Failure mode of LEDs in overcurrent situation

[Andrew T.]

I'm curious about what happens to an LED when it is exposed to a
moderate overcurrent situation.

This question came about as I was troubleshooting some small white LED
lights I built using 3 5mm LEDs and a current limiting resistor to run
on 12Vdc (supplied by a lead-acid battery charged by solar). After some
time, many of these strings would start to flicker or blink.

I was about to post a question about that, when I redid my resistor
calculation. I had originally calculated that a 100 ohm resistor
would be sufficient to keep the current around 20mA (assumed 12.5V -
10.5V total LED drop). But on recalculation, that would result in 30mA
(the absoulte maximum rating) at 13.5V, the low end of charging range.
Adding another 47 ohm resistor seems to have helped.

So now I'm curious what happens inside the LED when the current
exceeds the absolute max current by a modest amount (i.e., not the
obvious overcurrent of leaving out the resistor entirely). The LEDs
I'm using seem to return to normal operation when the current is
reduced, and seem to produce a momentary open-circuit condition during
overcurrent, presumably from overheating.

Anyone know what's happening to the semiconductor material in this
situation?

Thanks,
--Andrew

--
Andrew Turnquist, Short Tract, New York, USA (USDA Zone 5)
(Remove numbers and .invalid for email address)
"Do what you can with what you have where you are." -T Roosevelt

 

On Friday, 6 December 2019 19:08:00 UTC, Andrew T. wrote:

I'm curious about what happens to an LED when it is exposed to a
moderate overcurrent situation.

This question came about as I was troubleshooting some small white LED
lights I built using 3 5mm LEDs and a current limiting resistor to run
on 12Vdc (supplied by a lead-acid battery charged by solar). After some
time, many of these strings would start to flicker or blink.

I was about to post a question about that, when I redid my resistor
calculation. I had originally calculated that a 100 ohm resistor
would be sufficient to keep the current around 20mA (assumed 12.5V -
10.5V total LED drop). But on recalculation, that would result in 30mA
(the absoulte maximum rating) at 13.5V, the low end of charging range.
Adding another 47 ohm resistor seems to have helped.

So now I'm curious what happens inside the LED when the current
exceeds the absolute max current by a modest amount (i.e., not the
obvious overcurrent of leaving out the resistor entirely). The LEDs
I'm using seem to return to normal operation when the current is
reduced, and seem to produce a momentary open-circuit condition during
overcurrent, presumably from overheating.

Anyone know what's happening to the semiconductor material in this
situation?

Thanks,

Flashing I'm told is caused by wire bonds detaching & moving due to thermal expansion. If so I suspect at some level of reduced i a flashing LED ought to work again, just at reduced output.


NT

 

[Andrew T.]

On 2019-12-06, tabbypurr@gmail.com <tabbypurr@gmail.com> wrote:

Flashing I'm told is caused by wire bonds detaching & moving due to
thermal expansion. If so I suspect at some level of reduced i a flashing
LED ought to work again, just at reduced output.


NT

Thanks. I'd never really considered the actual wire bonds themselves.
That would make sense. Of course, now I wonder what the effect of a
sudden physical shock would be, if that might knock the bond apart....

--Andrew

--
Andrew Turnquist, Short Tract, New York, USA (USDA Zone 5)
(Remove numbers and .invalid for email address)
"Do what you can with what you have where you are." -T Roosevelt

 

[Phil Allison]

Andrew T. wrote:

----------------

I'm curious about what happens to an LED when it is exposed to a
moderate overcurrent situation.

** I have seen a beautiful example for this with a Boogie valve amplifier that had a bad manufacturing error.

The amp had 10 Vactecs inside, LED / Cds cell devices for channel switching but the screen printing on the PCB had all the resistor values for the LEDs wrong by a factor of 10. So 47ohm parts were fitted instead of 470ohm as needed.

The upshot was that after 50 or so hours of use from new, the Vactecs took on a mind of their own - changing the sound of the amp in a random fashion that drove the young lady owner mad.

All the red LEDs were damaged and would drop light level or cut out and then jump back to normal after cooling.

The repair was tedious but simple, fit all new LEDs and resistors to the Vactecs. Covered by the maker's warranty ??

No, cos the gal had bought the amp in the USA and traveled back with it to Sydney where no warranty applied.





...... Phil