Automotive LED failures/durability

cobb

Flashlight Enthusiast
Joined
Sep 26, 2004
Messages
2,957
I spend/loose up to 3 hours a day for transportation. Sometimes in transit, waiting at bus stops, etc. One thing I have noticed is that many of the circular led brake and amber turn signal lights fail in pie pieces. Rather the light is a cluster of lights or just 6 in a star shape.

Ive seen these on city buses, para transit vans, numerous deliver vans and commercial trucks, tractor trailers, garbage trucks, etc.

As for the 2 led side markers, Ive seen some that do not light, some that are dim others that are super bright. Mainly on city buses, paratransit vans and some comemrcial trucks.

Also those third led red brake lights. I am seeing more and more of those where an led or two in the middle has failed. I have seen this on suvs, cars, aftermarket wings for cars. So far, I have yet to see a cluster led lens on a car like a turn or brake light with any failures.

Last but not least, the traffic lights that have been switched out to led. Ive seen many red stop lights where they start to fail, some almost half or just a few round spots.

I have 2 amber and 2 red 6 leds each side marker lights on my wheelchair powered 24/7 since january this year and so far, no problems other than the ambers flickering and failing when they get wet from rain. The reds seem rock solid.
 

idleprocess

Flashaholic
Joined
Feb 29, 2004
Messages
7,197
Location
decamped
Electrical systems in cars almost never operate at the "12 volts" they're claimed to be. The typical LA/SLA (lead-acid/sealed lead-acid) battery is 14.4V when fully charged and ~10.8V when almost-flat. Just to spice things up, the voltage regulator in a car doesn't maintain 12V as much as it tries.

Where am I going with all of this? Most LED products I've seen for 12V systems are simply a series of LEDs with a resistor for each LED. If the voltage to the LED series were regulated, that would be fine, but... it's not. If the design has most of the voltage drop going across the LEDs, a small upward change in system voltage will dramaticly increase the current going through the LEDs.
 

Skyline

Enlightened
Joined
Aug 17, 2000
Messages
755
Location
New Jersey
10.8V? Hehe, a couple of a months ago I left my maplight on for three days and the battery drained down to 8.8V. /ubbthreads/images/graemlins/blush.gif

I have to agree that LED products in cars need to be regulated. On Friday I put in a couple of georges80's Luxeon Dome Light modules and I think they should achieve their full lifespan. They are fully regulated and can handle a wide range of voltages.
 

markdi

Flashlight Enthusiast
Joined
Jun 27, 2003
Messages
2,403
Location
Portland Oregon
I have a blue led that I need to replace in my car.
I have an aftermarket led 3rd brake light on my old 88 car.
I bought it from jc whitney in 1996 and all of the leds
are working and of equal brightness.

car baterys only have 14.4 volts across them when they are being charged.
 

gadget_lover

Flashaholic
Joined
Oct 7, 2003
Messages
7,148
Location
Near Silicon Valley (too near)
Leds on cars suffer from the same things that bulbs do, and then a bit more. Many lights are constructed so the mount is the negative connection to the battery via
the car body. In other words the sheet metal screw is the ground point.

As pointed out, most auto LEDs are simply resistored. Te keep prices down, they are often wired parallell + series. Depending on the color of the LED, there would be from 3 to 6 LEds in series with a single resistor to limit current. In a series circuit, when one goes, all 6 will go.

The neat patterns you see on a stoplight that fails simply reveals the part that has failed; An open ground path can take out a big chunk. If they use a dozen power transistors, you will see 1/12 of the LEDS die when one goes. The pattern of the LEDs is based on what was easiest to layout or wire, so the patterns cam be bizzare.

Daniel
 

NewBie

*Retired*
Joined
Feb 18, 2004
Messages
4,944
Location
Oregon- United States of America
During starting of a car, and other electrical events, the battery voltage can really spike up. On some vehicles, I have seen spikes as high as 40 volts. Without protection for the LEDs, they turn into expensive fuses.

Also, heat/vibration/humidity can put additional stresses on LEDs, and if the device that they utilize wasn't designed for this, or if they implemented the LED solution poorly, you'll have early failures.

Stoplights/walklights are a good example. Some designs last a very long time, with under a 2% failure rate (which were attributed to failures in the electronics anyhow) in a 3 year period, and some designs are lucky to have under 50% failure rate after two years, due to the LEDs themselves.
 

Big_Ed

Flashlight Enthusiast
Joined
Oct 28, 2003
Messages
1,768
Location
Sycamore, Illinois
I guess it shows that LED's, although durable, aren't yet the end-all be-all for lighting. I personally thought heat and vibration didn't bother LED's, after all aren't they supposed to be shock-proof? Why can flashlights be treated so roughly, being banged around, (and advertised as passing tests to the like with flying colors), yet possibly fail due to vibration and heat in cars? (Unless the vibration is causing a soldered wire to come off or something). It doesn't make sense to me. I think it's pretty much the fault of the voltage, and/or shoddy construction.
 

evan9162

Flashlight Enthusiast
Joined
Apr 18, 2002
Messages
2,639
Location
Boise, ID
LED installations in cars that use a simlpe resistored setup are asking for trouble. A proper design with LEDs incorporates a current regulator, filtering, and voltage clamping to get rid of all the nasties in a car's electrical system.
 

