Chip level LED failure mode

[JohnR66]

I'm aware that the epoxy can discolor blocking the light and the phosphor
can degrade on white LED's.
But for this thread, I am curious about how the LED die itself fails.
Thanks, John

 

[Don Klipstein]

Bond wire fusing out as result of over current is popular one, even
your humble 5mm LED can have half a dozen bond wires to the die
nowadays

Some sort of film failure of top electrode can seem to come into play,
random failure land where it just turns Dark Emitting Diode, DED.

Guessing on contaminats in wafer or coating for the gradual blackout
cases where the LED dims and the die ends up looking significantly
blackened, relevant more to low end commodity LEDs.

Sure someone has more insight.

One thing I have seen go wrong is electrostatic damage to GaN and InGaN
LEDs. Any reverse breakdown appears to result in a partial short that
typically conducts a few mA, occaisionally 10's of mA at the typical
forward voltage drop.

- Don Klipstein ([email protected])

 

[Don Klipstein]

On Jan 4, 10:03 pm, [email protected] (Don Klipstein) wrote:

I have a limited electronic background, so I wanted to clarify what
you are saying here. If I have say 6 InGaN leds in a series string
and one of them is damaged by static causing a reverse breakdown, it
will only partially short, and so I would still be able to run the
whole string, but the damaged LED wouldn't light up fully or would
something else happen?

It may light up less than fully, or may fail to light up at all if the
partial short is a more conductive one

(dropping less voltage than it takes to make the LED glow visibly at all
using the amount of current being used).

Is there a way I can check an LED that I suspect might have been damaged
via static?

Undamaged InGaN and GaN LEDs normally glow easily visibly at .2
milliamp, maybe except for violet and UV ones. At .5 mA, they usually
glow with a brightness typical of LED indicator lamps.

In my experience so far, static-damaged ones don't glow at all at .5 mA,
and I expect static-damaged ones to consistently not glow and to show
unusually low voltage drop less than 2 volts (usually some fraction of a
volt) at .2 mA.

- Don Klipstein ([email protected])

 

[JohnR66]

It may light up less than fully, or may fail to light up at all if the
partial short is a more conductive one

(dropping less voltage than it takes to make the LED glow visibly at all
using the amount of current being used).


Undamaged InGaN and GaN LEDs normally glow easily visibly at .2
milliamp, maybe except for violet and UV ones. At .5 mA, they usually
glow with a brightness typical of LED indicator lamps.

In my experience so far, static-damaged ones don't glow at all at .5 mA,
and I expect static-damaged ones to consistently not glow and to show
unusually low voltage drop less than 2 volts (usually some fraction of a
volt) at .2 mA.

- Don Klipstein ([email protected])

Interesting. I performed a "fade" test of 3mm and 5mm white LEDs of various
brands by driving them at 30ma for 192 hours. Most showed lumen depreciation
of varying amounts. One developed a condition where it wouldn't emit light
until current was raised above a couple ma. It was conducting at lower
current, just no light. I wonder if it had formed a conducting channel of
sorts like a static damaged LED?

FWIW, In another test, I reverse biased a white LED at 30 volts in series
(with a 1k Ohm resistor to eliminate possible fireworks). Current remaind at
0. No damage to LED.