In the history of LED lighting, it took quite a bit of work (hybrid phosphor technology, etc) to get them to emit warm color temperatures. The exact color temp can be a personal preference, but I think a lot of people aren't privy to the difference the color temperature of "white" light can make!
I want to go on a guerrilla campaign around my neighborhood and replace porch light bulbs with warm equivalents.
I just went through this with my car. All the OEM overhead interior lights were horrible 6000k+ type bulbs that not only made the interior feel like a hospital room but completely ruined my night vision turning me into the classic “Jesus!! You’re going to kill someone!” dad whenever my kids would fiddle with them at night.
I cannot overstate how much of a difference it made switching them all to 3000k warm whites. Of course instead of being replaceable bulbs they were all SMDs directly soldered onto little PCBs making it a bit of a project … but so worth it!
The bad LEDs can also be improved with yellow or amber film, which is more accessible to most people than desoldering.
I found a German company that sent me a free sample of their amber film. As I recall their market was semiconductor manufacturing, but I can’t find them again. Amazon has a variety of films that should work.
If you look towards companies like Rosco or Lee they have hundreds of options for just this application.
A sheet of Rosco 3408 takes you from about 5500K to 3800K so a much warmer colour.
A small sheet which should be enough goes for about £7 https://shopwl.com/rosco-roscolux-3408-roscosun-1-2-cto/ you can probably find it cheaper just by searching Rosco 3408
I happened to find someone on a German forum (cars a VW) that had taken his overheads apart to do the same thing who posted a p/n or rather an industry size. I had never dealt with replacing SMD LEDs before but it seems there is an industry standard sizing that most adhere to. Just ordered a few in that size from Amazon and they worked. From what I remember most in that size were rated for similar voltages so seemed pretty universal. I imagine as long as you can see the led and measure its dimensions finding a replacement should be easy.
Desoldering/soldering was a pain as they were tiny; like 2mmx2mm. One cool thing I learned though is that you can use a multimeter in its ‘continuity’ setting to provide enough current to light up these small LEDs which is very convenient when testing if you soldered them correctly.
Desoldering / resoldering is not pain if you have proper tools to do it and know the correct technique. Trying to do it with an iron is making things very hard or even impossible (especially for smds with contacts below not at the sides). Using hot air rework station + preheater and low melt solder paste is the way to go - and it’s a piece of cake then. And you can do the whole board at once in a few minutes.
I modded most LEDs in my house because 2700-3000K is too warm for me, but 4000K is too cool. So I mix them ;)
Oh that’s good to know! Yeah, I definitely didn’t do myself any favors trying it with just a cheap unadjustable iron. I didn’t so much desolder the existing LEDs but instead just melted/scraped them off and cleaned the contacts. Was ugly for sure.
Do you by chance have any resources (videos or tutorials) to recommend that show how this is best done?
Thing is, when you order a small PCB, the price difference between 1 and 20units is usually negligible, so I'd try to find the light sizes for all my friends cars :D
Truth is, aftermarket lights for cars (typical AliExpress or Amazon stuff) always focus on brightness, but never on quality. It's usual for taxi drivers to have aftermarket lights in the vehicle, so you can find your money or credit card more easily, but the light quality is always terrible. :P
I've become somewhat expert at identifying cheap prefabulated automotive LED modules on AliExpress that seem unlikely to be problematic, and modifying them to be something other than awful and stupid.
It seems like such needless work. All I want are ~3000k LED widgets in various standard-automotive sizes that aren't stupid.
What I get are ~3000k LED widgets that are overdriven and painful to be around at night, which also have extra "CANBUS" resistors that only serve as heaters.
It's all pretty repulsive. Some recentl examples I got consumed 2.6 Watts, and about half of that was deliberately wasted as nothing but heat.
(2.6 Watts is quite a lot for a little LED board to dissipate, which results in early failures and flickering and other nonsense.)
But with the stupid heater-resistors removed, that dropped to only an eye-burning 1.3 Watts.
With the current-limiting resistor swapped to a higher value, power consumption was reduced to a few hundred mW and the light output approximated the relatively-unintrusive incandescent lamp that was originally fitted.
And finally, after all of that work of finding and modding the things, I can open the car door at night without becoming blind -- just like it was the 1990s again -- and also preserve the battery for things like starting the engine.
I'm curious what is the percentage of the population that prefers 3000k warm light. I personally disliked it. My mind constantly notices the distortion in color accuracy, and it's an unpleasant feeling. My preference is 5000k "day" light.
Not OP but I think it can be either. Great for the safety of one driver who can see the road perfectly can be awful for the safety of another who is blinded.
