Hacker News new | ask | show | jobs
LED street lighting the newest challenge to old utility business models (blog.rmi.org)
67 points by thaddeusmt 4586 days ago
17 comments

NYC is starting to replace its sodium vapor street lamps with LED lamps, and expects that all 250,000 of them will be converted to LED lamps by 2017:

http://www1.nyc.gov/office-of-the-mayor/news/343-13/mayor-bl...

I wonder what percentage of street lights actually have activity below them? What about a motion sensing light that has dual-brightness levels? A lower level when no pedestrian or car activity is detected, and a brighter level when it is. Not only would this save energy, and look very cool when a car drives down a deserted street as lights brighten in advance of it, but it could enhance security as late night prowlers would have attention attracted to them by neighbors and police.
Most of the lights near my towns exit ramp work off of light sensors (dusk to dawn style). There are a couple where the sensor is either too sensitive or angled incorrectly (so it picks up vehicle headlights). As your car approaches, the light turns off. It's rather disconcerting when it happens.
That would be very annoying to people living nearby.

It could work for highway interchanges though.

Those already exist.
An interesting con to this LED lighting movement is that I have heard from a few CDOT engineers that the new LED lights do not put out enough heat to melt through snow that accumulates during storms and this has caused issues for traffic lighting. This is all second hand, of course.
Heat coils are very simple, as are thermostats. They could be added very easily to run when it's cold enough out that it would be helpful.
A simple light sensing trigger (coupled with a timer) would do the trick.

Light should be on but see no light and it's winter time? Turn on the heater.

Same situation in the summer? Send out a tech, the light's probably out.

And try to get a contractor to put them in out of the goodness of their heart or go through a rigorous study to prove equal compliance with federal and state regulations? Surely you jest.
They could be built as a part of the light, since we're changing all of that out anyway. Really not that hard.

Also, contractors shouldn't be studying compliance for new things. And you would pay them as you're installing the new lights. So not out of the "goodness of their heart" either. In fact, I have no idea what you're trying to get at here.

Most items installed on a transportation project must meet the standard specifications for the state which generally incorporates federal requirements as needed. Anything installed must be "per spec" and manufacturers or contractors installed items that are not per spec can be grounds for a refusal of payment or refusal of acceptance.

Specifications compliance is solely the responsibility of the contractor (who generally passes it on to the manufacturer).

In order to install your lighting the contractor would have to prove specification compliance, include the additional cost of the lights, and still win the low-bid process that almost all state and federal transportation projects are subject to. Alternatively, the contractor could win the project without the new lights and then come back with a value-added change order and these are not uncommon in my experience.

I guess what I'm trying to say is that it's a good idea, it just isn't feasible in the field in my opinion.

Well, what if specifications included something like minimum heat output? If government / society wants it, why wouldn't they include it in the specs?
Contractors don't spec roadway lighting, engineers do. The engineers provide guidelines that are (or should be) easy for the contractors to follow. Likewise, the DOT asking for bids could require certain types of lighting.

It's not always a quick process, but we do get innovations with some regularity.

So we'll never have any new lights?
Check out http://ledsmagazine.com/features/2/5/8 It says that:

The energy consumed by a 100-watt GLS incandescent bulb produces around 12% heat, 83% IR and only 5% visible light. In contrast, a typical LED might produce 15% visible light and 85% heat.

You can't always put a LED light in a enclosed fixture, as it might even overheat. It needs some air flow to cool it's heat sinks.

The question is a matter of power then. A 100W bulb at 5% light is 5W of light (assuming power vs energy can be compared this way... I don't do electrical engineering) which would correspond to a 35W LED producing less than one half of the heat when you consider that snow piling up in front of a light and the light enclosure will convert the IR to heat.
It can: http://en.wikipedia.org/wiki/Radiant_flux

Light measurement depends on the "use case":

Needed energy/power vs. How much "light" can be used by the human eye. Or, how good is the light suited for selecting pairs of (black?) socks: http://en.wikipedia.org/wiki/Color_rendering_index

That's why there are so many ways of measuring light: http://en.wikipedia.org/wiki/Light#Units_and_measures

My neighborhood in Los Angeles recently switched over to LED. I am happy that they are trying to save money/energy, but having a bright white light outside my window has been quite obnoxious. The older lights were dimmer with a more natural light color.
Contemporary LED lighting does seem much harsher with very sharp shadows and much less pleasant to the eye than older lights. I believe that lighting need to be assessed on it's overall merits, not just based on brightness / cost ratio.
Agreed. I am sure that they could tint/screen/soften the light to help make them less intrusive. I would even be happy if they dimmed the lights. It doesn't have to so bright for people to get around the streets.
They have been installing LED street lights in the UK for a while now. The LED supplier approaches a council, the council agrees to pay the supplier the difference between their normal and new lighting bill for X years. The supplier gives them the relatively expensive LED lamps (which talk to each other wirelessly) and management system. The council installs them. Supplier makes money from the price difference.
And they're horrible. They just put them in down my road and it's like prison searchlights compared to the old lighting.

