Comments on “Seriously bright light vs. winter blahs”

SAD therapy glasses

Jacq 2015-12-25

Hello David, have you experimented with SAD therapy glasses? (Me personally, I haven’t yet, but I have heard some positive mentions so far.)

SAD glasses

David Chapman 2015-12-26

Hi, no, I haven’t tried them. It’s an intriguing approach. If you do, or if anyone else reading this has experience, I’d love to hear about it!

Metal Halide - UV

Jim 2015-12-28

Hi David. Great lamp. I am 100% like you.. I start to get down in October, and then by December can hardly think. I take meds in order to function. It really gets to me. I recently bought a 1000w HPS and MH grow light to combat my SAD with. I tried regular happy lights but they weren’t the intensity I felt I needed. The 1000w grow lights are amazingly bright. It helps like nothing else. BUT - I can’t use it anymore due to the UV radiation. I only used it a few days for like 10 minutes. The MH and HPS bulbs emit UVA, UVB, and UVC which will cause skin cancer. They are grow lights afterall.. I am searching for a solution that is equally bright (100k lux) but will not emit the UV range. I’m trying to find either an enclosure to block the UV or just buy something else. Your post gave me a couple of ideas. I recently saw the LED strips the other day at a Walgreens display. They are super bright. Thanks for posting this. It really helps. I have been looking into the melatonin suppressing pills too. One of these days I will be SAD free!


David Chapman 2015-12-28

Hi Jim,

Thanks for pointing out the UV problem! I’ve added it to the list of issues with metal halide bulbs in the post.

I believe that you can block the UV with certain type of glass, but I don’t know details. If you find out, it would be great if you posted it here!

A whole lot of LED strips glued side-by-side on a board seems like a good approach. It would take a long day’s work gluing and soldering and wiring and so on, but probably not more than one day. If you try that, I’d also love to hear about it. The 30k lumens I have are great, but I wouldn’t mind augmenting them with a lot more!

More about LEDs

David Chapman 2015-12-30

(I had some questions by email, about building a 100k lumen light, that I thought I'd answer here in case they are useful to several people.)

The "bulbs" I used have a small fan built-in. It makes a bit of fan noise, which isn't ideal, but I don't find it bothersome. They don't claim to be dimmable. Most "non-dimmable" LED bulbs can be dimmed using a special LED dimmer, but the fan circuitry in these might not work with one.

I've done a little more thinking about the strip light approach. I experimented with these strips; they may not be the best, but they certainly produce a lot of lumens per dollar! 14,000 lumens for $17 (up from $13 when I wrote the post a month ago). So you could could get 100k lumens from seven strips for not much more than $100 (plus power supply). They are dimmable with a PWM dimmer; I used this one successfully in experiments, but you'd need a bigger one to handle seven strips.

Each strip is 5 meters long, so seven strips is 35 meters. You can cut them to custom lengths. (I've done this successfully; it's work, but works.) So if you mounted them on a meter-square back board, there would be 35 strips one meter long. A meter is about 35" so if you put the strips at one inch spacing they would cover the board evenly.

The main thing I'm not sure about is heat dissipation. Each strip takes 90 watts (apparently—it might be good to check both this and the lumen rating some more before committing). So that's 630 watts total; quite a lot of heat. Spread over a meter square, it will probably cool itself into the surrounding air OK, but I'm not sure of that.

(If any real engineer reading this can answer that question, I'd much appreciate it!)

If passive cooling is inadequate, some rethinking would be required—and this is the point where I decided to just get the bulbs for a quicker, easier, but probably less-good solution.

One approach would be to mount muffin fans at the bottom of the board, to blow upward over the surface. Would that do the trick? I don't know.

If not, another approach would be to use a metal mounting board with heat sinks on the back. However, the LED strips are rubbery plastic, and they might not transfer heat to the mounting board well. Also, there's the issue of how the strips are attached to the board. The strips I used come with an adhesive backing which is awful. It is not strong enough to attach the strips to a board, but it does get all over everything else. (Apparently other brands of LED strips do not have this problem.) Several people in Amazon reviews recommend using a strong double-stick tape instead. Anyway, the double-stick tape probably wouldn't transfer heat well; and whatever other adhesive you use might not either, and also might melt or burn if a metal mounting board got too hot.

A sensible strategy might be to start by mounting 3.5 strips at 2-inch spacing first, for about 50k lumens (which is probably enough!) and 315 watts (which I'm pretty sure would cool passively OK). And if that doesn't seem to be enough light, and it doesn't get very hot, add another 3.5 strips, running the new ones in between the old ones.

1000W Metal Halide Work Light - 110,000 Lux

Jim 2016-01-04

Hi David. Thanks for the reply. I’ve been doing some searching for ways to achieve 100,000 or brighter lux and came across this:

It’s a 125v metal halide work light. It’s designed to be used indoors and should have the appropriate shielding (big assumption). It is expensive, but for a one time purchase it might do the job nicely. And the bulbs should last a long time.

I also looked for other metal halide work lights on Amazon and found a bunch. Most are 400W but they have a few 1000W lights as well. This one in particular looks good: Designers Edge L4210 1000-watt 100,000-Lumens Industrial Metal Halide Power Light, Orange

Once again, it’s expensive, but I agree with you - You need more lumens! I’m going to contact the companies about a couple of these lights and see if they are safe to work in from close proximity. I will report back. Let me know your thoughts!

