[HN Gopher] I tried making artificial sunlight at home
___________________________________________________________________
I tried making artificial sunlight at home
Author : fouronnes3
Score : 197 points
Date : 2025-03-27 19:49 UTC (3 hours ago)
(HTM) web link (victorpoughon.fr)
(TXT) w3m dump (victorpoughon.fr)
| NotGMan wrote:
| Those in dark winter places can check this out for a home bright
| light setup.
|
| Some people report much better mood in dark winter days:
|
| https://meaningness.com/sad-light-led-lux
|
| https://meaningness.com/sad-light-lumens
|
| Buy a cheap lux meter if you will be doing this since otherwise
| you are flying blind.
| astrange wrote:
| This definitely helps for me, but it's hard to place the lights
| without glare, since they can't be anywhere that's ever
| directly in your vision! I'm also a little worried it's going
| to burn the houseplant near it.
|
| But it was a big change; my Silicon Valley apartment has little
| overhead lighting, and I mostly use Hue lights which just don't
| get very bright, so it turned out I'd accidentally built a
| depression cave in the winter.
| dylan604 wrote:
| Personally, I hate overhead lighting and much prefer lamps or
| wall sconce type of lighting. I did replace my overhead bulbs
| with Hue bulbs, but I don't feel the not very bright aspect.
| My overhead bulbs are most often at 25%. Anything over 50% is
| just too much for me. I love my Hue lights and am trying to
| switch over to all Hue bulbs. I still have a few regular
| bulbs in places that just don't get a lot of use, so they get
| ignored
| J_Shelby_J wrote:
| The 1600 lumen bulbs are worth it - if they fit your
| fixtures.
|
| Still nowhere near as bright as a window on a cloudy day
| though.
| proudestmonkey wrote:
| This is super cool, and a great project.
|
| As someone more on the software side as well, I'm inspired to
| take on something similar.
|
| But more importantly, as someone who rented a windowless room for
| a year, I would have loved to make a smart light like this to
| wake up to in the morning.
| fouronnes3 wrote:
| > windowless room
|
| Thankfully that's illegal where I live. I'm even tempted to
| suppress my own work here, because I don't want it to be used
| to unlock new levels of dystopic housing.
|
| Just kidding, thanks for the feedback!
| ge96 wrote:
| The interactive 3D models are nice
| woah wrote:
| Would it be possible to make it look like the light was coming in
| at an angle from above and the side like real sunlight?
| fouronnes3 wrote:
| I've given some thought to this, I agree it's one of the
| obvious improvements. I think with a refractive design like
| this one it's difficult, because of the way a lens works. Most
| other projects like this including commercial ones like CoeLux
| use a reflective design, and I believe that's partly for this
| reason.
| hinkley wrote:
| I've seen the video he links at the top of the article, a few
| years back. They took an old satellite dish, silvered the
| surface, and put a giant LED where the receiver had been so the
| light produced was nearly parallel. Otherwise the shadows look
| wrong.
|
| In his case he sacrificed a doorway or a closet to make a
| window with "real light"
| j_bum wrote:
| I was thinking of the same video! Link:
|
| https://youtu.be/6bqBsHSwPgw?si=nV7nGdRhGvkivc1p
| ericmcer wrote:
| There was a project posted here awhile ago with something
| similar, but the light was larger and he positioned it outside
| a fake window with a curtain across it.
| goda90 wrote:
| Would moving the box itself help provide that effect? Maybe put
| it on a track to move throughout the day?
| stephenpontes wrote:
| DIY Perks also attempted to make artificial sunlight at home,
| also focusing on things like Rayleigh scattering! This was a
| great watch:
|
| https://www.youtube.com/watch?v=6bqBsHSwPgw
|
| EDIT: After reading the article, I see the OP calls out DIY Perks
| specifically - the OPs design is much more compact :)
|
| > It's compact. The total size is 19cm x 19cm x 9cm. This is
| quite compact for a 5cm focal length and an effective lighting
| area of 18cm x 18cm. Reflective designs like the DIYPerks video
| or commercial products like CoeLux do not achieve this form
| factor.
