US Streetlights Are Turning Purple • Hacker News Board

Re: US Streetlights Are Turning Purple

#31

Post by addaon » Tue, Jun 17, 2025, 2:13 PM UTC

hedora wrote: post #16

BorgHunter wrote: post #8

Earlier quoted context omitted.

It depends strongly on implementation. LEDs can produce white light at any color temperature and CRIs ranging from terrible to great. Quality LEDs are quite nice indeed (when they're not turning purple). Most LED streetlights replaced sodium vapor lights, though, which produce the sickliest, most horrible orange color known to humanity. Just about any LED is an improvement over those.

Sodium lights aren’t even orange in the normal sense of the word. They only emit one wavelength.

> Sodium lights aren’t even orange in the normal sense of the word. They only emit one wavelength.

It's hard to think of a more normal sense of the word "orange" than "emitting and/or reflecting predominantly wavelengths between 590 and 620 nm." I guess you could argue that sodium is close enough to that lower edge to be yellow?

Re: US Streetlights Are Turning Purple

#32

Post by looofooo0 » Tue, Jun 17, 2025, 2:16 PM UTC

hedora wrote: post #16

BorgHunter wrote: post #8

Earlier quoted context omitted.

It depends strongly on implementation. LEDs can produce white light at any color temperature and CRIs ranging from terrible to great. Quality LEDs are quite nice indeed (when they're not turning purple). Most LED streetlights replaced sodium vapor lights, though, which produce the sickliest, most horrible orange color known to humanity. Just about any LED is an improvement over those.

Sodium lights aren’t even orange in the normal sense of the word. They only emit one wavelength.

What is the normal sense of the word? I only know the CIE standard observer to define colour and with this it is clearly in the equivalence class of orange.

Re: US Streetlights Are Turning Purple

#33

Post by bob1029 » Tue, Jun 17, 2025, 2:16 PM UTC

floatrock wrote: post #26

joshstrange wrote: post #14

> And yet some streetlights have suddenly turned a jarring shade of purple. It is hard to determine the exact cause without dissecting one of the defective lights, but scientists have a hypothesis: bright purple light suggests the phosphor layer around the lights has been “delaminated”—peeled off—exposing the blue LED light underneath, Brgoch says. What an annoying/bad article. "Here are our guesses of this when we c…

> Get a crew out there and figure it out. I mean, to be fair, the city maintenance crews don't generally run tear-down youtube channels with appropriate equipment to take a close look at a failing semiconductor-based light assembly. But the general point is correct. What could be done is connect with some of those crews' offices, understand what warranties are provided with the municipal purchase of those bulbs, and…

> Hopefully these bulbs weren't purchased from a RANDOMSYLLABLE amazon china dropshipper who disappeared after 3 months.

This is how most LED bulbs in use today are purchased. In theory, LED is absolutely superior when engineered correctly, but it rarely is on a statistical basis when looking at the available products.

It's hard to make an incandescent bulb that is shitty, other than making it not last for a very long time. I'd rather a dead bulb than one that turns my street into a nightclub.

Re: US Streetlights Are Turning Purple

#34

Post by PedroBatista » Tue, Jun 17, 2025, 2:18 PM UTC

modeless wrote: post #11

> people driving at night may notice a moving object in their peripheral vision more quickly under bluish-white light than under yellowish-white light, [...] Gaining improved peripheral vision under blue-tinged light comes with a trade-off, however: once the moving object comes into focus, it becomes harder to see. If we were switching away from LED to sodium vapor lamps instead of the other way around, they would ha…

No, sodium vapor lights offer great performance in terms of contrast and visibility. Especially in bad weather conditions.

Re: US Streetlights Are Turning Purple

#35

Post by robertlagrant » Tue, Jun 17, 2025, 2:19 PM UTC

PedroBatista wrote: post #34

modeless wrote: post #11

> people driving at night may notice a moving object in their peripheral vision more quickly under bluish-white light than under yellowish-white light, [...] Gaining improved peripheral vision under blue-tinged light comes with a trade-off, however: once the moving object comes into focus, it becomes harder to see. If we were switching away from LED to sodium vapor lamps instead of the other way around, they would ha…

No, sodium vapor lights offer great performance in terms of contrast and visibility. Especially in bad weather conditions.

Yes but that's the point: sodium vapour offers worse peripheral vision, so it could make some situations riskier.

Re: US Streetlights Are Turning Purple

#36

Post by looofooo0 » Tue, Jun 17, 2025, 2:21 PM UTC

xattt wrote: post #28

BorgHunter wrote: post #8

Earlier quoted context omitted.

It depends strongly on implementation. LEDs can produce white light at any color temperature and CRIs ranging from terrible to great. Quality LEDs are quite nice indeed (when they're not turning purple). Most LED streetlights replaced sodium vapor lights, though, which produce the sickliest, most horrible orange color known to humanity. Just about any LED is an improvement over those.

If orange was tolerated before, I am curious why orange LEDs aren’t being looked at again for resilience.

Well we cannot manufacture orange LEDs with good efficiency.

Re: US Streetlights Are Turning Purple

#37

Post by davidmurdoch » Tue, Jun 17, 2025, 2:23 PM UTC

Read somewhere a few years ago this was a manufacturer defect and they were in the hook for replacing them all.

I've (not seriously) considered buying a pellet gun to shoot out the 4 massive neon purple lights at the entrance to my quaint 1970s era neighborhood. They didn't remove the old light poles after installing the new ones about 3 years ago, so it's double lit with sodium vapor and purple now.

Curious is there is a single person on the planet that prefers the white (er, purple) street lights?

Re: US Streetlights Are Turning Purple

#38

Post by guicen » Tue, Jun 17, 2025, 2:24 PM UTC

It’s interesting how many people assume these purple lights are some kind of new design choice. But really, it’s just a side effect of the phosphor layer breaking down in the LEDs.

Makes you appreciate how tricky it is to balance cost, lifespan, and quality when you’re manufacturing millions of these for cities.

Re: US Streetlights Are Turning Purple

#39

Post by kragen » Tue, Jun 17, 2025, 2:27 PM UTC

davidmurdoch wrote: post #37

Read somewhere a few years ago this was a manufacturer defect and they were in the hook for replacing them all. I've (not seriously) considered buying a pellet gun to shoot out the 4 massive neon purple lights at the entrance to my quaint 1970s era neighborhood. They didn't remove the old light poles after installing the new ones about 3 years ago, so it's double lit with sodium vapor and purple now. Curious is there…

I haven't tried them, but they sound awesome.

Re: US Streetlights Are Turning Purple

#40

Post by ribcage » Tue, Jun 17, 2025, 2:28 PM UTC

[dead]