The Neon Bender's BBS!
The Archive no longer allows posting but search of hundreds of previous posts are available
If you have any questions please
contact us.
[Please Read]
| |
Re: LED Article
Posted By: SVP Neon Equipment In Response To: Re: LED Article (Boyd)
Date: Thursday, 11 February 2010, at 1:19 p.m.
Thanks for the info. May I ask where the link is to this article? I seems to be garbled in some places from the copy and paste attempt.
Concerning Dr. Kahn, "As president of LED Lighting Technologies" speaks volumes. It is like expecting the pro global warming climate researchers who are getting paid big funding to tell everyone they were wrong; they would lose their funding, so they will never say it.
That fact that she has finally put in print that the 100,000 hour life span of LED's is not true means nothing at this point. Everyone, including the LED mfgs. have known this for a long time. It is old news, which does nothing for her credibility to admit it after it has been proven time and time again to be a false claim.
There are a lot of other comments in the article that show bias also. Unfortunately I don't have the time to go into them right now. Perhaps someone else will.
Mark
> As president of LED Lighting Technologies,
> Dr. M. Nisa Khan educates the lighting
> industry and consumers about LED lighting.
> She has a bachelor's degree in physics and
> mathematics, and master's and Ph.D. degrees
> in electrical engineering. Email
> heratnisa.khan@iem-asset.com How Long Do
> LEDs Last?
> Nisa explains the "second-half"
> syndrome> Popularity begets popularity. Particularly
> today, because of the countless electronic
> communication systems that quickly dispatch
> popular news and information across the
> globe. Ask Tiger Woods. Today, the news
> travels - and aggrandizes - quickly.
> In the scientific arena, popularity
> describes a matter, concept or idea that
> races across a given field, and later,
> perhaps, the nation or world. However, with
> science, a rise in popularity does not
> indicate a rise in validity.
> For example, when a colleague recently
> remarked that LEDs last forever, I
> lightheartedly, but truthŽfully said only
> half will last a portion of forever. My
> reason? LightingŽindustry professionals
> comprehend that a specific lamp series may
> be publicized to have a three-year life, but
> also recognize that only half the lamps are
> expected to last that long. The remainder
> may last only a fracŽtion of three years.
> Some call this the "second-half'
> syndrome.
> The "LED lamps operate intermiŽnably or
> almost forever" belief unfurls because
> casual observers read LED manufacturers'
> claims of 100,000-hour lifespan.
> Unfortunately, such claims and beliefs have
> become popular, but nothing could be further
> from the truth. A classic falibility example
> is the partially darkened, LED-arrayed,
> traffic lights we sometimes see.
> Too often, the "second-half"
> failure syndrome frustrates signage and
> luminaire makers, especially if the lamps
> burn out within months or days of
> installation.
> If prosaic traffic-signal LEDs experience
> lifespan problems, what can we expect from
> modern, but not-se-well-tested LEDs,
> especially the much-lauded, white ones?> First-half lamps
> For many reasons, some scientific
> communities, as well as many ordinary
> people, believe LED lamps last a long time,
> or have such potential longevity. One reason
> is advertising claims; another is that LED
> indicator lamps on early electronic gear
> still produce light. Further, credible,
> laboŽratory "accelerated-lifespan"
> tests of current LED lamps extrapolate
> lifetimes which should exceed 100,000 hours.
> However, lighting professionals' views
> differ from that of the public. They
> recognize, for example, that a
> manufacturer's production run of two
> million, 100,000-hour lifespan, LED lamps
> produces roughly one million that will
> endure. Here, lifespan is akin to "Mean
> Time Between Failure" (MTBF). Some
> describe the long-life lamps as
> "first-half lamps," because they
> prevail.
> Therefore, assessing such a lamp-life
> characteristic shows that LED manufacturers
> shouldn't claim a consistent lifespan for
> generalŽpurpose, LED lamps, such as those
> sold for signage or illumination
> applications.
> Preferably, manufacturers would test a
> production run of general-purpose LEDs for
> all relevant light characteristics and
> announce that half, at least, had endured as
> expected. Meaning, no light characteristics
> had dropped below published specifications.
> This is rarely the case.
> To justify advertised lifetimes, diligent
> manufacturers perform certain equivalent
> tests via sampling, as well as accelerated
> temperature and aging tests. Such claims may
> be valid for smaller quantities, say for
> clusters of one hundred, rather than for
> tens of thousands, or millions.
