White LED bulbs are commonplace in households by now, mostly due to their low power usage and high reliability. Crank up the light output enough and you do however get high temperatures and corresponding interesting failure modes. An example is the one demonstrated by the [electronupdate] channel on YouTube with a Philips MR16 LED spot that had developed a distinct purple light output.
After popping off the front to expose the PCB with the LED packages, the fault seemed to be due to the phosphor on one of the four LEDs flaking off, exposing the individual 405 nm LEDs underneath. Generally, white LEDs are just UV or 405 nm (‘blue’) LEDs that have a phosphor coating on top that converts the emitted wavelength into broad band visible (white) or another specific wavelength, so this failure mode makes perfect sense.
After putting the PCB under a microscope and having a look at the failed and the other LED packages the crumbled phosphor on not just the one package became obvious, as the remaining three showed clear cracks in the phosphor coating. Whether due to the heat in these high-intensity spot lamps or just age, clearly over time these white LED packages become just bare LEDs without the phosphor coating. Ideally you could dab on some fresh phosphor, but likely the fix is to replace these LED packages every few years until the power supply in the bulb gives up the ghost.
Thanks to [ludek111] for the tip.
AFAIK it’s not UV but blue LED.
That’s what I’ve always understood.
You’re right, most are blue-pumped (450nm), but there are some manufacturers pushing violet-pumped (405nm) leds. The rationale is to reduce the blue peak in the spectrum which allegedly improves color rendering and is more pleasant to the eye.
We have some street lights on our state roads in Connecticut where that has happened. Weird looking, but no one seems to get around to getting the manufacturer to cough up a warranty replacement.
https://hackaday.com/2024/10/09/fail-of-the-week-the-case-of-the-curiously-colored-streetlights/
Same in our area. And supposedly they were going to get replaced, under warranty. That never happened. I was hoping to buy up a few of the old ones for cheap. I actually like the color, think it’s purty!
I live in a rural area, where there isn’t any regular street lighting, however it is possible to get the power company to put a light on a nearby pole for a monthly fee (or sometimes in your yard on a dedicated pole, but presumably that has some installation costs as well). One of my neighbors has one on the pole at the street in front of their house, and it’s failed in this manner as well.
It was obnoxious enough, having the only super-bright light around, but now it’s gone an eye-searing purple, and it’s even worse. I find the purple light painful to experience, but I have trouble with blue LEDs as well.
Blue LEDs are the optical equivalent of fingernails on a chalkboard.
It’s hilarious/annoying/sad to see streetlights slowly, one by one, shifting blue as they succumb to the phosphor shedding disease. Calgary, in particular, seems to have got a really bad batch of luminaires a few years ago, but several other cities I’ve seen have got it too.
The even more irritating failure mode is the random frequent flash. I swear one light on a road I frequent was sending Morse code for months before it got replaced.
They are quite bad for the ecosystem, attracting insects to their death. The insect collapse is real, and the consequences for us humans will be huge.
As a child in the 1960s in the Central valley California, I would always see tons of spider webs and bugs in those webs wherever there was a light on at night. By the 1990s there were never any bugs or spider webs around any lights that were on at night. Was it the insecticides used? The stress in the world is much higher than it has ever been.
Heh heh no one expected me to exfiltrate the Canadian defense data via THE STREETLIGHTS!!!!
Bwahahahaha
Hmm. Interesting. That morse-code blinking streetlight was near a local big defense contractor too.
Happens in Calgary as well – the theory was the temperature range causing delamination. It ranges from almost -40C to the high 30s C here:
https://www.ctvnews.ca/calgary/article/hard-on-the-eyes-defective-streetlights-on-some-calgary-roadways-have-drivers-blue/
Minor correction: The vast majority of white LEDs use a 450nm blue LED to pump the phosphor, not UV.
There are some specialty white LEDs that use a UV diode to get very good colour rendering (eg. Yuji VTC series) but they’re pricey.
High intensity only means the heat producing diodes are poorly liberating the heat they make. As far as I know they should only do 20mA per diode die. There are many manufactures that push this and they are filling the world for most of the LED haters with short life, flickering, and this. I bought several screw in mini spots and I now run the remaining ones at 80 to 90% power over the workbench. The first 2 had only a couple hundred hours life.
No LED should have any part inside that is too hot to hold your finger on. These ran too hot! There is a trailer park (go figure) near town that has gone purple, thankfully I’ve never seen this anywhere else. Duke Energy seems to get a warm light on the replacements on the street, but their bluer rent-a-lights go into flickering quite often. They probably aware of this which is a serious traffic hazard. For this duty they should be very conservative as it costs time to replace, and should fail-safe if intermittent.
