Lifespan of a led /flashlight

RemcoM

Enlightened
Joined
Jan 14, 2012
Messages
631
Hi,

I have one Fenix TK75, for 8 months now, and i use it almost every day, but how long is the lifespan of the 3 leds in it?

Does it maintain the same beamintensity from the begin, when i had it brandnew? Or does the beamintensity reduce slowly?

Or does it stay the same, until all leds are dying?

Remco
 

Yoda4561

Flashlight Enthusiast
Joined
Jan 22, 2007
Messages
1,265
Location
Florida, U.S.A.
There are a few failure modes for LED flashlights.

Let's start with the LEDs themselves. LEDs have a projected average lifespan from the manufacturer, usually in the 30,000-100,000 hour range. If the light is properly designed and the driver/power supply circuitry doesn't fail it will slowly dim to a percentage of its new brightness by that xx,000 hour mark. Here's a list of those numbers for most cree LEDs http://www.cree.com/xlamp_app_notes/LM80_results . The projected time to dim to 80% when driven at the 3amp max current is 25,000 hours, if you use the light for 4 hours a day every day on high/turbo then you're looking at around 17 years before the LED dims to 80%. The human eye can have difficulty seeing a change in output smaller than 50% even when comparing two lights side by side. Using it at lower modes most of the time will extend the LEDs life to levels longer than most people will live. If a light overdrives the LED beyond its rated current it may shorten its lifespan to anywhere between a few thousand hours to a few seconds.



Now the 2nd mode of led flashlight failure, the driver circuit. This is the part most likely to fail other than switches on most lights. It's a compact circuitboard filled with parts that basically acts as a power supply and voltage/current regulator for the LED. Soldering quality, defective components, and poor circuit design can all lead to premature failure. Even potting compound to fix the vibration issues could cause problems if its thermal expansion characteristics aren't a good match to the operating temperature of the driver board. A good LED driver should last upwards of 10 years, past 25 years and a bunch of things come into play like environmental aging of the plastics and insulators used in the board components.


Edit: Oops, almost forgot. The real killer for LEDs in an otherwise solid light is HEAT. Especially on the smaller lights, LEDs still put out enough heat that without adequate air movement over the metal body of the light the heat builds up and can damage the LED. Many modern flashlights are being designed to put out the maximum output in the smallest size, which is fine as long as you only run the light in max for a short period at a time. With exceptional heatsinking to the light body the LED will usually be fine, but this is where alot of the cheaper manufacturers cut corners when building lights.(no thermal compound between the LED/body of the light, non-thermally conductive PCBs, etc) Back in the day when Fenix was just starting to get popular they had this issue with their lights on turbo mode, the quality was fine but the lights were just too small to handle the heat buildup when set to turbo and left sitting pointing at what someone was working at for an hour. That learning experience has led to lights having better thermal protection built into the drivers that automatically reduce the output when an overheating condition is detected.
 
Last edited:

TEEJ

Flashaholic
Joined
Jan 12, 2012
Messages
7,490
Location
NJ
Its mostly not an issue. They last so long (As explained above) that you will have many better options to use instead way before its end of life. Even if, 17 years later (~ 2030-ish?) you insist on using the antique light, you can upgrade to a newer/better LED for a few bucks.
 

RemcoM

Enlightened
Joined
Jan 14, 2012
Messages
631
So, my Fenix TK75, i have, and use now for more than a half year on turbo mostly, and on low, does not reduced its 92000 cd candlepower, as stated by the manufacterer? Can expect it, not started to slowly dimming?

But i have 2TK75 seen, and the other one has had a brighter hotspot, and low was considerably brighter than the low on the other TK75.

How is that possible?
 

idleprocess

Flashaholic
Joined
Feb 29, 2004
Messages
7,197
Location
decamped
So, my Fenix TK75, i have, and use now for more than a half year on turbo mostly, and on low, does not reduced its 92000 cd candlepower, as stated by the manufacterer? Can expect it, not started to slowly dimming?
With every hour of use, a LED will dim imperceptibly. Hour-to-hour, this dimming will not be detectable by all but the most sensitive laboratory instruments. Over longer periods of time, this will be detectable by lesser instruments and perhaps an A/B "then and now" photograph would reveal the difference.

This seems to be the fate for most near-point light sources. Incans vaporize their filaments and that gets deposited on the envelope. HID's dim to <70% over time as they vaporize electrodes. Don't know about short-arc but I suspect they're similar to HID.

But i have 2TK75 seen, and the other one has had a brighter hotspot, and low was considerably brighter than the low on the other TK75.