Metal halide work lights

David Chapman 2016-01-04

Those are intriguing! I don’t have any experience with them, but some thoughts anyway:

The nearest experience I have is with “conventional” halogen work lights, such as this one. Conventional halogens run about 20 lumens/watt, so I ran two 1000w lights for 40k lumens. That was reasonably effective for SAD. A bit less so than my current LED lamp, because I can get much closer to the LEDs than to the halogens. They run very hot, so you don’t want to get too close.

The good thing about the halogen solution was that it was cheap—the ones I linked are $55 currently. The bad thing is that they burn out quickly; within maybe two or three weeks of regular use. Sometimes the bulbs fail; those can be replaced inexpensively but it’s a hassle. Sometimes the fixtures fail. I got tired of buying new ones every few weeks.

The 1000w metal halide lamps would also run hot, but at 100k lumens, you could be ~4 feet away and still get lots of light. That would probably work well. I sort of assume that at that price they are built to last!

A 1000w lamp is a fire and burn hazard. Just be sure not to leave them turned on unattended, or let kids get near them.

I’d love to hear how this works out if you go ahead with it!

LED strips

Ivan 2016-01-22

Thanks for the link to those LED strips. I think I finally have adequate lighting in my room for the first time and it’s life-changing. I was worried the ‘warm white’ 3000-3500K version would be too warm, but it works really well. My eyes think the color temperature is a lot higher than the spec, but a calibration test with a monitor suggests the spec is about right and there’s an optical illusion going on.

My lazy setup involved wrapping the LED strip around a water bottle, so it looks essentially the same as your corn bulbs.


Ivan 2016-01-22

The shipping takes longer, but AliExpress has the same LED bulbs and strips with lower prices than Amazon. A search for “100W LED Corn 10000” finds a quantity 4 lot of that corn bulb for $382.70 (3000K version) or $398.47 (5000K).


David Chapman 2016-01-23

Great to hear! Thanks for the info!

LED strip is probably closer to 4000 lumens

Ivan 2016-01-23

Talking it over with some people, I realized that 14,000 lumens for the LED strip is outside the realm of plausibility. An Amazon reviewer notes the strip uses ~38.6W and my measurement says ~33W. 14,000 lumens/33W = 424 lumens/W which is beyond the theoretical “maximum efficacy […] in the range of 250–370 lm/W”

Going by this graph of LED lumens/watt efficiency, and by Wikipedia’s maximum efficiency for a 5730 SMD LED, 33W * 110 lumens/watt means the LED strip is probably producing ~3630 lumens. Which is still ~4.5x brighter than the typical lightbulb.

I believe the Amazon seller is playing a game where they indicate “40-45 LM” for a group of 3 LEDs.

Now I know why the corn bulbs cost so much more!

Powering the LED strips

Ivan 2016-01-23 suggests that these LED strips have to be overvolted and powered at multiple points, which might bring them closer to ~7000 lumens.

I’m also going to try comparing the hkbayi Super Bright to LEDMO’s strip with 600 LEDs:

LED strip probably mis-spec'd

David Chapman 2016-01-23

Well, that’s disappointing!

If they don’t burn out quickly at 14.2V, it might still be a cost-effective solution (although more work to wire up). (BTW, your idea of wrapping around a bottle is clever!) (Also BTW, part of the expense of the corn bulb is the integrated fan cooling system and a fairly elaborate mechanical support for the LEDs.)

The LEDMO strip would be 9,000 lumens if 15lm/LED is accurate, so that looks good. I’ll be interested to hear when you compare them!

The first review of that compares with “premium professional-grade” ones. I did look into those a bit… they are a lot more expensive, but have higher CRI. Not sure how much of a difference that would make in practice.

There’s probably a business opportunity for someone to manufacture a high-lumen SAD light… although it may take a while for SAD sufferers to catch on that they are dramatically superior to the ones that have been available commercially so far.

Light spectrum is significant

Frank 2016-01-24

There is one critical bit of information you should put into consideration: a large part of the effect of light therapy for SAD is the restoration of the melatonin down regulation during the day. This mechanism relies on blue-green light (446–477 nm). So you should obtain a spectrum plot for the LEDs or tubes you use and check if there is not a “valley” in the plot at these light colors, which is the case on quite many products. In my experience more light intensity is less important then making sure the melatonin-supressing colors are there.

Another avenue you might want to check into is vitamin D3 supplementation. They compensate the other vital aspect of sunlight, the vitamin D generation from UV.

For more details see

Thanks everyone

Don 2016-01-24

I can’t thank everyone enough for the research and time. I have a commercial light box mounted next to my monitor at work and it does make a big difference. In addition, I wake up in the morning, take two 4000 IU of Vitamin D3 as a supplement to help and I can tell you, this winter is moving along MUCH better than ever before.

Now I am encouraged to step up my game!

The link to
is good advice.

Get some light, take your vitamins, get some exercise and talk to your friends and loved ones!

This will work and your life will be better.

The darkness you may feel isn’t you.

Don’t let it continue being an excuse for mediocre behavior, a shroud for hiding, or more importantly a voice of despair in your ear.

The voice of despair is a liar


Dawn Simulator

jim 2016-01-24

Don’t forget to use a dawn simulator alarm clock. In some studies the dawn simulator has been shown to be more effective than bright light therapy. Why? Because it starts working on shutting down your melatonin while you are asleep. So you’re getting light therapy and you aren’t even awake for it! It really does help. And it’s a lot better way to wake up as opposed to a traditional alarm clock. Hope this helps!