Haesslich

Flashlight Enthusiast
Joined
Nov 2, 2003
Messages
1,433
Location
Canada
[ QUOTE ]
Big_Ed said:
I guess it shows that LED's, although durable, aren't yet the end-all be-all for lighting. I personally thought heat and vibration didn't bother LED's, after all aren't they supposed to be shock-proof? Why can flashlights be treated so roughly, being banged around, (and advertised as passing tests to the like with flying colors), yet possibly fail due to vibration and heat in cars? (Unless the vibration is causing a soldered wire to come off or something). It doesn't make sense to me. I think it's pretty much the fault of the voltage, and/or shoddy construction.

[/ QUOTE ]

Heat is the primary killer of LEDs as far as dropping their lifetimes goes - like any circuit, more heat means the circuit's literally being baked alive, and that'll also cause the phosphors to degenerate, or at least that's my understanding of the situation. Any circuit that's overheated will eventually fail.

As far as vibration goes - LEDs are as susceptible to sudden shocks the way incandescent filaments are, so they're a lot more rugged in that way. However, that COULD theoretically work them loose or screw with the electronics by shaking them loose or breaking connections, so that's something to be considered.
 

NewBie

*Retired*
Joined
Feb 18, 2004
Messages
4,944
Location
Oregon- United States of America
[ QUOTE ]
Big_Ed said:
I guess it shows that LED's, although durable, aren't yet the end-all be-all for lighting. I personally thought heat and vibration didn't bother LED's, after all aren't they supposed to be shock-proof? Why can flashlights be treated so roughly, being banged around, (and advertised as passing tests to the like with flying colors), yet possibly fail due to vibration and heat in cars? (Unless the vibration is causing a soldered wire to come off or something). It doesn't make sense to me. I think it's pretty much the fault of the voltage, and/or shoddy construction.

[/ QUOTE ]

Shock is one thing, constant vibration at certain frequencies is *quite* another.

Additionally, you have the board mounting system, when in a vibration environment like a car. Boards can go into resonant modes which amplify the stress on the LEDs, depending on the design.

Producing and designing a reliable piece of electronics is not as easy as it first seems, especially when faced with a highly cost competitive environment.
 

dano

Flashlight Enthusiast
Joined
Aug 11, 2000
Messages
3,881
Location
East Bay, Cali.
I have to Google for the article, but locally there's a town that converted its stop lights to LED's as a cost saving measure. Evidently, the failure rate is so high on the LED's that it's costing more in repairs than it is to run incans. in the stop light fixtures.

-dan
 

NewBie

*Retired*
Joined
Feb 18, 2004
Messages
4,944
Location
Oregon- United States of America
Here is a report I referenced on this in another thread, various cities report widely different experiences, about as many as there are brands. You need to understand the technology and investigate the folks that make them very carefully, and it would be wise to have long term warranty coverage.

http://www.dot.wisconsin.gov/library/research/docs/tsrs/tsrledsintrafficsignals.pdf

There is more information in this thread:
http://www.candlepowerforums.com/ubbthreads/showflat.php?Cat=&Board=LED&Number=660414

I do have several hundred thousand white Nichia LEDs in the field, in a "high vibration" environment, which is in Boeing 747/757/767/727/737 aircraft with 10,000-20,000 hours on them, and have only seen one single failure to date, which was due to soldering on a part. I took great care to very aggressively remove the majority of the heat, and actually under powered the LEDs by putting only 30% of the normal amount of current in them. Additionally, by doing both of these things, in the lab test, I see less than a 10% degradation at +20,000 hours, and a sensor turns up the power to the LEDs to hold the light level to under a 1% variation. Eventually, the LEDs will reach their rated 20mA drive level, at which point the system will advise the maintainance crew that it should be replaced at the next maintainance step. The system will then further increase the power to the LEDs up to a maximum of 30mA per device.

Sometime in the future I hope to get one of the fielded units back to see how it held up in it's operational environment. For now I'm happy that they keep working as designed.
 

cobb

Flashlight Enthusiast
Joined
Sep 26, 2004
Messages
2,957
I am about as confused as i was when I posted this. Wonder if its more of a vibration/heat than voltage? Reason I say this I have them on my wheelchair. 2 red and 2 ambers in series to each other in parallel to the 24 volt pwoer supply while they are 12 volt units. The voltage after a charge is 26.5 and under load as low as 17 and with regen braking as high as 32 volts. The ambers seem to vary in color from a greenish yellow to an almost redish amber. Now these seem to have a circuit they plug into to and you can wire them up either way, they seem polarity protected.
 

James S

Flashlight Enthusiast
Joined
Aug 27, 2002
Messages
5,078
Location
on an island surrounded by reality
Allright NewBie, that sounds like quite an interesting piece of lighting or indication that you've got on those planes. Please post more info about that!

in another thread if you feel it's off topic for this one, but I am intrigued!
 

PocketBeam

Flashlight Enthusiast
Joined
Jun 22, 2004
Messages
1,009
Location
Southern California
As has been said above, conjecture is that most of these failures are because of improper or insuficiant heat sinking and over driving the LEDs.

Cobb, the LEDs have some circuit in line with them. The circuit is dropping the voltage down for the LEDs. They work good because they probally aren't pulling a lot of power and/or have good heat sinking. Most 5mm LEDS don't even need heat sinking at normal currents, and in small numbers.
 
Top