How to Build a Lumenator

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator

Once upon a time, a friend was sad. Specifically, they had seasonal affective disorder. They tried to fix it by adding lights to their room during the winter. It didn’t work. They tried adding a LOT of light. It worked. They called the giant bundle of lights they assembled a Lumenator. Other people wondered how they, too, might summon a sun into their living room. The task was not exactly *complicated *or hard, but it was a little confusing and inconvenient. Instructions were passed around by word of mouth, and individuals cobbled together lumenators in their own homes. Some of them has seasonal affective disorder, and some just liked their rooms to feel line sunshine all the time. Bit by bit, people’s lives grew brighter. Eventually, someone said "it’s silly that there are no instructions on how to do this on the internet and it has to be passed around as weird Cultural Metis." They said this, but then took no actions based on it. Then a second person said that, and they too took no actions based on it. And then finally someone asked me this weekend how to do it and I wrote this blog post. (Don’t be too impressed, because that *second *person who did nothing was also me) A lumenator is 24 lightbulbs, hung in a row from your ceiling. You may want a dimmer switch to control exactly how bright it is. You may want lantern-covers so that the light isn’t directly in your eyes (depends on the room in question). You will need the following material components (roughly $300 on Amazon). [UPDATE: many of these amazon links no longer work, and I think I got some pieces subtly wrong. I recommend checking out this recent UK post and this post to get more details)

Comment

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=yZNzQSQSZHHpAzAG8

It turns out there was an original article written on arbital.com, which I failed to find because I mispelled "lumenator". The links on what to buy are out of date, but more clearly convey which technical specifications are important

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=wB8JQFTEyXdqpYcdh

This is a good post, props for writing up a practical thing that people can refer to! This is potentially really useful information for people outside the community as well—lots of people struggle with SAD. Two small changes I’d want to see before I show this to friends outside the community:

  • Take out the word "rationalist" in the first sentence. This sounds like a small nitpick but I think it’s huge—It’s early and prominent enough that it would likely turn off a casual reader who wasn’t already aware or fond of the community. (And the person being a rationalist isn’t relevant to the advice.) Replace it with "friend", perhaps.

  • Add a picture, even just a crappy cell phone photo. How do you get the hooks to hang a cord from the ceiling?

Comment

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=Efp6XGX6CzBAkiu2f

Yeah. I think I’ll leave the current wording for a couple days for weird ingroupy signaling as the article gets it’s initial wave of attention, and then change it afterwards for posterity. I’ll do a picture in a day or two when I’m less busy.

Comment

pssst

Comment

lol I should stop promising this sort of thing. https://​​i.imgur.com/​​cPJNVo3.jpg

Comment

Ah, okay. It looks like your lumenator is hung from normal hooks on the ceiling. But if you wanted to use command hooks like you describe, you’d have to put it on the wall, correct?

Comment

Disclosure: this is actually my neighbor’s lumenator and I hadn’t checked which kind of hooks it uses. But yes, mine are hung from the wall. (The end result looks pretty similar)

Comment

Could you check? I’m worried the command ceiling hooks can’t take the weight from the ceiling.

Wow, this is pretty.

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=ndazhxeLPeuFPco2h

I wonder if anyone has considered re-purposing "broken" LED televisions for this purpose.
Often, as an LED TV ages, it becomes less efficient at eliminating the waste heat generated by its power supply, GPU, etc., as ventilation holes become clogged with dust. When the set is powered up, the internal temperature gradients become larger than they were when set was new, ultimately leading to localized thermal expansion cycles in the thin glass screen as the set is powered on and off. Eventually, the screen develops an internal crack (usually withing a month or six of the warranty expiring) which ruins the picture quality. When this happens, people usually just chuck the set. This happened to me several months ago, and, not wanting to waste some perfectly good hardware (I’d thought I might play around with re-purposing the GPU board) I cracked it open, removed the components I wanted to keep, and put it back together. Plugging it in, I found that I had an extremely bright back-lit tabletop, which immediately clicked in my head as something more useful than the GPU board (which, honestly was probably just going to end up collecting dust in my garage).
Reading this post got me to thinking, though. "Broken" LED TV’s (e.g., ones with cracked screens) can be had on the cheap, are super bright, and are wall- (or even ceiling-) mountable. Hell, they often can be had with a wall mounting bracket still attached. They aren’t unattractive, either. Thoughts?