Consider 100W COB leds

Walking by 2016-01-24

You can buy 100W led chip-on-board units for $10 these days, and a corresponding power supply for $30. Attach a computer cpu cooler unit and you can get a lot of lux.

Chip-on-board cooling

David Chapman 2016-01-24

Thanks! I did get one of those when they first came out (three-ish years ago) and did some experiments. They do put out a LOT of light!

Unfortunately, I wasn’t able to find a way to cool it practically. A CPU cooler was definitely not sufficient; neither was a really big passive heatsink. This interesting video shows a series of experiments with trying to cool one; he eventually concludes it’s not really feasible.

There are now commercial units that house high-wattage COB LEDs in a big metal box for use as an outdoor (night-time) floodlight. I expect these get quite hot, but dissipate the heat over enough metal that it doesn’t fry. (These didn’t exist when I got my 100W COB.)

I haven’t looked hard for the best price, but these aren’t particularly cheap. For example, this one is 250W, rated 25,000 lumens, for $359. That’s about the same price per lumen as the corn bulb solution.

For indoor use, you’d have to design and build a stand for it. I’d be really interested to hear how it works out if anyone tries this!

High-lumen SAD lights are a Kickstarter waiting to happen

Ben W 2016-01-26

I’m positive that a consumer-oriented high-lumen SAD light, accompanied by a smart explanatory video, would blow up on Kickstarter. And it would make a lot of lives better!

I'd like more evidence first

David Chapman 2016-01-26

Ben, that’s an excellent idea. I’d thought casually about how to create a business around this, but Kickstarter didn’t occur to me. (Duh. It’s the way many hardware startups have funded themselves recently!)

Before selling anything, I’d want more evidence for more lumens being better. It certainly seems to work for me; but placebo effect, and other cognitive distortions, are powerful. I could be fooling myself.

A well-controlled dose-response study would be ideal… but very expensive.

Perhaps if this post inspires many people to experiment, and many do find great benefit, that will seem good enough (perhaps to someone with more engineering/manufacturing experience than me)!


Evan 2016-02-16

This is a very accurate representation of the commercially available SAD lights. I purchased a desktop lamp that uses a 27w PL cfl ~5000k, just as mentioned it was not bright enough! After years of buying light fixtures on craigslist, as a hobby, I realized I could use a couple of these for light therapy! I now have a cozy futon and 32 inch led tv in my basement with 4x 400w metal halide/HPS fixtures in my basement, along with a 1000w MH/HPS ballast. Depending on my mood I can switch between a lovely golden glow or a lively day like white light. Over the years I had almost forgotten about their benefits during the winter months, as it is normal for me to spend 2-5 hours under these lights each night. Thank you for this article!!

Face Mask for Sleeping

Jim 2016-02-18

I’ve read a lot about SAD and it’s affects on melatonin and serotonin.
There are many theories on why this condition exists. I certainly have it and have been searching for answers for years. The other night when it was a full moon and there was snow on the ground I could not sleep because it was too bright inside my bedroom. I have white blinds that don’t do a very good job of keeping out the light. Instead of replacing the blinds I bought a face mask at Walgreens for a few bucks. After a few days of wearing it I can say it has made a huge impact on my sleep. I no longer lay awake tossing and turning until I fall asleep. I fall asleep rather quickly. And I’m also getting up early. Earlier than usual and without a dawn simulator or seeing outside light as a stimulus. I live in the Midwest and the days are getting longer. But it’s still dark out around 5:20 pm (sunset). We’re not out of winter yet. So I can’t attribute my sleeping better to the time of year. It has to be the mask. One of the theories of SAD is that people with SAD are just more sensitive to the daily light cycles than others. They’re not necessarily genetically flawed, they are just more sensitive to light. It sounds reasonable at least. Maybe this could be because of their hunter/gatherer heritage, who knows. But I thought I should pass this along in case it helps somebody. Maybe we should be chasing darkness as much as we are chasing lumens.. Just a thought. Hope this helps!

Dark is good

David Chapman 2016-02-21

Yes, sleep experts agree light at night is one of the main reasons people have trouble sleeping.

I used a sleep mask every night for several years. Then I had a retinal tear, which I think might have been caused in part by the pressure of the “lip” of the mask on my eyeball (when lying on my side against a pillow). I don’t know whether that’s true, but I now try to avoid the mask, and sleep only on my back when I do wear it.

Most bedroom curtains do not keep light out effectively. However, it is possible to rig a solution that makes the room absolutely dark at night—and I’ve found that really helpful.

I recently used these blackout shades successfully. They take ten minutes to install (just cut to length with a razor knife and use the adhesive strip) and are totally opaque.

They aren’t a good long-term solution (they are flimsy), but for $30 they are a quick, inexpensive way to get immediate relief, and to find out whether total darkness is effective for oneself.

Reptile Bulbs

Jim 2016-03-09

Hi David. I have never felt any different sitting around a sad SAD light, no matter how bright it is. I always feel like it’s just missing something. I know a lot of people claim they think it works, but I almost wonder if it’s the placebo effect. While there is probably some truth to bright light being able to suppress melatonin, I am thinking more and more winter blues have to do with the lack of UV rays in the wintertime. All SAD lights don’t emit any UV. It’s FDA mandated they do not. They just emit visible spectrum light.
Visible spectrum light will wake you up, and set your clock. But I think that’s about it. Your body doesn’t synthesize Vitamin D from light in the visible spectrum. A lot of people take Vitamin D pills in the winter because of this. The problem with Vitamin D pills is they are not metabolized the same in the body as vitamin d created from the skin. There are also some 280 genetic receptor processes that are linked to Vitamin D produced by the skin. So taking a pill is not the same as getting it from the sun. 90% of our vitamin D should come from the sun. The rest comes from food.

