Semi OT: LED traffic light failures

It is amazing how accurate observations you made on the green traffic lights. I assumed that you should be driving through the crossing, when the green light is on :-).

For security reasons, in some countries, if the red lamp fails, all the traffic lights in the crossing should go blinking yellow. For this reason, you should never see a completely failed red light.

Never seen it in the UK yet. But our summers are not very warm. This one is positively cold 11C and raining outside at present :(

True green is a bit higher Vf than amber or red so dissipation +50% or so but that is probably not enough to explain the difference unless the environment is already close to their maximum operating temperature.

How does an overheated LED fail? Bimetallic strip movement of one of the internal contacts might explain the independent LEDs flickering.

LED drivers over here in the UK are close to the flicker frequency for peripheral vision and cause ghosting when you turn you head suddenly.

--
Regards, 
Martin Brown

Could that be mains 50/100Hz flicker?

piglet

On Mon, 27 Jul 2015 23:59:07 -0400, bitrex Gave us:

Idiot. I'm not sure how you can write about something, then f*ck up what someone responds with by guessing what portion of your shit they were responding to.

I was talking about the flickering green remark, where they all are working, but one sees the flickering just as condition transitions are taking place, not a few stray burned out LED drivers. In such cases the LEDs are usually still fine, but the circuit that drives them has failed.

On Mon, 27 Jul 2015 21:09:04 -0700, Jim Thompson Gave us:

Jim Thompson is yet another retard who has foisted this bullshit for years.

That pretty much makes you the troll, dumbass.

On Tue, 28 Jul 2015 10:08:16 +0100, piglet Gave us:

Just like a handheld flashlight is driven by a pulsed drive, they (signal indicators)are NOT anywhere close to line frequency.

Sheesh.

I don't know how they are made today, but there was a requirement that, if an LED failed, it would not make the entire signal dark. Hence the requirement that there be multiple series strings in parallel.

Ten years ago they were using SSRs. The reason I know is that there was leakage through the snubber causing problems with the lower current LED signals.

If so, it doesn't make it into the signal.

That's all in your mind, these modules are too well made:

Possibly, the green light is on longer than the red and the yellow: Scroll down to Figure 5-2 and notice the much longer time that the green is on. The following sections define how the "minimum green" should be calculated.

Ignoring the yellow light, it would seem that for every green light at an intersection, there is an equal duration red light on the perpendicular roadways. That suggests that if you find burned out green LED's on a main road, you should also find burnt out red LED's on all the perpendicular side roads. So, the longer green theory doesn't really work for simple intersections. (assuming the reliability of red and green LED's were equal). However, if the signals were timed to favor the main highway, which would be green most of the time, then even if the red LED's on the side roads were equally burned out, you would never see them because you're only traveling on the main highway.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

Indeed, confirmation bias! But also the reds appear more reliable (lower Pd is a believable mechanism), so that one need not expect complementary failures on the intersection.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: http://seventransistorlabs.com

It is a big mistake to think that a business has to be efficient to survive. It doesn't even have to sell reliable products. The one, single and only requirement is that it brings in more money than it hands out.

This is in my mind?

That's along a mile or so of Folsom street, on my drive to work. About a third of the greens have partial failure, some almost 50% out, many blinking.

It is tricky to get a good pic, through the windshield of a moving car.

The manufacturer has agreed to replace the green LED modules if the city will do the installation. It's expected to take several years to replace them all.

I have a friend in the LED lighting (not traffic light) business, and he's also noted the high failure rate of green LEDs.

--

John Larkin         Highland Technology, Inc 
lunatic fringe electronics 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

I see that frequently on auto taillights. I turn my head and that one Cadillac two lanes over becomes a dozen unexpected cars my subconscious alerts me to. It really bothered me a lot when driving in town. I am not in the city as much now so I can more easily turn off the false alarms.

I wish there was a standard on the flicker rate that would prevent this. Anyone know what rate are they flashing? I'm pretty sure it is below

1000 Hz. I'm thinking that were it above 1 kHz I wouldn't notice it. What speed do they flash those LED sign sticks?

--

Rick

I can't answer that authoritatively, Rick, but the guys I work with say that about 150Hz is the absolute minimum PWM frequency we will provide. I always go for more.

This does not take into account lights with more than two phases. Many intersections have separate phases for turning where the main road through traffic is red. It would seem to me that most of the time traffic signals should be showing red other than the ones with auto detectors which only turn when a car is present on the side road.

--

Rick

I have not seen any such problem with Canadian LED signal lights. I vaguely recall some issues years ago, and the multi-million dollar freeway information signs had to be refitted with better LEDs (10-20 years ago), but they seem to have gotten it now. Lots of LED streetlamps too, more glarey and blue-white but getting used to it from the old orangey Na vapor lamps doesn't take long.

Maybe someone foisted a bunch of crappy bulbs on different municipalities.. they tend to have a herd mentality.

--sp

--
Best regards,  
Spehro Pefhany 
Amazon link for AoE 3rd Edition:            http://tinyurl.com/ntrpwu8 
Microchip link for 2015 Masters in Phoenix: http://tinyurl.com/l7g2k48

I have a vague recollection that the british railways had a problem with early LED railway signals - wrong shade of green - more of a blue/green tinge. They are quite particular about that sort of thing...

Den tirsdag den 28. juli 2015 kl. 21.45.55 UTC+2 skrev Spehro Pefhany:

-Lasse