| softbuilder wrote:
| This is a great build and report. Even though you don't have
| great photo gear, you should try capturing a picture of steam in
| the light (as DIY Perks did).
| fouronnes3 wrote:
| That's a great suggestion, thanks! I'll give it a try. It
| already works with dust quite well though, hehe.
| brcmthrowaway wrote:
| Wait, you can order custom fabrication of lenses online?
| fouronnes3 wrote:
| Turns out you can yeah! I'm not sure the quality would be
| anything useful for imaging applications though. But for
| lighting / non-imaging stuff go for it!
| isoprophlex wrote:
| Well, there goes my idea of DIYing a telescope...
| hatthew wrote:
| DIY telescope discussion on HN 2 weeks ago:
| https://news.ycombinator.com/item?id=43351988
| brcmthrowaway wrote:
| What was the cost details?
| jclarkcom wrote:
| Very cool. I'm the CEO of Innerscene (https://innerscene.com) and
| we make a commercial artificial skylight that uses some of these
| concepts. Actually the coelux ht25 model is almost identical to
| what you made but using smaller lenses and more LEDs - however
| the effect they were able to achieve still isn't that great, the
| sun looks like a giant orb and once you get a few feet away you
| can make out a sun at all. We spent a lot of time working on
| perfect collimation and hiding lens edges and making sure the
| view into the sky was seamless and artifact free. I'd say the
| last 10% of that problem is 90% of the work. :). I think we
| successfully cracked the nut but currently using a lot of
| expensive parts so working on brining the cost down. If you
| search Innerscene patent many of our approaches are spelled out.
| We also spent a lot of time on simulation and software...
| fouronnes3 wrote:
| Are you hiring? I'm looking for a job currently. Contact info
| is on my website :)
| jclarkcom wrote:
| Let's chat! Are you the article author?
| fouronnes3 wrote:
| Yes that's me! Happy to chat :)
| the_arun wrote:
| This is a great way to pitch for a job & half the
| interview is already done.
| rambambram wrote:
| > If you search Innerscene patent many of our approaches are
| spelled out.
|
| Subtle, nice. Maybe you can give the man a job. ;)
| sheepscreek wrote:
| "Virtual Sun" looks really cool. How much does one cost today?
| Say if someone wanted to buy a single piece or a few (2-3).
| dan-robertson wrote:
| Pricing is 'contact us' and going to be region-dependent. I
| would expect to pay ca $ 10^4.5 for a panel that produces the
| 'virtual sun' effect. Same goes for competitors (I know of
| Coelux but maybe there are others. Not sure how quality/price
| compares). But maybe GP will reply to you.
| jclarkcom wrote:
| Yes, it's region specific. In the USA sales generally go
| through a partner. But it is a fraction of the weight,
| depth, and cost of the large coelux units that run $40k
| plus install. We also realized there is a need for
| producing just the sky with great photo metrics and that is
| even thinner and less expensive. A lot of interesting
| reading can be found in our interactive spec sheet here:
| https://www.innerscene.com/products/circadian-
| sky/CircadianS...
| kulahan wrote:
| Make them small, add heat, attack the pet market >:)
| jclarkcom wrote:
| Haha - actually one of our customers is a doggie hotel
| chain that has installed them in multiple locations
| anthk wrote:
| Hope Oracle doesn't sue them somehow ;)
| klysm wrote:
| How do you guys go about doing sheet metal design? It looks
| pretty clean
| jclarkcom wrote:
| Thanks! It's powder coating that gives it a super clean look,
| great for sales demos but I'd suggest that anything not
| visible from the room side should be cost optimized
| klysm wrote:
| Yeah definitely, I'm more referring to the bending process,
| material selection, fastener types, etc. It seems like
| quite the art form
| Intralexical wrote:
| This is simultaneously really cool technology, like a hyper-
| specialized analog light field display with health benefits,
| and yet there's also something dystopically unsettling about
| the sun being faked by a machine in a box.