> Truth is, equivalency tests and samplings
> don't always scale, because Significant,
> temperatureŽrelated, performance variations
> exist among LEDs built from the same
> compound semiconductor epitaxial wafer, as
> well as those from wafer batches. Other
> process variations may also cause
> perforŽmance variations within the entire
> lamp manufacturing line, but to lesser
> degrees.> Connections matter
> Aside from manufacturing-line variŽations,
> consider that other LED lifeŽtime and
> failure points may affect different LED
> types and, of course, the applied driving
> methods can affect endurance. Also, higher
> ambient temperature, humidity and operating
> current cause LEDs to degrade faster.
> Generally, two types of LED lamps exist in
> the marketplace: high-power units that use
> elevated current (200mA or higher) or
> low-power units which use modest currents
> (tens of milliŽamps). Designated LEDs have
> their own, current-voltage requirements and
> corresponding light-current
> charŽacteristics. Expectedly, all diodes
> burn out if you apply excessive current,
> meaning, you shouldn't direct high current
> into low-current lamps.
> Because excessive current flow will burn
> them out, LEDs need to be properly wired. If
> you accurately choose your voltages,
> resistors and currents, both series and
> parallel connections are feasible. If you
> choose parallel, all the LEDs must be
> sufficiently identical in relation to
> operating current, resistance and diode
> turn-on forward voltage (transistor load
> time, for practical purposes). However,
> without the proper binning of equivalent
> LEDs, such matching is difficult to achieve.
> Red, blue and white LEDs inherŽently require
> different voltages; therefore, parallel lamp
> connections may require each LED to have its
> own series resistance, to limit the current
> passing through the diode.
> High-power LEDs (but not necesŽsarily
> high-brightness LEDs) require high current,
> and this causes them to operate at a higher
> diode-junc-
> temperature, Wl1iCl1, in [Urn, leads to
> faster aging and a shorter lifetime.
> For their various uses, sign manufacturers,
> and others, require white-light LEDs to
> produce high lumen output. And, for decent
> illuŽmination of channel letters, cabinet
> signs and electronic displays, signŽtype
> lamps should consistently produce several
> hundreds of lumens throughout their
> predicted lifespan. However, unlike
> tradiŽtional indicator lamps, signmakers
> require their lamps to correspond in color
> rendering and temperature during their
> entire lifespan. Such requirements
> inevitably shorten the lamps' lifetimes.
> Most manufacturers of high-power, white LEDs
> estimate a lifetime between 30,000 to 50,000
> hours, to maintain 70% of initial lumen
> level. This estimate assumes constant
> operŽation near 350mA of constant CU1Tent
> and a consistent junction temperature not
> higher than 900 c.
> I.azproved tech.nologles Improving
> technologies have produced longer-life LEDs
> that, to some extent, tolerate higher drive
> currents and operating temperatures.
> Presently, certain manufacturers offer LEDs
> rated for 100,000 hours while operating at
> over 700mA with junction temperatures up to
> 1200 C. I suggest you consult with these
> manufacturers, to see if such LEDs are
> available and cost effective, at high
> volume.
> As the LED lighting industry develops, the
> engineers consider both 70% and 50%
> lumen-mainteŽnance levels as lifespan
> definitions, but for different applications.
> For example, a traditional lightbulb ends
> its usefulness when the filament breaks and
> the light ceases. Oppositely, an LED lamp's
> efficiency drops over time, which causes the
> light output to diminish gradually. When it
> falls below 70% (or 50%) of the original
> output, the design engineers say it has
> expired.
> mceresCfl'igtf, ate Kve-'LYg'C' (7c?:TaI7
> can't judge when the LED lamp has reached
> this point, which is the drawback of this
> standard.
> To better understand LED lifeŽspans, ask
> your LED lamp provider for lifespan data
> sheets that describe the gradual behavior of
> the lamps' luminous-output degraŽdation over
> time and under specific operating
> conditions.
> Also ask them to provide test data that
> relates to the specific LED chip or die, or
> junction temperaŽture, over the lifespan
> test period. Not all LED manufacturers
> publish such data, but such information can
> help you calculate a specific LEDŽbased
> sign's life.
> In my next column, I'll write more on
> finding application soluŽtions, and LEDs'
> thermal designs
> and analysis.
| |
Neon Bender's BBS! is maintained by Kenny with WebBBS 5.12.
This page Copyright © 1996-2005 Krypton Neon