Ever sit in a tanning bed? Feel different afterwards? Tanning beds emit mostly UVA, which tans. But, there is a little UVB as well. The reason a lot people either get addicted to tanning or just go a lot is because it releases endorphins, which feel good. But, it also produces Vitamin D. I don’t think commercial off the shelf SAD lights are the answer, no matter how bright they are. They don’t mimic the sun and are only a fraction of the spectrum. I think they help suppress melatonin, which will wake you up, but I really think you need the UV just as much as you need the brightness to combat the depressive and energy effects of winter.

I’m looking into Mercury Vapor Reptile Bulbs to tread my SAD. I also purchased a Vitamin D lamp (Sperti). We get UV rays from the sun in the summer, so why would we not want to get UV in the winter? Of course the FDA and suntan companies will warn UV radiation causes cancer, and that’s a whole separate conversation in itself, especially since the sunblock companies are trying to sell a product. But naturally, we lack UV in the winter, and that’s just fact. Many many people claim happy lights just don’t do anything for them, and I am one of those people. If you read the reviews of UVB reptile lights on Amazon, a lot of people are using them for winter blues and vitamin D production, as well as psoriasis and other skin conditions. I think we have to be careful about how much UV radiation we are exposed to on a daily basis, but I do believe we need the UV. And sitting under a Mercury Vapor Reptile bulb actually feels like the sun, and is sooooo different than sitting by a 10,000 lux light. It’s warm, you’re producing vitamin D, and it just feels as close to the sun than anything I’ve ever tried. I’m going to research these types of bulbs, as well as pick up a UV meter and maybe even a lux meter.
This one looks like a good one: 300R/E27/ULTRA VITALUX OSRAM SUN LAMP MEDICAL TANNING BULB 300 WATTS UV BULB 230 VOLTS (Amazon)

And I guess a good question would be, why do you think you need more lumens? Do you think it’s because something is missing? Perhaps the UV? I’m interested in your thoughts on this. Thanks David!

Placebo, and is UV the answer?

David Chapman 2016-03-09

Placebo effect is certainly possible! Subjectively, a bright enough lamp is effective for me, but n=1, and it’s easy to fool yourself.

Given that sunlight is about 100k lux, and SAD is caused by lack of sun, it seems like common sense that most commercial SAD lamps shouldn’t work; as used in practice they typically deliver about 3k lux, or less, which is a small difference compared to summer vs. winter. But, this certainly doesn’t necessarily imply that 100k lux is an adequate replacement, since the wavelengths are different.

As far as I can tell from limited reading of the literature, no one knows what the right UV dose is, or how much vitamin D in the body is optimum. Because the effects of vitamin D seem to be large, it would be really good for someone to do this research (carefully), but so far there doesn’t seem to be much.

My n=1 data point, for what it is worth, is that I live in a place (Tahoe) that gets a lot of sunny winter days. I go outside when I can, and I probably get an average of 30 minutes of full sun per day during winter. According to what I’ve read, that’s enough to produce enough vitamin D. Also, vitamin D persists in your body for weeks, and is released as needed, so day-to-day variations in sunlight shouldn’t make a significant difference. However, I find that 30 minutes in the sun is only mildly helpful with SAD. As soon as it get gloomy again, so do I, within minutes or hours. The short day length, and dark periods during the day, are more significant.

These observations make me think vitamin D is probably not the key. (But, they are subjective, and n=1, so I have low confidence in that.) I would guess that if UV is the answer, it is probably for some reason other than vitamin D. Sunlight may have other important effects besides vitamin D, which may be UV-mediated, or based on some other difference from artificial light that we haven’t identified.

Your experience with the mercury vapor bulb is really interesting. Again n=1 but it seems to be a powerful effect. I hope someone is motivated to do the necessary research properly soon!

UV Readings Over a Year

Jim 2016-03-09

“No one knows what the right UV dose is.”
I agree, but would tend to think one would want whatever average dose is present during the month(s) where you feel the best. Most likely the summer months.

I also agree that 30 minutes in the sun is only mildly helpful with SAD.
Once it gets gloomy, I start to feel it too.
Cloudy days seem to be the worst. The last time I was on a plane it was a grey, gloomy day in the winter. We hadn’t seen the sun for weeks living in the Midwest. Once we rose above the clouds it was a beautiful blue sky. It changed my mood rather quickly just seeing some sun. I can’t imagine that has anything to do with vitamin D. So, does SAD have more to do with color? Or could it be the possibility of being closer to a UV electromagnetic field?

More lumens = a greater electromagnetic field.

They also use electromagnetic devices (cranial stimulation) for depression with success.

It’s almost like those of us with SAD are not getting the energy we need.
And that’s what it feels like - we have low energy, and depressed moods.

The guy on this website took UV readings almost year round for his pet tortoise. Check out the UV readings.

I would say that directly correlates with mood and energy for SAD patients.

What about full spectrum 96 CRI Fluourescents?

Ryan McDermott 2016-08-25

Check this out:

Coupled with this:

at $219 your looking at just over 40k lumens, 5900k, 93+ CRI… I’ve been looking around quite a bit to find this solution. Double that and you pay $438 for 83,200 lumens .

You see anything wrong with this solution?

Overhead light with high-CRI fluorescents

David Chapman 2016-08-25

That’s intriguing!

