LED traffic lights can't melt off snow and ice from their lens.

Seems like a major, disastrous design flaw too me. They can’t be serious about having workers manually remove that snow from every traffic light? :confused: A major city might have say ten thousand traffic lights? And they are going to manually clear off the snow and ice? Good luck with that plan.

Yet, they plan to install *more *LED traffic lights? WTH is wrong with this scenario? Lets implement more of these failed things? Have these people lost their minds?

The obvious solution is to retrofit heaters on each light. Imagine the cost of that. Or just go back to incandescent.

This has been known for a while. Given the choice of using ~100 watts per signal light set, 24/7/365, and having integral snow-melt capability and a way to keep some clever birdies warm, and reducing that to ~10 watts per light and needing to find a way to keep lights clear of snow and ice - like, including a thermostatic heater of very low power - I’ll take the new problem over the old one.

Time to put the kami to work.

I read somewhere (no cite, though) that retrofitting a heater into them negates the cost/green savings of the LEDs using less electricity.
Since building a moisture sensor into them makes them more expensive & more complex (& therefore malfunctioning easier) the heating element would basically run all winter, whenever it’s below x° out.

I’d think that’s only an issue for a relatively small subset of cities around the world.

Here in Dallas (and parts south), the LED lights make great sense; they are energy efficient and they last a lot longer than the old incandescent ones, and it snows/ices about once a year, so it’s not a huge problem that they don’t melt snow.

That said, how tough would it be to engineer a heater that detects icing and only kicks on when there is actually ice? Just turning the heater on below a certain temp seems like a relatively inelegant solution to the problem.

I would have thought the consequences of snarled traffic would concern officials more. A traffic light that can’t be seen is similar to a traffic light in a power failure. Traffic becomes a nightmare. Every car treats that intersection like a 4 Way stop sign. (At least you hope and pray everybody does that).

It always scares me really bad because you never know when some dumb ass will just sail through the intersection. I dread driving when the traffic lights are out. It happened locally by my house three weeks ago. A big area of town (including my neighborhood) lost power. One of the major 6 lane avenues had traffic lights out.

Traffic lights that are on a horizontal pole over the road or (even worse) a wire strung across the street are approx. 15’ to 20’ feet high. Properly aimed headlights aren’t focused at this height so when it’s dark (& there’s less daylight hours in the winter) they are invisible. I could very easily see cars sailing thru intersections where the drivers don’t ‘see’ a traffic light.
Stop signs are at least lower in height & frequently painted with reflective paint, making them much more visible than an iced over LED traffic light.

Clearly, that’s not true if it’s thermostatically controlled.
ETA: LEDs are 5x as efficient as Incandescents - even if the heater had to be on 1/2 the year, it’s still a win.

is this really a problem which needs solving? we get plenty of snow where I live and I’ve never seen any of the LED traffic signals get a build up of enough to block the light. Pretty much every signal I’ve encountered uses an open bottom or “baseball cap” style of visor, so snow shouldn’t be building up in front of the lens.


besides, the bulbs typically used in a 12" signal are ~ 116 watt incandescents, and some of that energy is lost via filtering by the lens. are they really on for a long enough interval to appreciably melt any snow w/o it re-freezing right there?

You could just install heaters with a cheap RFID switch in each light, and then install an activator on the plow trucks. As plow trucks go by, they would automatically activate the heater for, say, 36 hours.

Too simple.

everything is simple if you’re not the one who has to do it :wink:

How many LEDs are bundled in the Japanese signals? Could it be just 12 or so, like a brake light on the back of a school bus?

Winters are long and harsh here at 7,200+ feet of elevation, yet I’ve never noticed such a problem with our LED traffic signals. Ours have (I’m guessing) about 120 bundled LEDs. Could 120 LEDs produce sufficient heat?

It’s absolutely not rocket surgery to embed a heated ring on the lamp assembly.

I do not notice this as a problem here in Calgary and we have had LED traffic signals for years. Our winters tend to be made up of cold days well below freezing and warming trends where everything melts and drys off in a day or two. In a climate with more wet sticky snow, like parts of Japan, it might be more of an issue.

Okay, back in the land of reality, we know two things:

  1. This really isn’t a big problem.

  2. We can trust the Japanese to come up with a very engineering-intensive solution to this problem. I’m thinking 15-foot tall walking robots that have hair dryers for arms, but look like cute cartoon characters and talk in a very high pitched, girly voice.

This looks a lot like a very minor issue made out to be a big problem by those who have an issue with LED lights (as many on the right seem to have.) I live in Colorado where we get big snows and cold with regularity and it just isn’t an issue. We had 13" of snow this weekend and it hasn’t come close to the freezing mark since and everything is perfectly fine. I’ve never been unable to make out a streetlight signal since I started driving.

the hard part is making sure it works properly, will work for at least 10 years (or else the repair/replacement comes out of your pocket,) and won’t damage anything else if it fails. 80% of product development isn’t getting it to work, it’s getting it to work to your customer’s expectation.

Japan can’t afford that anymore.

You may still have the regular incandescent traffic lights. It’s going to be quite expensive to change all the lights to LED. I doubt many cities have started the conversion.

Rather than retrofitting. It seems like they need to design and implement the heating ring first. Fully test the new traffic light assemblies. Then install them. Rather than install something that will still need retrofitting later.

They’re LED. They just have longer, deeper shrouds that are painted black. No heaters needed. They installed them when the old incandescents burned out (which was all the time.)

You still are thinking the lack of excess heat is an actual problem?

Anyway, the cost of conversion is lowered significantly by the saving of replacing bulbs constantly. The city of San Jose (10th largest in the country) and its surrounding suburbs have been replacing lights for years now on a rolling basis, and now pretty much every signal in the area is LED.