Question about the new LED street lighting

[davelew.livejournal.com]

This is a long shot, but there's enough random information on this community that I thought somebody might be able to help.

The city of Somerville is switching from old sodium vapor street lights to new, more energy efficient LED street lights. That's great and I support it, it's good for the environment and for the city's finances. The problem is with my telescope. I have a filter for my telescope that cuts out light pollution from old-style lights, by blocking light at precisely the wavelengths of the emissions spectrum of sodium atoms. Unfortunately, that filter does nothing for the new LED lights. It's nice to have a filter for urban astronomy if you're looking at anything dimmer than the moon, and I don't know what filter(s) to get.

So, here's my random question: Does anybody know the emissions spectrum of Somerville's new LED lights? Or even a single particular wavelength that they don't emit? I'd even be happy with a model number so that I could contact the manufacturer.

Comments

[davelew.livejournal.com]

Thanks! That's very helpful. Out of curiosity, how do you know those are the models being used?

If they are using LEDs to simulate a 4000K blackbody color temperature, that's good news. That's usually done with a combination of blue and red LEDs, which leaves a valley in the spectral response right around the Oxygen (III) bands, which happen to correspond to the color of certain nebulae. I don't have an O(III) band pass filter, but they are at least readily available.

[http://users.livejournal.com/_meej_/]

I'm a landscape architect who's worked with the City on past projects; these were, as of a few years ago, the "standard" models the City let us know about at that time (public record, so I feel comfortable sharing them). No guarantee they're still current, but I haven't noticed a change in the overall look.

[starphire.livejournal.com]

I don't know where you heard that a 4000K color temperature LED light is usually accomplished with a combination of blue and red LEDs, but as Xuth noted above most white LED lighting products actually use a blue LED in combination with one or more phosphors which convert much of the light into a broad spectrum of longer wavelengths, centered in the green and yellow part of the spectrum (or with a secondary peak in the orange-red if a warmer white color is desired).
LED streetlights are optimized for efficiency and visibility, which means cool or neutral white, not warm white. Here is a typical emission spectrum for a single-phosphor white LED excited by a blue LED:
http://en.wikipedia.org/wiki/File:White_LED.png
Formulations vary a bit between white LED manufacturers, but this is probably very similar to what Somerville's LED streetlights put out, whatever brand they use. They almost certainly do not use pure blue and red LEDs (which would be very magenta-looking and be a poor match for human eye sensitivity, which peaks in the green).

As you can see, there's a fair amount of light emitted at the sodium wavelength your filter blocks out, but most of the light is both longer and shorter wavelength and covers most of the visible spectrum. In other words, your filter would be fairly useless at blocking extraneous light from LED streetlights. And since such filters usually block a decent percentage of all other wavelengths as well, it'll probably do more harm than good to use it for looking at nebulae if the city switches over to LEDs completely.
The passband for an Oxygen(III) filter does happen to coincide nicely with an emission minimum around 500 nM.
Unfortunately, there is still a fair amount of light energy emitted even at 500 nM, because the overall spectrum emitted by these LEDs is so broad - unlike the sharp yellow peak emitted by sodium vapor streetlights. So again, you probably don't want to invest in one if you're trying to block light from LED streetlights.

I'd say your best hope is the fact that LED streetlights are typically much better at putting light on the ground rather than losing it out to the sides and upward, compared to sodium vapor lights. Hopefully that will reduce light pollution overall and improve your chances of seeing dim objects in the heavens, even without filters.

[davelew.livejournal.com]

Sorry, I meant to say 4000K LEDS are made with a combination of blue and red light, not blue and red LEDs. I was referring to the peaks at 450nm and 550nm in the spectrum by "blue" and "red", not to individual diodes.

I consider 4000k to be fairly warm light, similar to Illuminant A. To me, cool light is light at a color temperature above about 6500K.

[starphire.livejournal.com]

Ah, OK -thanks for clarifying (although 550nm is green, not red). I think my previous comment should still answer your original questions.

There are various reasons why phosphor-based white LEDs are actually pretty poor emitters in the red part of the spectrum - low efficiency, cost/rarity of raw materials, and degradation over time among them. In the LED lighting business cool white (~7000k) and neutral white (~4000k) LEDs don't have peaks in the red part of the spectrum - it's mostly blue and yellow-green, with smaller amounts of adjacent wavelengths. Warm white LEDs (2700-3300k) do generally have an extra peak in the red, but it's still secondary to the blue and green peaks. However, those aren't used in LED streetlights, due to their shortcomings - they basically exist only to satisfy consumer demand for something that simulates the warm yellowish glow of a household incandescent light bulb. In that market, aesthetic concerns are important enough to triumph over efficiency.