|
| Please don't have your marketing department destroy the real
| sun. Doing so would break antitrust regulations.
| cannonpr wrote:
| I adore the lights by your company, though they seem to be
| incredibly hard to source in general except for high end
| architectural projects. I wish there was an easier way to order
| them directly for DYI inclined engineers willing to pay the
| price.
| jclarkcom wrote:
| Yeah - channels can be a pain, reach out to me directly if
| you run into issues: Jonathan @ Innerscene
| kulahan wrote:
| It's infuriating that you only have "accept all cookies" and no
| other options.
|
| Oh but I do absolutely love this concept. I'm curious if you've
| had lots of interest from these billionaires building bunkers,
| if that story is even true!
| nmarinov wrote:
| Hi, I couldn't figure it out on your site but on the picture
| where you have your hand under the virtual sun[1], do you feel
| the warmth of the "sun" on your hand as you'd feel it with the
| real sun?
|
| [1] -
| https://www.innerscene.com/_next/image?url=%2F_next%2Fstatic...
| huzeifad wrote:
| Great project! Instead of routing ground traces, I would
| recommend to use a ground plane on the bottom layer of the PCB
| Const-me wrote:
| > the main thermal issue when scaling up would be the cooling of
| the power supply itself, not of the lamp
|
| If I would be scaling up that device, I would consider an ATX
| power supply. These are relatively large and typically include an
| active cooler inside, but they can easily supply hundreds of
| watts at 12V, often have an on/off switch on the back, are
| relatively inexpensive (at least unless you need much more than
| 500W of power), and are available everywhere. Usually, you just
| need to connect the PS_ON wire with the ground to make them turn
| on once powered.
| brcmthrowaway wrote:
| Could you wire them in parallel for more watts
| Const-me wrote:
| Personally, I would avoid it if possible. Even if they are of
| the same model, small discrepancies may cause their +12V to
| be slightly different. At the very least, will cause very
| non-uniform load distribution.
|
| Luckily, seems the OP only needs one. The current light only
| uses 36W @ 12V, even if they make the new light 10x more
| powerful, a single 400W PSU should do the job nicely.
| foobarian wrote:
| It's not like the fixtures consist of a single LED either.
| Should be trivial to parallelize.
| brcmthrowaway wrote:
| What about wiring them in parallel with a voltage regular
| after.
| mystified5016 wrote:
| Then you waste a ton of energy on the conversion and
| still have the same problem. You won't ever get two
| regulators at _exactly_ the same voltage. One will
| _always_ take most of the load.
|
| You get around this with load balancing resistors, but
| that comes with its own set of problems.
|
| The way you get around _that_ is to chop up your load
| into multiple independent power domains. That way each
| segment is powered exclusively by one supply.
|
| If you can't do that, you will _always_ be better off
| with a single larger supply.
| wtallis wrote:
| At that point you might be better off getting a power
| distributor from an old server that was already designed to
| operate off redundant 12V power supplies. But you wouldn't
| want the server PSUs themselves due to the tiny loud fans.
| petsfed wrote:
| Is there a reason why you went with traces rather than pours? I
| count 7 signals per board, and they're all meant to be low
| impedance. You could even expose copper on the back of the board
| to be used as ad hoc heat sinks without spending any extra money.
| The weird little triangle loops on the back really stand out to
| me, even though you probably don't need to worry about the
| impacts of a loop in your circuit.
| fouronnes3 wrote:
| Thanks for the feedback. Everything I know about PCB design is
| self-taught. I'm a total beginner, so I'm sure there's a lot to
| improve!
|
| There's only 2 routes per board, VCC and GND. I initially
| planned for SMD header pins that I didn't end up using, because
| soldering wire on bare pads was good enough. I also planned for
| 8 connection pads per PCB, but only used 2 to 4 in the final
| assembly. So yeah, lots of room for improvements in the PCB
| design! Definitely would need to spend some time on it for a
| higher power version 2.