I’m planning to start working soon on a high-powered overhead light, to supplement the desktop one I built last year and described in the post. Tentatively I was planning to use LED strip lights, but from experience I know those are a hassle. This looks like it might be better solution!

If you go with it, please report your results here!

It comes with bulbs

David Chapman 2016-08-25

Hmm, I see the Amazon fixture (Hydroplanet™ T5 4ft 8lamp Fluorescent) actually comes with 6500K bulbs. Although customer reviews say they aren’t great, so replacing them with ones like those you found might be good/necessary for the job. On the other hand, the fixture+bulbs is inexpensive enough that one might get it first to see if it’s basically the right thing.

This one is cheaper:

Ryan McDermott 2016-08-25

Doesn’t come with bulbs and is only $130

That other site also sells a 6 bay with bulbs for $175:

Looks like you need to buy accessories though if you don’t want to bother with wiring it in.

What drove me to look for such a bright solution was the fact that I was trying to find out if it would be possible to light an entire room up with enough lux for SAD. Based on my calculations the practical answer is “No”. The reason for this is I have a toddler. It’s easier to keep her in a room rather than 1-2 feet from a light. I’ve been using this calculator here:

Based on a 180 degree angle (which I think is right) I would need to buy 6 of those light setups to hit close to a measly 10k lux for the entire room. Not happening

Lighting an entire room. In blue?

David Chapman 2016-08-25

I was trying to find out if it would be possible to light an entire room up with enough lux for SAD. Based on my calculations the practical answer is “No”.

Yes, I came to that conclusion too. (A pity!) I’m thinking of using something like this as a supplement to the close-in lamp, not a replacement.

Random intriguing possibility: I’ve read a suggestion that diffuse blue light, plus a narrow white one, may fake your brain into registering “sky with sun,” and that this might be helpful for SAD. There was essentially no evidence for this, but it seems plausible, and I’m tempted to try it (using blue LED strips I guess).

Blu Blockers

Jim 2016-08-31

You can fill a room with enough light to mimic the sun. Hollywood does it all the time using special LED panels. It’s totally expensive, but the photography gear is out there. I don’t think that’s going to solve anyone’s SAD problems though.

One thing I’ve found helpful is yellow, Blu-Blocker type sunglasses. Seriously. They make a grey day look sunny. You don’t even have to buy the Blu-Blocker brand. Mine are fishing sunglasses. But the lenses are completely yellow. I can practically wear them at night because they seem to make everything yellow and lit up.

Psychologically, yellow is the happiest color. Coincidence? All I know is it helps. It really does. When it’s a grey, yucky day and I’m driving to work, I put on my yellow sunglasses and it’s like a sunny, summer day again. It really helps my outlook. I have no idea why except for the fact that yellow is what we’re used to seeing in the summer and missing in the winter. I wish I had yellow plastic that I could put over my windows at home because it makes a huge difference. It keeps the grey out and lets the yellow in. I don’t have the sunglasses with me now since I’m at work but I can post the brand later if anyone’s interested. I got them at a gas station for around $17. I don’t know if it’s just the color of the lenses, or if it has to do with UV blocking, or polarization. I just know that everything is prettier with the glasses on grey days.

I doubt it’s going to solve anyone’s SAD, but it does make a gloomy day look brighter, and may shed some light (no pun intended, well ok maybe) on what we need to beat the winter blues.

Good Bang for the Buck

chris 2016-09-22


First let me say thank you very much for all the info, describing your experience and for creating a forum where people can cooperate in finding solutions.

After reading everything posted here and researching on my own, I found this lamp (link below) that I am considering trying. However, there is not info about CRI quality. Should I be concern about that variable with it being LED technology? Are ~6000K LEDs inherently broad spectrum enough for SAD treatment purposes? Would you have any reservations about trying this lamp?

200W Outdoor LED Flood Lights 20000LM Daylight White 6000K for $130 at the time of this posting.

Hockey Arena Lighting

Jim 2016-09-23

Last winter I was severely depressed and had low energy levels due to SAD here in the midwest. One night I went to a local hockey game. I think the stadium recently replaced the metal halide bulbs because they seemed super bright. The game lasted a couple of hours but the next day I felt amazing. I felt like I got some sun. During the game I felt a lot better also. I felt like I was getting a charge almost. It was the wierdest thing. This year I would like to take a lux meter to a game to see how much light the stadium is emitting. Unfortunately you have to go through metal detectors in order to get in so I don’t know if I’ll be able to get in and get a reading. I know there are recommendations and some numbers out there for how much light hockey arenas should have but I want to know exactly what worked for me. It really made a big difference. It might be worth trying - go to a stadium and sit under the lights. Hope this helps!

LEDs, high-CRI fluorescents, metal halides, ...

David Chapman 2016-09-25

Interesting to receive several recent comments, just around the equinox! This is probably a good time to build/buy SAD lights, because it’s about when SAD first starts, and (in my experience) if one delays until December, one hasn’t got the energy to deal with it. Better to be proactive.

Sorry to be slow to follow up on these recent comments. I’ve just started thinking about this again myself, and plan to do a round of research soon. Lighting technology is improving dramatically every year—2015 was the first when high-output LEDs seemed practical to me. Prices will have dropped again, I expect, and there may be new products.

Based on very preliminary reading, the high-CRI fluorescents found by Ryan McDermott look very good. Unless I find some problem with them, or an unexpected better alternative, I’m probably going to try them as an overhead fill-in/supplement to the LED corn-bulb lamp I built last year.