Comment

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=bHkDEgTknSZ5zoJiv

You can now just basically buy this! https://​​www.lesswrong.com/​​posts/​​ZAMA4y6SbtFCTjjY4/​​a-new-option-for-building-lumenators

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=rA2RqTSxwEAcuQ9ds

https://​​www.youtube.com/​​watch?v=8JrqH2oOTK4

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=HQcn5ohjY9TZkRurv

There’s some concern in the other comments about the aesthetics of this solution, and some call for a pre-built solution from an installation-labor perspective. For those people, I suggest getting a "High Bay LED light". These are really bright hanging light fixtures… most rooms would be well served by 1-2 them, which come in two shapes: round "UFO" and "linear". I think they look pretty good, as the need to have good heat-sinking capabilities makes them one of the few products where even the budget producers have to use "quality construction". These are cost-competitive with the lumenator build suggested by the OP:The lumenator ($80 of bulb sockets, $40 of command hooks, $200 lightblubs) totals $320, consumes 380W, and produces about 40000 lumens.Modern high bay lighting solutions generally cost about .9$/​watt and produce about 130-140 lumens/​watt, so two 150W high bay light will cost about $300 and produce 39000 lumens. (As an added bonus, the higher bulb efficiency saves $4/​year @ 150 days x 3hrs/​day per year) Disadvantages: The CRI is generally a bit lower, around 85. The linked bulbs have a CRI of 92. The availability of 2700K fixtures is very poor. Most high bay bulbs are 5000K, with good availability of 4000K lights. Dimmer switch wiring is by a 0-10v logic voltage. This can be left unconnected to run at full brightness, or for one light, a 100k-ohm potentiometer works… for a single control operating two or more lights, the hydroponics industry seems to have produced a large number of inexpensive controllers. These are largely marketed at an industrial market, so be careful to buy one that already has a cord installed, or be prepared to do some minor wiring.

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=uYCS9PeHCT6ohfyng

Has anyone who has gotten relief by using luninators done rigorous a/​b testing with different temperatures/​colors or intensity or duration or other possibly important variables?

Not just gold standard clinical trials, something like "I tried color a for a week and logged 3 episodes, but color b for a week resulted in 8" could be informative for people deciding which type of bulb to get.

Comment

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=B4Qaqf4ie8LhLji2G

I thought the original builds of this used 5500k bulbs instead of 4000k bulbs. If you aren’t price sensitive, I would try kitting this out with programmable wifi RGB LED lights. That many LED lights in close proximity to each other can create cool effects and you could make a more pleasant light palate by having each bulb at a different color temperature, or having them change to simulate the… sun.

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=dp2fxXkFHFB38FpsL

It looks to me like you’re paying a substantial premium to get high CRI lights. It looks like the original models used lower CRI. Has anyone compared them, either in pleasantness or efficacy? I’m happy with 80 CRI Ecosmart (8-pack of warm from amazon, 2-pack of daylight from homedepot) about half of the per-bulb price you linked. The Home Depot web site is confusing and inconsistent, but I think that I was able to get free home delivery of a small order, just a single 2-pack, from that link. But if you live in California, HD may well enforce the ban on such bulbs.

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=EFgNCEgk3LcdNFaky

There’s also David Chapman’s version which is more neat-looking and possibly brighter, but not dimmable.

Comment

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=5ciKQLgCGcnZ6Q2FK

As well as his original: YOU NEED MORE LUMENS (Mon, 12/​07/​2015).