I have a neighbor that rents a light from Duke. It’s been obnoxiously purple for at least a year. I guess I should be glad it’s not flickering. It’s painful enough with just the purple.
I’d be surprised they haven’t done anything about it, except that they don’t do anything if they can avoid it. Our pole at the street is leaning at a pretty decent angle. It does this every few years. They keep doing band-aid fixes (pull it straight, shove a bag of expanding foam into the newly-expose gap). The latest tech agreed that it needed to be guyed from across the street, but it’s been months, and that hasn’t gone anywhere. It continues to lean, held up mostly by the wires to the other poles.
I design LED drivers. We have plenty that drive 2A through individual dies, which are easily capable of handling it. I’ve worked on a die that could handle >30A. If you can extract the heat well enough, the bondwire can apparently handle well in excess of 2A. Which probably isn’t surprising: they’re easily visible, definitely larger than a human hair. The die bonds on our drivers are just about 0.025mm and can handle an amp each, so our 2A parts use multiple pins in parallel. We used to see a lot of led modules that had parallel dies for higher powers, but now they’re visibly single-die.
There is a parking lot near me that the entire lot is purple from every single LED failing. They have not done anything about it in years.
We had numerous warranty claims over this on brand new lamps in cars l. PC converted Leds are very well known from this
Too hard they Crack and flake off … too soft and jelly like they get torn
My former workplace decided to increase efficiency they would route these things off into individual boards and send them skyrocketing down a roller ramp loose in a tray…
Claims went up 30% and they are nearly out of biz
White LEDs are NOT UV LEDs + phosphor coating !!!
They are blue LEDs, close from the end of visible spectrum (but not beyond except for a little fraction of the light emitted (LEDs are not monochromatic)
If it was the case, UV LEDs would be as cheap – or cheaper, not having the phosphor coating – than white LEDs.
Ultraviolet is per se NOT visible.
LEDs (the individual diodes) emit light with a very narrow bandwidth. They are nearly monochromatic; so for almost all purposes they are just considered monochromatic light sources.
And don’t forget the case of the purple spots from the problematic LEDs in the backlight of many TVs, mine included.
My LG tv started getting purple areas that were mostly visible while watching black and white movies, I replaced the backlights and hopefully the replacements don’t suffer the same fate.
A co-worker gave me a Vizio tv that the business was going to throw out due to blotches from dead leds and the backlights are blue, I was a little confused at first because the picture colors looked normal when watching the set before disassembling. I put the screens back in front of the backlights and the colors returned to normal again. This tv has a phosphor layer mixed in with the screens.
That’s called a QLED TV. It’s using Quantum dot technology to convert the blue into green and red.
The failure problem isn’t with the phosphor per se, but the matrix that the phosphor is embedded in. These are typically organic polymer based (though ceramic /glass matrices have been investigated) and are subject to debonding/fracture/failure with the extreme conditions that they operate in. I don’t work deeply in the field and there may be someone out there who can comment a bit more knowledgeably about the specifics .
It may also be a typical case of corporate “Why fix it if it sells more product this way?” similar to the Phoebus Cartel nonsense with filament bulbs.
https://www.led-professional.com/resources-1/articles/new-glass-based-phosphors-for-white-light-emitting-diodes
https://www.sciencedirect.com/science/article/abs/pii/S1369800124006218
https://en.wikipedia.org/wiki/Phoebus_cartel
A lot of cheap leds just used an organic dye that flouresced, but also bleached out fast. This was very common in torches, where they could go blue in hours.
So the phosphor is/was definitely an issue at the low end of the market
Some led products use a remote phosphor, where the phosphor is embedded into the diffuser and doesn’t get nearly as hot. It’s less of a point source though so not as good for focusing into beams.
Although it is possible to produce beams with remote phosphor, Laser projectors also often use a remote phosphor to generate the green light (red and blue are done directly with laser diodes)
Same reason you have purple street lights. I’m just surprised it happened to home LED lights as well since that’s much less common occurrence despite the higher penetration. Most of my lights just die, not go purple. But purple street lights seem to have one somewhere that everyone knows about.
The odd thing for me is that I’ve never experienced this, all my lamps either die or go rather dim with age.
And I don’t recall seeing it outside either.
I’ve seen the blinking issue plenty though in commercial places, where the driver gets into that loop.
Noticed this has happened to some of the street lights in and around Boston. They developed a clear purple color at night!
Led bulb claimed life seems to be total fantasy. I now write the date, shop, invoice#, and claimed lifetime in hours, on every led bulb when I buy them. We have actual consumer law, and I have zero trouble getting a bulb replaced with all that written on it.