How is that possible?
There are numerous possibilities such as different flux bins on the LED's, different cell performance, and different product revisions. I would suggest that if you find the performance of your TK75 satisfactory that you not worry too much about the difference in output...

Flashlight manufacturers specify a certain minimum level of performance in their output specifications. A manufacturer like Fenix with multiple products might find it advantageous to buy a better LED flux bin to benefit some other products and give some of the TK75's they're making a free performance boost - with clever purchasing this can save the manufacturer a good deal of money. Pretty much no one can actually measure lumen output from a flashlight and would not much complain about it being brighter than advertised so long as it hits its claimed runtime.

How many cells were you running your light on, and what type? How about the other one? Running the TK75 on 4 cells vs two will halve the current draw from each set of cells. Better cells can supply more current with less voltage drop.

Fenix seems like it does quiet product revisions and perhaps the other one is a different revision with slightly different current/PWM levels for each mode. Alternatively, perhaps one of the two is regulating current/PWM out of spec causing the brightness difference.
 

StarHalo

Flashaholic
Joined
Dec 4, 2007
Messages
10,927
Location
California Republic
The first thing to fail on a serious duty flashlight will be the switch; if you're going to keep using this light for a long while, make sure you know where to get a replacement.
 

zespectre

Flashlight Enthusiast
Joined
May 21, 2005
Messages
2,197
Location
Lost in NY
In testing sample lights, some of which go back to roughly 2005, my personal experience has been that the vast majority of failures (not attributable to mishandling, drops, or other accidents) have been mechanical in nature Leading the way with switches, controls (digital and non digital), going next to "O" ring or other seal failures (allowing water or other contamination leading to failure) and occasionally actual structural failure (metal fatigue in threads, splits, cracks, or degraded plastics).

Out of the many lights I have tested I had some heat induced failures early on before I added a cooling fan to my test rig. Once I did that those failures stopped except in lights with very poor heat sink designs.

After that came a few (very few) failures due to the power source such as leaking cells, voltage protection issues, or one set of "runaway" CR123 cells that blew the head off a light. Obviously a problem with the cells and not the light, but regardless of the cause the light was still deadlined.

And finally.....waaaaaay past the tail end of the train, are the two or three lights I've had where the LED failed/dimmed for an undetermined reason. This behavior is so rare that I can only attribute it to some kind of manufacturing error. In some cases it was just a case of obvious reduction in output but in a couple the emitters in question made an abrupt and rather startling transition from blue-white to emerald-green and dropped 1/2 to 3/4 of their brightness in a couple of hours.
 

blah9

Flashlight Enthusiast
Joined
Mar 10, 2011
Messages
2,105
So, my Fenix TK75, i have, and use now for more than a half year on turbo mostly, and on low, does not reduced its 92000 cd candlepower, as stated by the manufacterer? Can expect it, not started to slowly dimming?

But i have 2TK75 seen, and the other one has had a brighter hotspot, and low was considerably brighter than the low on the other TK75.

How is that possible?


Is one light the newer TK75 with updated LEDs and the other is the older type? If that's the case then the newer one should be brighter in each mode too, right?
 

AnAppleSnail

Flashlight Enthusiast
Joined
Aug 21, 2009
Messages
4,200
Location
South Hill, VA
Since when do Maglites not have switches? I assume you might be referring to the mini mag, but it has an internal switch activated by the turning of the head on the light.

One of my favorite things about the old Eveready/Energizer 2xAA lights was the small size and self-cleaning switch. If you flooded the light with salt water you had to replace batteries. If you could EVER move the switch again, the slide would clean the corrosion off and work almost correctly with very few cycles of sliding.
 

idleprocess

Flashaholic
Joined
Feb 29, 2004
Messages
7,197
Location
decamped
One of my favorite things about the old Eveready/Energizer 2xAA lights was the small size and self-cleaning switch. If you flooded the light with salt water you had to replace batteries. If you could EVER move the switch again, the slide would clean the corrosion off and work almost correctly with very few cycles of sliding.

I believe that maglite is so proud of their self-cleaning switch on the C/D maglites that it's patented.
 

Etsu

Enlightened
Joined
Jul 1, 2013
Messages
783
I believe that maglite is so proud of their self-cleaning switch on the C/D maglites that it's patented.

Yes, it's certainly a good switch. Never had a switch problem with a C or D maglite. Only issues I've had is compressed head springs, and alkaleaks.
 

Canada

Newly Enlightened
Joined
Dec 29, 2012
Messages
22
You'll want another light before your The end of your TK ;)


Sent from my iPad using Tapatalk - now Free
 
Top