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=3CixGWWfowk6xtKaq

Other things worth trying are infrared saunas and infrared heaters. I think it’s great people engineered a solution that works for their seasonal affective disorder, but I’m skeptical this is a long term population level treatment. I am biased by how subjectively unpleasant I find such lighting systems. In terms of the true physiology of the human body, the visual light energy of the sun impacts the body much less than infrared and ultraviolet light. Therefore, a true treatment for SAD suggests light therapy on non visual spectrums. An infrared heater has much improved my winter quality of life. Now when it’s cold, I look forward to turning on the space heater, because it puts out pleasant sauna like energy instead of feeling like a blow dryer. An infrared heater puts off more negative ions than a conventional space heater. A meta review showed negative ionization of air impacts depression but their prior has them skeptical of the association without current biological mechanism. https://​​www.ncbi.nlm.nih.gov/​​pmc/​​articles/​​PMC3598548/​​ https://​​www.amazon.com/​​Dr-Infrared-Heater-Portable-1500-Watt/​​dp/​​B002QZ11J6/​​

Comment

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=EdGtj88FGMEag59L4

Are you suggesting blacklightboxes?

Comment

The science says unfiltered UV light is bad for you and unfiltered IR light is good for you… darklight box? But very unclear that staring around infrared light does anything, massive exposure to energetic IR light to you seems to be supported by evidence and human behavior in the winter.

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=K7NAXL6G2mFEdusR5

A lot of the links are broken, so here’s what I built recently with valid Amazon links as of November 2021. 5x these: https://​​www.amazon.com/​​gp/​​product/​​B01CEXCGDO/​​ref=ppx_yo_dt_b_asin_title_o04_s00?ie=UTF8&th=1&fbclid=IwAR3jjBCjqfiZ1lH1HEDOidodi2kOeoGYmYqdcsVwcDNG8b73NnGNJLSZAuo1x this: https://​​www.amazon.com/​​gp/​​product/​​B01D3SKICK/​​ref=ppx_yo_dt_b_asin_title_o05_s00?ie=UTF8&psc=1&fbclid=IwAR1_KO5KVMk1qd8bp716VBBQTOE2K_qEs5O4X6wABRGT1wvlKR0OmGeh_yw2x this: https://​​www.amazon.com/​​gp/​​product/​​B079QM8V1Y/​​ref=ppx_yo_dt_b_asin_title_o06_s00?ie=UTF8&psc=1&fbclid=IwAR2aPW2EmfEMrAJPoMtgbPkpwDA3Y3Yg7H4NIfAmSJiCwz0YyZi8QZ4pF5Q 2x this: https://​​www.amazon.com/​​gp/​​product/​​B079QQNY7T/​​ref=ppx_yo_dt_b_asin_title_o06_s00?ie=UTF8&psc=1&fbclid=IwAR3yotRrJLuhjrmiFeaRD4thnPoKhD2ArjKlD6D40SiYu7GSn5RQorDojyQ

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=BZoFDQeSPbadDkSFa

How useful would it be to have someone who produced luminators that were pegboards with lights mounted via zip ties or something equally aesthetically bad? If the labor of collecting and assembling the components can efficiently be outsourced into buying a nonstandard light fixture, it might be more accessible.

Comment

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=3D5uv2piKwXwgskT5

The main source of complexity for this sort of thing is diffusion: bright LED lightbulbs are unpleasant to look at, and light the room in an unaesthetic way. My thought would be to build lamps you can buy in sets of four, stick in the corners, and throw as many lumens as possible onto as wide an area of ceiling as possible without the bulbs being directly visible. White paint usually has an albedo of around 0.8-0.9, so you don’t lose much on the first bounce, and it’d light the room evenly with a much simpler setup.

https://www.lesswrong.com/posts/hC2NFsuf5anuGadFm/how-to-build-a-lumenator?commentId=2bmtvE8XkQa5XiZHS

There are just not very many components here and the design isn’t solved enough that I think that would make sense. There is however a business opportunity in solving the design challenge and then releasing an infomercial marketed light product better than existing solutions.