I hate lighting that flickers and will often notice it when other people do not. A few friends also notice flickering that most people don't detect. It takes tens of kHz before I won't notice a stop-motion effect like a strobe light with moving objects.
LEDs do not have to flicker, but supplying low-ripple, constant-current DC power is not as cheap as ways of powering LEDs that do result in flicker.
I rarely notice PWM strobing with most LEDs at home.
However, once place where I do notice it and it's annoying is with taillights on some cars - sometimes when I sweep my eyes across the cars ahead, I see the strobe effect from LED taillights.
It doesn't seem to happen with all cars.
I've noticed it in a few bike taillights, but not nearly as much as in cars. I used to have an early LED bike headlight, and the strobing was very apparent when the light was dimmmed, I could only use it at its highest level.
I recently walked along a road with some Christmas lights (installed by the city) with the most horrible flicker. When looking at them straight on you could just notice it, but from the corner of my eye it was really bad and distracting. I wonder if there have been any studies done on the dangers for drivers of LED flicker.
I've noticed this with the LED strip on the lower front bumper of newer Mercedes cars.
I'm pretty sure the DOT requires that your headlights don't blink, so I wonder how low the PWM frequency can be until they consider it blinking?
I also wonder whether they do it on purpose. How hard would it be to run the PWM at a few kHz instead of the < 100 Hz frequency it seems like they're using?
Tens of kHz is higher than our "normal" temporal resolution. I see crafted experiments push the critical flicker fusion rate up to almost 1KHz. You are still order of magnitude off. You should blind test that number.
That being said, obviously the fact we probably can't see 10KHz flicker, does not imply it has no effect on our well being.
In the right conditions, very high flicker frequencies can detected by humans. It doesn't require any special abilities, just the right conditions.
Consider a flickering streetlamp outside your house. Now imagine you have two sets of net curtains covering your window, 10cm apart. The net is very fine - say a 0.1mm mesh.
If you sit next to the window working, and move your head back and forth, you will see a moire effect. If you sit 10cm from the curtains and move your head at 1 m/s, the 'flashing' you see is 50 Khz. If the light source behind the curtains is flickering at 49,995 Hz, you will see the resulting 5Hz flicker when the flicker from the moire effect and the streetlamp get multiplied.
As a regular human, you have now made a 50kHz flicker detectable with no special equipment.
Flicker at high enough frequencies isn't visible directly, but only as phantom array effect with very fast eye movement and high-contrast light. Flicker fusion frequency isn't relevant. I couldn't see 10kHz flicker when I last tested it (with an LED and 555 timer), but I could see 5kHz flicker. The exact threshold will depend on the observer and the test conditions. I find high frequency flicker very annoying because the appearance and disappearance of the phantom array effect looks like unexpected motion to me, and if something is moving unexpectedly then it needs attention.
Yep, and assuming we can sweep our eyes 150 degrees in 300 ms, we get a maximum angular speed. That means that the light source flickers on and off once in (150/0.3)/10000 = 0.05 degrees. The human eye can resolve at best 0.02 degrees. This confirms the proper order of magnitude.
Edit: if you turn your head at the same time, my numbers above are too conservative.
It got even worst when shopping malls decide to follow the trend of Green Power and LED lightning while buying some absolute crap to use. The results are there are places I can't go, worst one I will feel dizzy and may want to puke, lesser ones are super annoying I don't want to go.
Problem is not everyone have this problem. I have this what I call latency intolerance syndrome.
This. I work in lighting and most of the people doing specification have never even heard of flicker. I usually get brushed off as some kind of impractical visionary.
I am on a mission to make lighting healthy. That's why I made Bedtime Bulb: https://bedtimebulb.com/ It has the lowest flicker I've ever seen in a bulb form factor—even less than those claiming to be "flicker-free."
I think you mean a secret. I'd be very surprised if you're having LEDs custom made for your application. I'll go ahead and guess it's the Nichia NVSLE21A.
