Hacker News new | past | comments | ask | show | jobs | submit
My pet theory is that the dimmable LEDs repel insects (when used as a grow light).

The thing that insects can see and we don't is the rapid flashing of a PWM power controller.

I wonder if these streetlights have power saving fade-ins as the nightfall approaches instead of clicking on at sunset.

Like - https://doi.org/10.1109/CSNT51715.2021.9509652

This. It likely depends on the driving technology. Some repell me too :P

Note you can make LEDs dim without PWM via constant current drivers, just less efficient as the duty cycle is 100%, they look much nicer though

PWM drivers should be banned, the quality of lighting is horrendous and it murders LEDs. You have an rms currernt that is in spec but the peak is way too high.

anecdote: I work in a factory, we have thousands of light fixtures. The older constant current LED fixtures have lasted a decade, the new PWM driven lights have had a ~1% failure rate.

How do you know the difference? Constant current is almost certainly a PWM controlled buck/boost regulator. The flicker is attenuated by inductance, more or less depending on the design.
Use a fidget spinner to see the difference. Spin up a 3 lobe fidget spinner and catch the reflection of the LED on it's surface. As the speed changes, a flickering LED will freeze the motion of the spinner. Even more fun with spinners :-).

Yes, adding an inductor will fix the problem, but that costs money. Not enough customers are complaining about the light flicker to justify the higher price.

I guess the distinction is 'filtered' vs 'unfiltered' then. If you look at the lights that fail early using a high framerate camera (1k fps) they flicker between full brightness and fully off. The lights that last longer have some ripple but its 5-10% brightness.
One of my favorite things I read growing up was a passage about what it's like to be a fly watching TV.

In those days TV's were big CRT boxes, illuminated by the electron beam which would race across the screen row by row.

A housefly's "refresh rate" on its eyeballs is much higher than that of a human, so the fly wouldn't see an image as we do, but would presumably find it quite relaxing watching the dot of light slowly move across the screen.

In those days. Oh no! I'm old.
Don't forget about flicker free dimmable LEDs. I install them in my house.
Yeah, while a lot of LED bulbs are bad, some LED bulbs have ever so slightly less flicker than incandescent bulbs.
The streetlights in my neighborhood still have the round photo sensors on top, that have something like a watt of standby current for the sensor itself. I don't think they're trying hard to save powwer.
The ones that have a watt of standby are the really old models (probably not sold for the last 30yrs) where you'd have an LDR controlled relay with no electronics

Anything more modern than that, even if it has a similar case, will use less power

A low-voltage photocell uses like 5mA @ 24VAC these days, it’s nowhere near a watt. You can buy a wireless battery powered photocell or occupancy sensor that will last a decade.

This Acuity occ sensor/photocell combo draws 4mA @ 24VDC: https://pdmassets.azureedge.net/library/SS/documents/specshe...

The article bluntly states: "Moonlight contains ultraviolet (UV) and infrared wavelengths because it is merely reflected sunlight. Insects cue on UV, believing it to be the moon. . . . Incandescent lights transmit UV but LEDs do not. Insects are far less attracted to LED."

However, my mind developed a similar conjecture to your's: what if the common frequencies of PWM occur within a kinetic visual blindspot of insect vision, like human saccades [0] in the novel Blindsight [1], so that insects are not aware of the LED at all?

0. https://www.ncbi.nlm.nih.gov/books/NBK10991/

1. https://www.rifters.com/real/Blindsight.htm#:~:text=%22There...

Are we giving too little credit to the insects?

They've had many generations to evolve alongside all human false light sources. Insect generations are usually pretty short. Perhaps they just find it evolutionary useful to cluster around human sources of light. Presence of human food, liklihood of bumping into other insects of the same species, warmth, etc.

If human sources of light were interfering somehow, they've had hundreds or thousands of generations to breed out that proclivity.

You assume it is possible to breed out without harming other advantageous behaviors which might not be case.