Hacker News Reader

Comment by wruza | Hacker News Reader

Ah, the good old "human eye can't see N Hz" that will make night lights insufferable for those who happen to see it. My grandma has a couple of lightbulbs in her apartment that annoy nobody except me. I feel like laser-blinded when they are on, no matter where I look. It's blindingly bright and visually dark. Idk if it's flicker frequency or light spectrum tbh, but it's that same "hey you shouldn't see it, cause nobody can" attitude. I also shouldn't see quick static-y shimmering on any lcd panel, because their backlight frequency is in a range that my eye couldn't even register.

View on HN

I feel the same way. If I'm sitting still I find I don't care much... but if I'm working with my hands and moving around it drives me crazy and gives me headaches, especially if I'm focusing on something spinning.

Just for fun, I replaced the LED dimmer in my garage with a homemade board that has a frequency knob in addition to a duty cycle knob. I've found that the cutoff for bothering me is around 10KHz, which is far higher than I'd have ever expected.

You should consider talking to the researchers who developed IEEE 1789. 120 Hz is problematic for many people, but 10kHz seems unusual unless you’re a camera.

It's specifically fast spinning objects that get me: I was spending a lot of time staring at a 3400rpm grinder wheel with a fine repeating pattern on it.

If I'm just sitting typing on my laptop, even the 120hz ones don't bother me.

Could it be that you inadvertently smoothed the visual flicker on the electrical side? Consider fast brightness sensor with known (ideally approximately linear) sensitivity to the AC component of brightness, and check with an oscilloscope or so. A soundcard with 96 kHz sample rate should probably have a high-enough filter cutoff to be quite and easily useful there, though I'd see to check the frequency response with an edge generator to confirm that it has linear sensitivity from low-ish frequencies to the >=10 kHz ones of particular interest.

That's a great point: above some frequency, the inductance of the system starts to average the voltage like the output of a switching power supply.

I did think of that: I used a camera with a rolling shutter image sensor and manual exposure settings to prove to myself the LEDs were really still getting the full swing at higher frequencies. Also, to be clear, I'm talking about 12VDC COB LEDs with hard on/off PWM dimming, not A/C ripple.

I don't know, but I'd expect the cutoff frequency to be a lot higher. The self inductance of 30ft of 16awg wire is about 20uH. I should test it...

> It's blindingly bright and visually dark

Well put. You can easily see flicker if you wave your hand in front of a flickering light source. The trail of your hand will look discrete, rather than continuous as it does with natural lighting.

I experience the same "blinding bright but visually dark" on a battery powered light I own. I wonder if it depends on the length of time the light is off.

Most light that flicker below 400hz really bother me, but I still perceive the surrounding as illuminated. While this particular torch light feels like "blinding darkness".

I wonder if your eyes move around a lot. It's known that we can see flicker in things like LEDs on a clock when our eyes move fast enough that subsequent flashes are sufficiently separated in the field of vision, particularly in our peripheral vision, and I've also experienced this with LED headlights while driving.

Dogs apparently see at a higher frequency. For those who want to torture animals...

Most LEDs already blink at high frequencies to control power output. This idea just synchronizes the blinking.

Reply on news.ycombinator.com