To add insult to injury they put CCTV cameras on every alternate lamp as well.

And they're fine. They've just installed them in my area in East London.

The only disadvantage I've spotted is that, with the old Sodium vapour lamps, it was easier to spot when a car was coming around a bend since the headlight colour was significantly different to the ambient streetlighting. I'll adapt.

Good point.

Perhaps they put good ones in your area. I live in a cheap borough.

These ones are so bright and not diffused at all. It makes silhouettes of everything through my windows.

I never realized the scale of energy consumption by streetlights, though when I give it some thought, it's not all that surprising. I wonder if there would be the possibility of using these lights to only illuminate the road and say 1-2m to each side, avoiding light pollution for people living near streetlights. As it stands currently, the nearby streetlights keep my room lit at night.
This doesn't make any sense.

A high pressure sodium vapor lamp has an efficiency of around 150 lm/W. An LED has an efficiency of under 100 lm/W.

Where is the energy savings? If anything an LED uses more energy, not less.

The psychology of light is much more complicated than just number of lumens being thrown about. Because HPS lamps throw light in pretty narrowly defined bands, people feel less comfortable, and aren't able to notice things as well compared to light that is lower wattage, but more full spectrum ( http://www.lrc.rpi.edu/researchAreas/pdf/GrotonFinalReport.p... ). Like a lot of things, quality of light counts more than quantity.
LED lights have much longer lifespans: 50k to 100k hours, compared to ~30K hours for high pressure sodium lamps [1]. Reduced labour and parts costs could account for the decision.

Also, the referenced document shows LED lights as emitting 70 - 150 lm/W, so power consumption may be better than you think.

[1] http://www.grahlighting.eu/en/street-lighting-technology-com...

And sodium light is invisible for insects and very easily filtered out of astronomy pictures.
> A high pressure sodium vapor lamp has an efficiency of around 150 lm/W

Source?

Allegedly, Cree has developed a white LED with up to 254 lm/W. [1]

[1] Cree Sets New Record for White LED Efficiency http://techon.nikkeibp.co.jp/english/NEWS_EN/20120423/214494...

> Source?

It's poor form to ask for a source of something that would take you 15 seconds to google.

> Allegedly, Cree has developed a white LED with up to 254 lm/W

Interesting considering that the theoretical maximum is 251 lm/W - so I wonder which law of physics they violated to do that. (Most likely it's not actually white, but reddish.)

And this time I will give you a source since this is a tiny bit harder to google http://physics.ucsd.edu/~tmurphy/papers/lumens-per-watt.pdf

The hps operates at higher power and comes in models that scatter the light in all directions (so you need a reflector). The spectrum is constrained (as is for led) but you'll find it difficult to change it. All those lumens are in an unfavorable part of the spectrum. LED is much better at fooling our eyes, their spectrum can be tuned to sit close to the receptiveness maxima for our eyes.
Under scotopic (night) viewing conditions, it is lower than that. Also when you take into account the losses due to the reflectors to get the light out of a lamp, the nice little lambertian LED emitters require much less optics to do a good job of providing light on a wide road area.
This is how you add redundant non-information to a sentence: "semiconductor-based, solid-state LED lighting".

Someone who doesn't know what a LED is surely won't understand that "clarification". Marketing mindset!

Someone who doesn't know much about LEDs might come away thinking that these are extra special LEDs, much better than the normal LEDs they're used to, without the marketing team having to say anything that can possibly be challenged. Marketing mindset!

(Se also the Mad Men episode where they come up with the new slogan for Lucky Strike cigarettes)

Apparently the predicted cost benefit of doing this in Manchester (UK) will take 17 years to recoup[1]. If that's representative it doesn't seem hugely compelling for municipalities to switch at current energy prices. In the UK at least.

However I can see that it might well be disruptive for utility companies if those municipalities should decide to switch anyway(as per the article).