Replying to chris—That’s 154 lumens/dollar, which is (as you note) an excellent bang/buck ratio! It looks good. If you get one, please report back here about how it worked for you!

The one thing I might wonder about is heat dissipation. It’s an outdoor fixture, meant to be attached to the side of a building; so it might get pretty hot. You’ll want to mount it in some way that makes it not a fire hazard. However, it’s only 200 watts, which should be manageable. Before I started working with LEDs, I used 1500 watts of halogens, which were pretty scary fire-wise, but survived.

Unfortunately, I don’t know how important CRI is, and so far as I know no one else does either, because the research hasn’t been done. Subjectively, I find standard fluorescents awful (and this seems to be the opinion of most people who don’t have SAD, as well). They produce greenish light that feels sickly. Incandescents and halogens produce yellowish light, which is much better, but feels like sunset (to me), which I find less than ideal. There may be individual differences in this—Jim reports that yellow light is better for him! There’s weak research evidence that blue light is particularly important for SAD, but I wouldn’t place a lot of confidence in that.

Subjectively, I find standard ~6000k LEDs taste like sunlight, close enough, although apparently their CRI is typically in the low 80s. That seems to be mostly due only to a sharp dip in the red wavelengths—the output is pretty even otherwise. I don’t notice the lack of red, and I suspect it isn’t relevant for SAD. (However, a complicating factor is that I have mild color-blindness.) There are high-CRI LEDs, but they’re quite a bit more expensive.

The corn bulbs I used last year are no longer available. They produced 90 lumens/dollar, and 10,000 lumens for 100W. 100W was the highest available.

This 150W corn bulb produces 17,200 lumens (or so it claims) for $170, as of today, or 100 lumens/dollar. Not a large price/performance improvement, but the higher wattage would make construction easier. I’m tempted to buy one or two to add to my existing hand-built fixture. CRI=80, which is not great, but I’m not sure how important that is.

Jim, your report about metal halide arena lighting is really intriguing. I wish someone would develop a metal halide fixture suitable for indoor use! I’m not confident enough of the safety issues to try it myself—at least not without a lot of research first.

When I’ve bought/built something new, I will follow up. Anyone else who does, I’d love to hear what you learn!

SUV LED light bar

David Chapman 2016-09-26

So after a few hours on Amazon, I discovered a new category of lighting I hadn’t seen before, which is “light bars” for off-road night driving in your SUV. These have the best lumens/dollar ratio of anything I’ve seen so far, by a long shot!

This one is 30,000 lumens, 6000k white, for $42 currently. That’s 714 lumens/dollar—less than 1/7th the cost of the corn bulbs I used last year! (CRI not specified, but probably in the low 80s like most current generic white LEDs.)

It seems like these should make an effective overhead lighting solution. I’ve ordered one to try; if I like it, I’ll get several!

You do need to feed them automotive DC voltage (10-30V). I ordered this power supply for $19.

I’ll report on how this goes once I’ve tried it out!

Agomelatine tablets (Valdoxan) & Mutated Melanopsin Gene

Jim 2016-09-27

David those automotive light bars look great. Excellent find. It looks like it would be a fair amount of light for the money. Definitely report back on your findings.

One theory behind SAD is that people with SAD symptoms have a mutated melanopsin gene and so their hormones and neurotransmitters get out of whack in the winter compared with those who don’t have a mutated melanopsin gene.

From what I’ve read (and I wish I could find my source - maybe?) people with the mutated gene make more melatonin in the winter and darker months which leads to delayed sleep phase syndrome and SAD.
In other words, their bodies are in hibernation mode due to too much melatonin circulating throughout the bloodstream.

I have an appointment with my pdoc on Thursday and I’m going to ask him for a prescription for Agomelatine.
My doc is pretty open-minded and knows my history with this so I’m hopeful he’ll let me try it.

The following was taken from Wikipedia regarding Agomelatine:

“Agomelatine is a melatonin receptor agonist and a 5-HT2C receptor antagonist.

Agomelatine resynchronises circadian rhythms in animal models of delayed sleep phase syndrome.
Agomelatine has shown an antidepressant-like effect in animal models of depression (learned helplessness test, despair test, chronic mild stress) as well as in models with circadian rhythm desynchronisation and in models related to stress and anxiety. In humans, agomelatine has positive phase shifting properties; it induces a phase advance of sleep, body temperature decline and melatonin onset.”

Now by all means I’m not trying to push drugs or anything like that. I just know, for me anyway, that I definitely need to be medicated during the winter - my SAD is that bad. I’ve tried Wellbutrin in the past, and it helps. Unfortunately it also gives me terrible acid reflux and makes me more angry/edgy. I know this forum isn’t necessarily for pharmaceutical drugs. I just want to help people since it can be such a debilitating condition.

I would most certainly rather do a light bar solution or something more natural but at least by me experimenting with melatonin supression I might be able to report back on how much melatonin plays a part in the SAD symptoms.


David Chapman 2016-09-27

Thanks, I had not heard of that before! It certainly sounds plausible and worth a try.

(Does anyone else have experience with this?)

Light bar news

Tino Mclaren 2016-10-05

Hi David

Did you try the light bar?

Im just about to create a SAD setup for this Autumn/Winter here in the UK and was wondering how you got on.

This thread has been most interesting to read. Many thanks.