LED bulbs exacerbate an already existing problem caused by vehicle noses being too tall such that the headlight is exactly eye-height for other drivers. Sometimes they're poorly calibrated too, aiming too high. Many drivers also retrofit super bright LED lights on older cars that do a very bad job at properly focusing the beam on the ground (same thing was happening with drivers modding their old mirror reflector headlights with xenon lamps and giving everyone on the road an artists impression of a supernova up-close).
I've seen cars with lights that were obviously very bright but very well calibrated to point as much as possible to the road instead of my eyes. Those are maybe marginally more tiring than the warmer and much dimmer lights used in the past, but this is more than compensated in safety so probably a worthwhile trade. Proper calibration and sensible placing make all the difference.
Unfortunately I can't compare between 2 modern cars equipped with similar lights save for the color temperature. I haven't seen modern LED headlights with warm color, so when I think of warm color headlights my brain immediately goes to my experience with cars as old as the mid '70s, with filament bulbs. This evokes memories of almost blindly searching for the road right in front of me, compared to today's "I can see 200m ahead".
So by this measure, not having to squint to see anything ahead, my (aging) eyes feel much more rested and reassured today than decades ago.
While the color temperature situation has improved, the actual spectral quality of most consumer LED lighting still leaves a lot to be desired. CRI makes an effort to measure this, although it's a low-granularity measurement.
I personally buy surplus cinema lighting equipment and use it to light my house. I have a bunch of fancy cinematic LEDs with high CRIs that produce decent light, although they still can't fully compete with tungsten bulbs (e.g. Arrilite series)
Ideally you want to be looking for something with a color temperature in the 3500K (mid-morning) to 6500K (clear blue noon day) range and a CRI of 95+. Also bug manufacturers to start using better color quality metrics like TM-30.
Just get modules or strip from Yuju; I have zero bad experience over many years with them.
Sure, it's not cheap, but IMO ~1$/W plus simple PSU (either some standard voltage, or dimmable current, depending on what you got) for good and 1.5~2$/W for very good CRI is quite fine.
Yep. I’ve been replacing all the bulbs in my house with Yuji SunWave, and the difference is astonishing. I just put in four of their PAR30 4000K bulbs above my kitchen island, and the first time I turned the lights on at night I gasped at how much it looked like sunlight. (Then again, my wife said she couldn’t tell much of a difference and suggested my reaction was possibly confirmation bias, so YMMV.)
They’re quite expensive compared to normal LED bulbs, but it’s hard to argue against the quality of life improvement, particularly in the winter where I work remotely from a home office all day.
I was actually considering setting up an high brightness array of their full spectrum lighting above my desk. I know the article mentions diffuse lighting is best, but for some odd reason I prefer “spotlight” style. I don’t know why—it just feels cozier to me.
One thing I’ve noticed as I’ve learned more about lighting is that the brighter the light, the better cool color temperatures start to look. For a long time I hated anything above 3000K (too “office-like”) but 4000K actually starts looking pretty good around ~7,000 lumens directly overhead, and I imagine 5000K would look alright above 15,000 lumens or so (perceived brightness is a logarithmic function of luminance). IMO 5000K looks downright ghastly in a home setting for the typical range of 500-800 lumens that most bulbs sell at nowadays.
Craigslist. I just have a bunch of alerts for various cinema lighting brands (among other industrial goods I like using in my house, like pelican cases).
For mounting, it depends. Some things, like Pavotubes, come with screw-in mounting solutions. Other stuff may require dedicated photo mounting hardware. Manfrotto Superclamps and similar are a great way to mount lights to random pipes, beams, etc.
Also, most cinema lighting supports DMX control, which is nice for automation. It's harder to set up than Zigbee, but it works a lot better once you get it set up. That bit is definitely not consumer-friendly though.
Also, MediaLight provide great high CRI options bias lighting for your TV or monitor, further reducing your eyestrain when watching TV at night: https://www.biaslighting.com/
Ah mate, I'd settle for manufacturers to even publish colour quality of there lights, even if it's just CRI. At least here in NZ, like 99% of the consumer lights say nothing about colour quality, and many are pretty average (not to mention flicker).
If you're patient, it's crazy how much money you can save off of new pricing. I frequently see discounts north of 75% off new price for stuff in pretty good condition.
In general I think the population at large, and especially programmer types, know way too little about lighting given how ubiquitous both lighting and photography/videography is in our everyday life.
