UPDATE LED Life Test at 2 Months

[Mike Harrison]

On Wed, 07 Jul 2004 07:25:35 -0700, "Watson A.Name \"Watt Sun - the Dark Remover\""
<NOSPAM@dslextreme.com> wrote:

Rune Christensen wrote:

Hello

I have found this link on eBay
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=2020&item=5106449490

He claims that the white leds in the flashlight can last 100,000 hrs. That
means 24/7 in over 10 years.

What's claimed and what's actually going to happen are two different things.

...and what do they mean by 'last'... it may still produce _some_ output after that time, but how
much?

 

[Tim Shoppa]

"Rune Christensen" <rune.christensen@adslhome.dk> wrote in message news:<40eba3f5$0$225$edfadb0f@dread12.news.tele.dk>...

Hello

I have found this link on eBay
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=2020&item=5106449490

He claims that the white leds in the flashlight can last 100,000 hrs. That
means 24/7 in over 10 years.

"Can last 100,000" hours is awfully vague. Is that a lifetime or a MTBF?
(The two are not the same... the MTBF of a US Soldier in Vietnam was
280 years, for example.) Under what conditions?

Agilent and Nichia have some really good documents on their websites
about how they test their LED's. They probably don't apply to E-bay
sellers .

Tim.

 

[Mark Mcmillan]

"Don Klipstein" <don@manx.misty.com> wrote in message
news:slrncf8qum.l7n.don@manx.misty.com...

In article <onv6f0pef668eok0tekrei72hcc15flchi@4ax.com>, The Real Andy
wrote:

Today I read an article that claims the die heat contributes to colour
change and long term reduction of light output. So I gots to thinking,
if you pulse the led at some reasonable rate, would this reduce the
die temperature and prolong the life of the led? Maybe you could bung
two of you're cheap hong kong leds next to each other, on pulsed at
the lowest duty cycle possible to prevent visible flicker, and the
other on dc.

White LEDs tend to require a higher average current and have a higher
voltage drop (for equal average klight output) when pulsed than when
operated continuously. Unless the average current is less than a few
milliamps, you get more light and less heat with continuous operation than
with pulsed operation at a frequency high enough to apear continuous.

By and large, only LEDs that are more efficient at higher instantaneous
current benefit from pulsing. This was especialy true of LED digital
displays with GaAsP on GaAs substrate, where a segment had an average
current of only a few milliamps but the efficiency was maximized at
instantaneous currents of 50 milliamps or more, often at least 100 mA.
Many people were not aware of this nonlinearity of those LEDs and believed
that the benefit of pulsing was due to a quirk of human vision.

Some LEDs, namely at least some InGaAlP ones, have a degradation mode
that is a function of temperature and duty cycle. I suspect this is from
some sort of diffusion of an ingredient from where it belongs to someplace
else and dependent on electric field around the boundary or zone that the
diffusion occurs across. Maybe a boundary between different layers
(having/lacking whatever diffuses) should be sharp but gets "blurred".
LEDs with that chemistry, primarily at higher temperatures and lower
average currents, can have a life extension from pulsing. One
manufacturer, Agilent, publishes an application brief where they encourage
pulsing of their LEDs with that chemistry if the average current is less
than 10 mA, and recommends instantaneous currents 10-100 mA (but average
current not exceeding 30 mA).

I discuss pulsing LEDs in an attempt to make them brighter (or appear
brighter) in a web page of mine, http://www.misty.com/~don/ledp.html

By and large, if the average current is already close to the maximum
rated constinuous current, there is at beast little to gain from pulsing.

- Don Klipstein (don@misty.com)

I remember being told once in the case of normal LEDs (no phosphor) that
over time (powered) and power cycling and at higher currents that, micro
fractures occur in the die and as these accumulate they prevent the light
from escaping from the die hence less light output. True or not I do not
know.