The light bar

David Chapman 2016-10-05

Yes, I got one and it works and it’s bright. Supposedly it puts out the same number of lumens as the 3-bulb light I built last year. I’m not convinced it is quite as bright (but I don’t have a meter to check). It’s pretty darned bright, however!

I have ordered two more, for a total of 90,000 lumens supposedly. I’m planning to hang them above my workspace. It will be interesting to see just how bright that feels!

Light Bar Source

Tino Mclaren 2016-10-06

I couldn’t find any 300w versions on Amazon or Ebay UK sites but found one on AliExpress, its 400W and has a theoretical output of 40,000 Lumens.

Cant post a link as getting refused as spam but not hard to find.

Actually there’s quite a lot of other spec light bars there too, so there may be a better Lumen/Price ratio.

Ive ordered from AliExpress a few times so think I might give this a go, will report back if anyone’s interested.

light therapy, brightness

Rob MacLachlan 2016-10-09

I’ve covered my kitchen wall with 300 watts of florescent lights. Something I maybe didn’t pay enough attention to is that it is the brightness (lux) that matters more than the total light output (lumens). A bigger light just lets you achieve high brightness at a larger distance. So make sure you are close enough. Also, most smartphones have a somewhat calibrated lux sensor. On Android I have something called androsensor.

more light thoughts

Rob MacLachlan 2016-10-09

This is a page I wrote on my use of fluorescents, which I’ve been doing for over 20 years. [Spam filter didn’t like my link, but look for diy_light_box on]
I think leds are the way to go now, especially for a small box. The led fixtures have come down to where they are not all that much more than a florescent setup, and there is no issue with uv. I might worry a bit about excess bright spots harming your eyes if you have no diffuser and are looking directly at the chips, because they are very bright.
Recently a bought a square “troffer” led fixture (like used in suspended ceilings) and used the same trick of putting a cord and switch on it.

I have messed a lot with leds, though not so much for light therapy. You don’t want to overdrive fixed voltage strips because most of the power will go into the current limiting resistors, overheating things and degrading the leds. Led power dissipation is roughly linear with current, while resistors are I^2. If you get current driven led assemblies (offboard current source), then overdrive would work better, but still keep it cool. Heat reduces the output and shortens life.
I made a fish tank light by attaching led strips to aluminum plate (maybe .062 thick) using electronics grade thermal double adhesive tape. This has held up well. You can buy from digi-key, “thermal management products”.

I’ve used for diy led stuff, but fixtures are now something you can buy from lighting supply houses. The commercial fixture has two led strips rivitted onto the thin steel housing, so not much heat sink, but runs pretty cool.

The “corn bulb” design is really dumb for a leds because of thermal issues. Really, any led any”bulb” is problematic, but high power only makes it worse. Unless you need super high brightness for a projector or spotlight, it’s much better to spread things out, and to use a separate driver/ballast module.

yellow/Amber glasses

Rob MacLachlan 2016-10-09

The thing about yellow glasses making outdoor light better doesn’t jibe with my understanding, though whatever works…

If you have sleep disregulation, they can help you calm down at night because they block the blue that drives your circadian rhythm. We got some yellow safety glasses. I find it pleasant, and use them if I’m up in the middle of the night. Good if you’re staring at a screen. FWIW, we found about them at the psych hospital when my son went manic, so not a fringe thing.

uv blocking

Rob MacLachlan 2016-10-09

Ordinary window glass blocks uv significantly, though if you are staring at high intensity sources, I’d research it. Plastic diffusers do not block uv.

Lux sensor

David Chapman 2016-10-10

Rob, thanks for the many suggestions! Especially the phone light meter app one. I’m going to get that and follow up once I’ve had the time to do some measuring.

Follow up

Chris 2016-10-11

I just bought this industrial LED floodlight. Wow! 37,000 Lumens (claimed) for $125. Very bright, neutral light.

Not as cost effective as the lightbar, but a nice package. It does get hot. The cooling fins are hot to the touch, have to pull away after a few seconds. Bright, silent, neutral color. I got the cool white which is supposed to be between 5700-6500K. CRI >75. It’s BIG. Like countertop Microwave big. But half the depth of a microwave. So far so good.

BTW, I scoured the web for a definitive answer on whether CRI mattered for effectiveness of SAD mitigation, and the impression I got is that it hasn’t been specifically studied, but the CRI quality did not seem as important as the brightness/exposure time. And the CRI quality was more a matter of it being more pleasing light.


Ladrhyn Bexx 2016-10-18

Actually you dont need more light to produce more LUMENS and LUX , its about the quality of the lamp itself and what it is able to do. I work with a team at where we sell and work with a brainwave entrainment stroboscopic light machine for brainwave enhancement, using light impulses to change consciousness for better sleep etc. It works by strobing light with closed eyes....By changing the duty, (how long the light is on for). So its not how much light there is, its how much light can enter the pupil itself. So you get more pupil dilation. We have read the results on a LUX meter for SAD, we are getting around 10,000 - 25,000 Lumens, the highest 25,000 lumens for a 40hz programme at only 10% duty

Unethical advertising for a quack treatment?

David Chapman 2016-10-18

This sounds like made-up pseudoscientific advertising hype to me.

Your claims make no sense: lumens and lux are straightforward physical quantities, and you can’t get more of them by flashing the light.

Am I missing something? Have you got any controlled clinical research to show that a flashing light is effective for SAD? I couldn’t find any support for these claims on your web site.

Your web site promotes your product to “consciousness researchers, personal development trainers, holistic and metaphysical practitioners, spiritual retreats, float tank centres.” In other words, gullible marks for monist woo.