I don't understand why everyone insists on bundling a transformer with each LED bulb and doesn't just make a standard for LEDs so we can buy one good transformer and then just change the LEDs when they die.
The transformers are usually less reliable than the LED's themselves.
Also, LED's unlike nearly every other electrical device are constant current devices (as opposed to constant voltage, which is what a USB port, battery, or nearly any other power supply offers for example).
Constant current devices must be wired in series rather than parallel, meaning that if you wanted more than one LED per transformer, you would need to break the existing circuit to add an extra LED. That isn't really compatible with standards around electrical plugs - you would need to short out all unused electrical outlets in a constant current system.
Constant Voltage vs Constant Current is another one of those 'DC vs AC' or 'electron has positive or negative charge' issues - We have chosen one standard, and everything is built on it. We could go back and redecide, but all infrastructure would need updating.
LEDs are not constant current devices. They are diodes with non-linear I-V characteristic curve.
The question of supplying LEDs with CC or CV power supply is only the question of efficiency and cost.
Most LED strips are made as a serial/parallel combination of LEDs with an added resistor for each segment, to get the desired current through the LEDs in a segment from a ~12V CV power supply.
You don't really have that much choice when it comes to LED strips. CC power supply would need to generate thousands of volts or you'd need a bunch of them. And that would be costly.
LEDs are better modeled as constant current because it's the I part of the curve which is typically more unreliable. The junction voltage is largely consistent, but due to the exponential increase in current, to balance the amount of power it's easier to regulate the I component of P=I*V for consistency than it is the V portion.
You can run any current through the LED. (well, up to a point :D) It's one way to regulate the brightness. I'm not sure what "modeled as constant current" really means when talking about a passive device.
If anything I simplify leds as maintaining a "constant voltage", no matter the current I send through them (in the right direction). Just like other diodes.
> it's easier to regulate the I component of P=I*V for consistency than it is the V portion.
I understand, and we're on the same page. But it doesn't change the fact that almost no led strips do this. You can iron out over individual LED differences by putting 3-4 in the series, and regulate the current using a resistor.
In fact, most of the LED indicators you'll see are driven like this. LED + resistor [+ optionally PWM or multiplexing].
Only for power LEDs you'll usually see anything more complicated invlving DC/DC CC switching power supplies.
No, I was talking about the LED light bulbs that replace incandescent light bulbs. Every one of those comes with a transformer, which we throw away when the LEDs die (or vice-versa). Could we not separate the two and have a transformer supplying constant current, and then we could swap that or the LEDs out when one died?
You don't see it directly, but you see its interaction with eye movement. See "phantom array" effect. I can see this up to at least 5kHz under the right conditions.
When there's multiple flickering LEDs in the same room, they create interference. So even if we don't notice a single one flickering, when all of them go from black to full at the same time, and this happens 20x a second, we can notice that.
While there are certainly conditions high frequencies (tens of kilohertz) can be detected by a human, this explanation isn't it.
In general, light is linear and time invariant. That means in a flickering light system, a flicker of blue light then of red light, or red and blue at the same time, are indistinguishable if the 'sampling' system doesn't have sufficient bandwidth to resolve the flicker at all.
There do exist nonlinear optical components[1], but none that exhibit nonlinearities at the light intensities you'll find at home!
Additionally, even if the above were not true, flickering LED sources tend to flicker at a multiple of the AC supply, so it's very unlikely you'd see one flickering at 1000 Hz and another at 1001 Hz for example. For battery powered devices, it could easily happen though.
Plenty of cheap LED light systems flicker at 50 or 60Hz, see elsewhere in this thread. Also, the flicker can be indirectly visible, for example by turning motion blur into a sequence of discrete strobe events. My favorite experiment is to shake the end of a spoon (because it is curved and reflective) back and forth in an area predominantly lit by strobing LEDs. You will see a number of spoons, instead of a blurry spoon like you would under incandescents.
LEDs do not have to flicker, but supplying low-ripple, constant-current DC power is not as cheap as ways of powering LEDs that do result in flicker.