All white LEDs are actually royal blue on the inside. The phosphor coating turns the blue light into white. If the LED overheats, the phosphor coating can burn off and voila… blue-a-mundo.

I had several hundred of them die! The story is in the first post here: The Tao of Chinese LED Strip Lights

Never seen one die under appropriate usage conditions. When I was an intern at a lighting company and was working on the development of their first LED light I did manage to kill many LEDs, mostly by human error or not knowing what I was doing. Definitely dedomed a few, accidentally melted the dome with a soldering iron (tip directly into dome, not too much heat around it), tried soldering some Seoul P4s to star boards and didn’t realize that the part of the package under the LED needed to be reflowed as well, so saw some big time color and intensity shifts etc. Also didn’t know that the P4 was not isolated from the star so shorted some out, overall the P4 was an extremely shitty design. Also killed some LEDs hotplugging them because thats how we had to switch the lamp, we did devise a fix for that though.

So, like I said, in proper usage, no, in dumbassery, big yes.