| null_name wrote:
| Really neat, thank you for posting. I took a different angle when
| trying to figure this out - my goals were high spectral
| similarity to the sun, high brightness, adjustable warmth, and
| low cost. The existing lower-cost solutions people have posted
| about end up with high brightness, but the lights are evil.
|
| I ended up with a photography light that's /alright/. It's not
| nearly as bright as I want, and I can't automate changing the
| warmth. When I next take a crack at this, I'll look deeper into
| some of what you've posted about here.
| fouronnes3 wrote:
| Thanks! Feel free to reach out, I love to chat about this
| stuff. Contact info is on my website.
| jddj wrote:
| What were you missing for the colour temp control?
| sberens wrote:
| Really cool! I'm working on a lamp that gives you daylight levels
| of light indoors (albeit no raleigh scattering and columnated
| light). On the bright side (pun intended), it's 50,000 lumens
| instead of ~4500. https://getbrighter.com/
| polishdude20 wrote:
| Oh this is cool! Reminds me of led grow lights. You can get a
| panel with hundreds of LEDs on it for growing plants for like
| $200
| izuchukwu wrote:
| Had the chance to see one of these in person - did not believe
| you could achieve "daylight indoors" before I did.
| polygot wrote:
| Very cool! I'd be curious to see an HDR version of it using
| bracketed exposures--it might give a better sense of how it
| actually looks in person. It seems really bright in the photos,
| so the shortest exposure would probably need to be very low to
| capture a good dynamic range for the HDR.
| alejohausner wrote:
| For me, the problem with this setup (and with most high
| efficiency LED lights) is the lack of red wavelenghts. Real
| sunlight has a substantial amount of energy in the very red end
| of the visible spectrum (700 nm) and also of course quite a bit
| in the infrared. These lamps have two spectral peaks: a narrow
| peak in the blue range, around 450 nm, a broader peak in the
| green, centered around 580 nm. That greenish peak falls off
| sharply, and has almost no energy in the red end.
|
| The color sensitive cones in our eye have three peaks of
| sensitivity, the S cones in the blue range, the M cones in the
| green, and the L cones in yellow. The L cones are what your brain
| uses to see red colors, but they are actually pretty insensitive
| to deep reds like 700 nm. That's why you THINK that LED lamps
| produce red, because they stimulate your L cones, but they do so
| without actually emitting much red energy at all!
|
| Our bodies are sensitive to deep red light. The cytochrome in our
| mitochondria respond to it. There's an experiment where shining
| red lights on the skin improved sugar metabolism. That makes
| sense, because we naked apes evolved under red-rich sunlight.
|
| So these lamps may look like sunlight, but they're missing some
| crucial wavelengths.
| fouronnes3 wrote:
| That's a good point, but not much I can do about that. Such a
| DIY project is limited to off the shelf LED suppliers. It would
| be cool to do such a lamp with both the high CRI and some
| infrared, also for heating (infrared lamps are a thing after
| all).
| crazygringo wrote:
| > _So these lamps may look like sunlight, but they're missing
| some crucial wavelengths._
|
| Which means they're also not going to give me a tan... bug, or
| feature?
| lll-o-lll wrote:
| Tan is the other end of the spectrum (UV)
| crazygringo wrote:
| They said crucial wavelength _s_ , plural. I took that to
| its natural conclusion. ;)
| TeMPOraL wrote:
| It also won't be a great disinfectant, both in literal and
| figurative senses of the phrase.
| drewolbrich wrote:
| My issue with this setup is that it doesn't emit as many
| neutrinos as the Sun.
| ketralnis wrote:
| Have you considered taking neutrino supplements?
|
| The nice thing about them is that they're flavour changing
| aljgz wrote:
| Good that you get your sunborn nutrino fix even in cloudy
| days (and much of it even at night)
| ankitml wrote:
| Yep, that was my first assessment. Cant call it sun without
| NIR.
