|
Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets. |
|
Thread Tools |
31st Jan 2019, 11:54 pm | #1 |
Dekatron
Join Date: Mar 2008
Location: St. Frajou, l'Isle en Dodon, Haute Garonne, France.(Previously: Ellesmere Port, Cheshire, UK.)
Posts: 3,183
|
Blue/white LED failure modes
Hi,
Apologies if I've posted this in the wrong place. I was looking at an old LED wall lamp which has two LED units, One shining up, and another shining downwards. Each unit consists of twenty one blue/white LEDs in series across a simple rectified & smoothed supply via a capacitor dropper and series resistance. What I'm curious about is that the LEDs are now very 'tired' and most of them have become very dim, also, one does not light at all, but still passes current to keep the others alight. I'd always assumed that a LED either lights, or not, and always failed open circuit. But this is not the case here. I'm curious about the failure modes of LEDs. Cheers, Pete.
__________________
"Hello?, Yes, I'm on the train, I might lose the signal soon as we're just going into a tunn..." |
1st Feb 2019, 12:05 am | #2 |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Blue/white LED failure modes
It is very common for LED's to fade out or go to a low brightness over time.
When LED's first hit the scene in great numbers in the early '70's, all looked well. But a decade later, many had faded especially things like red 7 sement ones on alarm clocks etc. Of late I have found that blue LED's seem particularly bad in this respect. In the facility where I work we installed synchronized large segment blue LED clocks in many rooms, after a year, many segments have faded, some more than others. Also a mains power energy meter I have, some of the segments have died and others faded over 2 years, they are blue too. Oddly I have a nixie clock that has been running for decades, still good light output from the nixies, making me wonder if they are out-doing seven segment LED's ! Someone might offer up a technical explanation of the mechanism that degrades the light output. |
1st Feb 2019, 1:21 am | #3 |
Dekatron
Join Date: Jan 2012
Location: Brentwood, Essex, UK.
Posts: 5,339
|
Re: Blue/white LED failure modes
According to a lecture on LEDs I went to a decade or so ago, white light LEDs use phosphors to determine their colour temperature. Possibly phosphor deterioration is a factor. At the time the lecturer said that the colour temperature of the lamps then in production was affected by the actual temperature. This had delayed their adoption in some commercial premises which used air conditioning via ceiling vents, as the colour of the lamps near the vents was different from the colour of the lamps remote from the vents, something which architects found unacceptable. I don't think this is a problem with the phosphors of present lamps.
Last edited by emeritus; 1st Feb 2019 at 1:29 am. |
1st Feb 2019, 10:50 am | #4 |
Nonode
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,677
|
Re: Blue/white LED failure modes
The phosphor in white LEDs at the cheaper end of the market certainly does degrade, turning them more and more blue over time. I've done a lot of work with white LEDs and observed several failure modes:
1. phosphor degradation leading to a change in colour temperature, not necessarily the same in all directions, so you get a blue light with a yellow ring round it, for example 2. reduction of overall light output, loss of efficiency. I think this is due to something going on in the gallium nitride semiconductor itself. 3. failure of the bond wire, which can often lead to flickering and flashing. Very common with small, cheap LEDs connected in series 4. complete lack of light output but similar forward voltage and current characteristics to a working LED. This one's a mystery. All of these are brought on more rapidly by excessive heat, which is the enemy of LEDs. The better quality the LED, the more robust it is, in my experience. From some suppliers it's possible to buy LEDs which are graded as 'only suitable for toys and novelty items' which gives you some idea of the depths to which it's possible to sink! I had one notable case of a Lumileds Luxeon V 5-watt white LED which, at the time, cost about £25, I remember, and was only rated to last 500 hours. Its heatsink mounting failed due to a hard knock, and it died in mode 4 from the list above. It still had about the right forward voltage at its rated forward current, so it wasn't open circuit, but simply no light came out any more! Chris
__________________
What's going on in the workshop? http://martin-jones.com/ |
1st Feb 2019, 11:22 am | #5 | |
Dekatron
Join Date: Dec 2004
Location: East Yorkshire, UK.
Posts: 3,987
|
Re: Blue/white LED failure modes
Quote:
Maybe I have been lucky with my home made clock, made in 1975 and featured in this 11 year old post. https://www.vintage-radio.net/forum/...ad.php?t=23781 #1 It is still in daily use and no sign of failing. This clock uses MAN 3640 Red 7 segment LED. Blue LEDs were one of the last colours to be introduced, I wonder if there are more complicated chemicals and production procedures involved that led to some LEDs being unreliable. John. Last edited by 60 oldjohn; 1st Feb 2019 at 11:31 am. |
|
1st Feb 2019, 11:27 am | #6 |
Moderator
Join Date: Jun 2003
Location: Oxford, UK
Posts: 27,947
|
Re: Blue/white LED failure modes
The cheap ones also tend to be run at close to their maximum ratings (and sometimes beyond them). There were some very dodgy lamps on the market a decade ago when LEDs first started to replace CFLs in places like pubs.
|
1st Feb 2019, 4:08 pm | #7 | |
Dekatron
Join Date: May 2008
Location: Derby, UK.
Posts: 7,735
|
Re: Blue/white LED failure modes
Quote:
It's good, really, that the LED has come to the point of being cheap enough to use in toys and novelties. The first generation of visible LEDs for use as indicators, as opposed to fibre-optic sources, were only seen on really expensive, up-market equipment; we plebs had to make do with tungsten filament lamps.
__________________
If I have seen further than others, it is because I was standing on a pile of failed experiments. |
|
1st Feb 2019, 8:38 pm | #8 |
Dekatron
Join Date: Dec 2007
Location: Haarlem, Netherlands
Posts: 4,199
|
Re: Blue/white LED failure modes
I agree about keeping the out of the landfill, it's nothing new either: manufacturers sold off the lowest grade semiconductors for hobby purposes.
|
1st Feb 2019, 8:55 pm | #9 |
Octode
Join Date: Aug 2006
Location: Stoke-on-Trent, Staffordshire, UK.
Posts: 1,082
|
Re: Blue/white LED failure modes
I have recently changed about 40 cool white GU10 led lamps 5 years old and the colour has changed to a warm white
So not as stable as promised Trev |
1st Feb 2019, 10:23 pm | #10 |
Dekatron
Join Date: May 2008
Location: Derby, UK.
Posts: 7,735
|
Re: Blue/white LED failure modes
That's interesting! I guess the main, blue LEDs must be giving off less light now than when they were new, but still enough to excite the phosphor -- which must not have not degraded so rapidly -- into emitting enough green and red to swamp out the blue and give a warmer shade of white, albeit with less candle power. Or maybe the emission spectrum has shifted, and there is now less visible blue but more UV, which of course will still excite the phosphor.
__________________
If I have seen further than others, it is because I was standing on a pile of failed experiments. |