[1] http://www.manchestereveningnews.co.uk/news/greater-manchest...

> and the fact that bugs are not attracted to LED lights as they are to the ultraviolet light cast from conventional street lights

Why is this listed as a benefit? A bug that is at the streetlight is a bug that is not at my house. Is there some problem with bugs congregating at streetlight that I'm selfishly ignorant of?

What's with the anti-nature attitude? Bugs perform tons of ecosystem services like pollination, food, decomposition, eating pests, fertilization, etc. A complete food web is dynamically stable, preventing infestations. This is why monoculture agriculture necessitates pesticides/herbicides/fungicides, btw.

Insects could live without us. We couldn't live without the insects. My interest is entirely selfish. ;)

Of course, this is getting into much bigger issues than just streetlights...

Further reading: http://lighting.com/light-pollution-wildlife/

Get LED lights in your house then?

http://news.discovery.com/animals/zoo-animals/bugs-streetlig...

> People with homes closer to street lights may be at greater risk for exposure to certain insects, and even the diseases some of them could harbor. For example, another recent study showed that Triatoma dimidiate, an insect that can carry Chagas disease (an inflammatory, infectious illness), infested houses that were in closer proximity to street lighting, Davies said.

You have to constantly clean the lights that are full of baked on bugs, for one.
The Lion's Gate Bridge, which I believe is the busiest bridge in Vancouver, BC, relatively recently upgraded their lighting to LED. It reduced power consumption by 90%, and saves the city $30,000/yr in energy and maintenance.

In addition to the power savings, the old bulbs lasted an average of 72 hours between replacements, while the new ones have an estimated lifetime of 12 years, and replacing them is the 'hidden' cost that a lot of people wouldn't normally consider (and replacing bulbs mounted on a bridge is even worse than replacing street lights).

http://en.wikipedia.org/wiki/Lions_Gate_Bridge#History

You've accurately quoted the cited article, but that "72 hours" in the article is itself uncited and has to be in error. Normal mercury bulbs are good for 10,000+ hours, and 100W is well within normal. Maybe it was supposed to be 72 months? If you are strong enough to wade into Wikipedia politics, the edit came in here: http://en.wikipedia.org/w/index.php?title=Lions_Gate_Bridge&...
Won't the blue in the spectrum from these disrupt people's sleep even further? http://www.health.harvard.edu/newsletters/Harvard_Health_Let...
If you're sleeping under the streetlights, you have bigger problems.
It doesn't seem to take much exposure to damage your sleep. A friend reported that the fluorescents in the subways made a real difference.

Our society is still really stupid about sleep.

Curtain lining.
Our town just voted to buy our street lights back from the utility. It's projected to save us millions even before we upgrade them to LEDs. It's notable that the utility is forced to offer to sell them back by a state law, otherwise they would have no interest in it.
The statement that Christchurch, New Zealand is leveraging 12 percent of the $30 billion reconstruction fund for this type of project is wildly inaccurate. The article referenced only discusses a potential $5m project.
Here's my plan: use street lighting for a wireless mesh and stick camera's in each one so the NSA can keep us all safe.
csmuk's comment says they do just that in the UK (consider this comment as a GOTO statement)
More info here for those with an interest: http://www.leafnut.com
I call BS on this:

"But in most cities around the country, the local electric distribution company provides overhead street lighting as a basic service at a flat monthly rate per light, which includes the light itself, maintenance, and electricity. Therein lies the rub—regulated utilities often have little incentive to invest in more efficient streetlights, which offer a reliable, consistent, and often lucrative revenue stream that comes at a time of day (or night) when demand is low."

If their getting a fixed rate, LED's would be quite profitable for them due to lower servicing costs, and lower power consumption - meaning for that fixed rate, they take in more income.

Beyond that, in many cities - running and maintaining the streetlights is included in the franchise rights to sell power in a given city - meaning to sell power, the power company had to install and pays for the lights.

They cover that in a later paragraph: changing that means they go back to the community and renegotiate that flat rate, eating up any cost savings that the utility would turn into profit.
It's also a statement only applicable to the USA.

In the UK the Roads Service ( NI ) or Highways Agency are responsible for the emplacement and maintenance of lighting. So they'd be quite grateful for any power saving, as the electricity distributors charge them for actual usage.

Another solution would be to eliminate most street lighting, but that's a rather touchy subject.

I think that's only on trunk roads isn;t it? The local authority has the responsibility in towns.