| fouronnes3 wrote:
| Can't call it sun without gravitational confinement nuclear
| fusion either.
| firju55 wrote:
| If the metric you're looking for is most accurate spectrum
| including deep red, near and far infrared (heat), then a good
| old incandescent filament light is the most efficient device.
| Not LEDs.
| ludicrousdispla wrote:
| Nice, Mojang should set one of these up at their headquarters.
| vincekerrazzi wrote:
| Honestly this area is so interesting to me because of how
| incredibly strongly light (or lack thereof) affects in the
| winter. It also seems to affect me more strongly than most in
| regards to sleeping patterns. Too much blue light in the evening
| and bam, brain refuses to sleep until 2am or later.
|
| So for me the biggest factor missing in these kinds of projects
| is a dynamic color temperature. While we get that from products
| similar to apples adaptive lighting, that's missing in products
| like this. It seems we can only really have one or the other.
|
| My dream is something like this build but with full adaptive and
| programmable color temperatures based on time and seasons.
|
| Honestly I want to build something like this but my area of
| expertise is too far into the software domain, and not very far
| into the hardware or electrical engineering domain.
| fouronnes3 wrote:
| This is the reason I decided to use 4000K instead of the more
| physically accurate 5400K. I picked a less middle of the day
| color temperature and closer to evening time, basically.
|
| I spent a lot of time looking at LED suppliers and there's
| quite a range of options in terms of color temperature. Some
| manufacturers even have 2 color temps on the same LED package,
| meaning you can sort of do what you describe by mixing two
| light sources at a distance less than a millimeter.
|
| > Honestly I want to build something like this but my area of
| expertise is too far into the software domain, and not very far
| into the hardware or electrical engineering domain.
|
| That was me last year before starting this project! You'd be
| surprised what you can achieve with time and effort!
| johnea wrote:
| Yea, I can relate. I tried making an artificial Big Bang at home.
|
| But my neighbors kept complaining 8-/
|
| Haters!
|
| Probably the same people that keep flagging comments on the
| articles I'm interested in...
| hubraumhugo wrote:
| That's the most HN headline I've read in a while
| 14 wrote:
| Very cool. I do think the DIY perks version just pops a little
| nicer but it could also be the photography of the project. But
| the firm factor if your project is much nicer as well. I think
| both projects would be better suited for different needs so again
| very cool. If you had lots of space the diy perks would be the
| way to go. If you only have a thin wall or ceiling then yours
| would be best. All in all thanks for posting this.
| fouronnes3 wrote:
| It's partly the photography but definitely the reality as well.
| Mine is only 0.03m2, the DIYPerks parabolic reflector is 1.13m2
| :)
|
| One of my goals was to explore the feasibility and scalability
| of a refraction based design.
| trallnag wrote:
| I've just mounted tons of full spectrum LEDs in my home office.
| Off the shelf, fairly cheap, easy to replace. My plants like it,
| except for the Goeppertias that keep dying on me
| davidw wrote:
| Ah... the intensity, not actual nuclear fusion. Well, still a
| nice project!
| simonebrunozzi wrote:
| At first I thought the same! Well, we're safer now.
| Molitor5901 wrote:
| Sunrise alarm clock lamps are a really great way to wake up,
| especially if you have a larger room.
| spl757 wrote:
| I'd be interested in seeing the optical spectrograph of the LEDs.
| If you want to simulate sunlight you want a full-spectrum LED
| like a Samsung LM301 series LED which are popular in grow lights.
| Not all LEDs are created equal, and even the LEDs in many "grow"
| lights only show two sharp peaks at red and blue wavelengths. A
| full-spectrum LED will output colors across the visible spectrum
| of light. You can't tell by looking at them, so you can either
| buy ones from which you trust the manufacturer or do what I did
| and build a cheap optical spectroscope using a raspberry pi with
| a small camera attached, a spectroscope lens, and some python
| code. I'm sure there are guides you can find with a quick web
| search if interested in making one.