They claim to know a lot but then none of them build proper Cf Lumen or F.Lux style tooling (i.e. super easy default instead of requiring tons of setup) for turning ALL blue light off (red shifting) IN THE DEFAULT OS as the primary way to use your device at night. None of that wussy shifting the color temperature brown shit that doesn't meaningfully reduce blue light hazard.
I can literally preserve my whole night vision and have zero eye strain in pitch black conditions by red shifting my computer screen. No one supports this because we are stupid.
The laser pointer community of all groups understands this stuff well. They recommend green lasers because they use the lowest power for the highest visibility at night. Blue lasers are easy to get super powerful, but unlike a lower powered green one, which usually will only temporarily blind you if you make a mistake - a powerful blue laser will straight up destroy your eyes forever. Permanent blindness.
I don't really get that impression. To the contrary: How dangerous blue light is supposed to be is commonly repeated, with many products for filtering it out, like glasses, and software in all devices that does the red shifting. Android has it by default, Linux does, macOS - is Windows missing? Or is there something in the implementation of those measures that you object to?
But note that the wikipedia article you link states clearly that the eye strain attributed to blue light has no scientific evidence. That's also what I found out last time I looked into it - all those very strong statements about how dangerous blue light emission is supposed to be has no equivalent scientific studies that prove them. I think there was also a prominent article about that here on HN.
I think the effect might be completely made up, as it is something that is easy to believe and to subjectively feel that it works, by a placebo effect basically.
It's insanity to be told that your claim is basically psudoscience or fake when night vision preservation, the mechanism of prevented eye-strain, is a well confirmed and relied upon part of our biology. Go talk to astronomers. I'm aware of Wikipedia claiming what you say it does. That's an incitement of Wikipedia and the big blue light crowd.
Further, actual studies of filteirng blue light needed the right display tech. Before OLED, I couldn't actually "turn off the non red pixels", and thus never fully eliminated blue light from my display. Most of the "science" around blue light is trash. Yes, I say this as someone who actively published at top AI conferences and is intimately familiar with peer review.
I'm going to keep on preserving my night vision and browsing the internet for hours with zero issues at night while being told that my red shifted phone is placebo. You try turning your OLED phone on at full brightness in the middle of the night with and without the red-shift and tell me which hurts your eyes more...
There is no big blue light crowd. There has always only been the "blue light is bad" position, which was completely ignored by the industry (with display colour defaults especially) until that turned around completely, after the success of f.lux broke a barrier I think.
Please notice two things: 1. It's not even about whether blue light might be harder on the eyes in general, but whether it matters on the brightness levels consumer devices are usually used (so astronomers are completely irrelevant to that discussion, and so is a modern phone at night at full brightness) and what the actual effects are in practice 2. It's an astonishing position to take, to at the one hand realize there is no scientific basis for the claim, but still act 100% convinced. Especially for an academic.
It's also not obvious why "night vision preservation" might be relevant. We are not talking about looking at things during the dark, but at (more or less brighly) illuminated objects.
I don't have skin in this game and likely won't respond further. I even have a weak red light shift active while writing this (I feel like it might help with the sleep rhythm a bit). But I am absolutely saying that claims like this need scientific evidence and that it is foolish to believe in such things only because it's currently a popular claim, to take that position that absolutely. This is not the correct way of thinking, and there is no need to not leave space for doubt.
There are old studies on the hazard of blue light. I found a few of them on https://scholar.google.com - I suspect they’re hard to find if you don’t know about specific keywords to include in your query.
The blue light problem has to do with how high energy blue photons interact with specific substances found in minute quantities in our eyes, leading to inflammation.
> They claim to know a lot but then none of them build proper Cf Lumen or F.Lux style tooling (i.e. super easy default instead of requiring tons of setup) for turning ALL blue light off (red shifting) IN THE DEFAULT OS as the primary way to use your device at night.
Probably not what you are aiming for but for Linux this is a somewhat solved problems. The default DM, Gnome, has built-in color shift with a time schedule AFAIK. For Wayland and xorg there are also numerous other solutions that do a display-level redshift, I can recommend wl-sunset for Wayland.
> The laser pointer community of all groups understands this stuff well
This has been my major frustration also. Different enthusiast communities all discover these things independently and talk about them in their own domain language. Night photography and astral photography talks about red shifting as well.
Along light temperature there's light cri as well, which measures how faithful to real life a light is. New standards such as Tm30 came out recently as well
> I think a lot of people aren't privy to the difference the color temperature of "white" light can make
Go out and buy cheap Christmas lights and expensive ones, and put them up next to each there, and show them that. The difference is staggering. To the point where they look dumb if you hang them up together.
I want to go on a guerrilla campaign around my neighborhood and replace porch light bulbs with warm equivalents.