| atoav wrote:
| This is covered by the CRI95+ value, note that the LM301 you
| mentioned only has a color rendering index of 70. Maybe it has
| predominant wavelengths that are relevant for plant growth, I
| don't know, but a CRI of 70 isn't impressive at all.
|
| For an accurate rendering of the suns spectrum you basically
| would like to simulate the spectrum of a blackbody radiator
| with a surface temperature of 5500degC minus the absorption
| bands of water vapors, atmospheric gases thst are typically
| inbetween the sun and us. Also note that the suns spectrum
| extends both above and below the visible range, which gives you
| the feeling of warmth (infrared) and tan/sunburn (ultraviolet).
|
| In reality most commercially available LEDs still have a
| extremely spikey spectrum compared to sunlight -- this can be
| somewhat fixed by mixing different LED types and adding
| filters. But this is only done in extremely expensive movie
| lights like Arri skypanels.
| nayuki wrote:
| Nice project. A few miscellaneous notes:
|
| > Kinda cool that you can see a lens flare effect in the shape of
| the lens grid array.
|
| A lens flare is just a copy of the scene you are photographing,
| but strongly attenuated in brightness, and possibly rotated
| 180deg.
|
| > 100,000 lux
|
| Your notation is correct, and there are other ways to write it.
| The SI unit "lux" has the symbol "lx". Your quantity can also be
| written as 100 kilolux or 100 klx.
| reaperman wrote:
| I saw "artificial sunlight" and thought "oh wow I'd love to see
| the spectrogram of the lighting solution this person came up
| with". I was disappointed to merely see "CRI 95+".
|
| https://www.youtube.com/watch?v=lH_owRxupC0
|
| This is great video on the shortcomings of "CRI" - it explains in
| detail CRI, CRI extended, TLCI, TM-30, and SSI.
|
| Brightness and color temperature are only two small parts of
| lighting - more people should start investigating the utility of
| taking their own spectroscopy measurements to figure out what
| lighting works best for them personally. My friends have very,
| very diverse opinions on what spectral distributions they
| like/hate, but they lack the language and experience to identify
| or communicate their preferences except for "Ooh I like/hate this
| bulb".
|
| I mostly use LED bulbs to keep heat generation down (I pay for
| the heat twice in Houston: once to generate it and again for the
| A/C to negate it). But I always mix in a bit of incandescents /
| halogens (2400-3000K) which provide full-spectrum blackbody
| radiation to see ALL the wonderful colors in my world.
| kulahan wrote:
| I think most people at least know the difference between warm
| and cool light, which helps a lot, but otherwise I agree. As
| I'm reading this, I'm realizing I have no vocabulary for this
| topic. That's... kinda strange to experience!
| fouronnes3 wrote:
| Do you have any insight on how to take spectroscopy
| measurements at home on a <1kEUR budget? And how to select an
| LED manufacturing supplier when CRI is often the only thing
| available on the datasheet?
| reaperman wrote:
| Starting points for first question: Look into i1Pro (later
| models of the first generation), which can be had for
| $200-400. Combine with some free or $99 Windows/Mac/Android
| software [0] [1].
|
| Second question: It is still too hard even to find CRI for
| most offerings. It's pretty much a "buy, test, return the
| ones you don't like" situation. If independent reviewers
| start publishing spectrograms and making YouTube/etc videos,
| perhaps the industry will move forward some day.
|
| 0: https://www.argyllcms.com/doc/instruments.html#i1p2
|
| 1: https://rmimaging.com/spectrashop_brochure.pdf
| Vox_Leone wrote:
| This project is fantastic, but you also deserve congrats for the
| 'Lens Maker'[0].
|
| "If we have a compound optical system made of a series of lenses,
| mirrors, etc., we can treat each optical element as the layer of
| a neural network." Kudos.
|
| [0]https://victorpoughon.github.io/torchlensmaker/
___________________________________________________________________
(page generated 2025-03-27 23:00 UTC)