Unless I am missing something, it sounds to me like you are making a false advertising claim for a quack treatment. That is unethical. You should stop.


Jim 2016-10-18

Yes, I agree. Please stop promoting junk science and trying to sell a voodoo/quack product here. This forum is not for that purpose. It disgusts me when people try to take advantage of others in need of medical help for financial gain.


Jim 2016-10-18

Half of the links on that website don’t even work. There are no technical details nor are there any links to scientific studies. It’s pure quackery. So annoying..

New post with probably better solution

David Chapman 2016-11-22

I have a new post today on using LED “light bars” for SAD phototherapy. These are much less expensive than the corn bulbs I used in 2105, and much easier to deal with than LED strips. I think they will be a better solution for most people. (The post explains why.) They’re not practical for all purposes, though, so I’m leaving this post up!

LED Strip Lights on Glasses or Hat

Jim 2016-11-26

Hey David. Nice work with the Light Bars. They look pretty nice and functional. Since you posted this I’m beginning to feel the winter coming on and have been thinking about what to try this year. I’m already behind the curve as always it seems…

When I was having problems sleeping at night because there was too light in my room I decided I could either try and block out as much light from the outside by buying new curtains (expensive) or buy a $10 face mask to wear when I sleep. I chose the face mask. It turned out to be a great buy. Especially since I can take it with when I travel. I’m thinking about using this same approach to get more light during the day.

I’ve seen light visors and even glasses that are being sold to combat SAD. They look really silly and would certainly not be something I would wear at work. But I would wear them at home on the nights and weekends. I’m thinking about buying some cheap led strip lights and rigging them up to a small (maybe rechargeable) battery and fastening it to a hat or pair of eyeglasses (the kind without a prescription lens). This would be a lot cheaper and would allow me to go from room to room without losing exposure.

Have you or anyone else tried anything like this? My concerns are how bright the leds would have to be to simulate a 100,000 lux sunny day. And obviously how close/far they would have to be. And then there’s the safety issue of this in general. I don’t want to wreck my eyes by overexposure if it’s not safe. So any help or thoughts would be appreciated. Thanks everyone.

Light visor

David Chapman 2016-11-26

I haven’t tried this… if you do, I’d love to hear how it goes!

(Or if anyone else has experience with something like this.)

UV Light, retinal dopamine in SAD

Jim 2017-01-27

So I’ve tried regular light boxes and they don’t work. It just doesnt feel right. It feels like my eyes were missing something. I can tell immediately. I’ve always suspected the reason was because of the missing UV rays. My pdoc put me on Adderall recently, and I immediately felt like I did in the summer. I felt like I had just solved the SAD puzzle. I started reading up on Adderall. Apparently it targets primarily dopamine, along with some norepinephrine and serotonin. But since I’ve had Wellbutrin and Paxil before which do not target dopamine primarily, I could finally FEEL what I had been missing. Knowing this feeling, I started to wonder about the eyes, dopamine, and SAD. I searched and found a lot of PubMed articles that talked about retinal dopamine and SAD. I also stumbled across this crazy neurologist’s website: Jack Kruse dot com. The guy is either a genius or not from this planet. I could not even to begin to wrap my head around the stuff he talks about. It’s one of those sites where the dude seems to know what he’s talking about, but has a cult like following and some of the info is so out there that it is either beyond recognition or just not true. Or maybe a little of both. And then of course he sells things. So that makes me question all of it.. But as far as the info goes, wow. About all I can say is check it out and decide for yourself. I think UV light, tyrosine, and dopamine are all related and am researching it more. Trust me I’m definitely not affiliated in any way. I just want and need to find a cure for winter depression. After feeling what i believe was missing (dopamine) I started researching. So I would be interested if anyone else knew or ever felt like this. I think we do need more lumens, (UV lumens I think), which creates retinal dopamine and turns off melatonin? Here is the link to his facebook page which got me reading more:
Also checkout the Pubmed Articles related to retinal dopamine, SAD, and melatonin.
Thanks for listening everyone. And thanks for making/keeping this thread open David. Spring is around the corner!


Steve 2017-02-02

I am not sure if it’s all about wattage and brightness of the LED lights. I’ve used pretty bright LED lights (up to 50 watts) and I do not get the same replacement of the sun. It does not wake me up at all. However, if I open the blinds and feel the sun. Literally I can feel the warmth of the sun and the brightness and it seems to perk my mood up right away.

I’m thinking some of the heat I feel from the Sun makes a difference in my mood as well. Is this something that people are noticing as well? I wonder if LEDs even at 100 or 200 watt would get that same Sun feeling or maybe we should go in a different direction with the type of light bulbs we are using.

I think there is some validity to this as I read above about the guy at the hockey game and also the person promoting the reptile light. Has anyone tried different types and seeing if that makes a difference?


Ryan 2017-02-03

I believe that in order to get the full effect you need a more full spectrum solution. Some UV light is needed for other biological effects on the body. Someone mention Jack Kruse earlier, definitely something to look into but it’s difficult material for the novice to understand.

I abandoned LED and CFL lighting because they are just not close to natural and can have deleterious effects on your biology because your cells literally consume and use the light. When you get sunlight, it’s like eating a salad. When you get CFL or LED light it’s like eating a cake.

Dave Asprey lives in Canada and likes to use a tanning bed in the winter months to get the valuable UVB to allow his body to create its own vitamin D among other valuable biological processes.

For me personally, I just got one of those double halogen work lamps. It’s 20k lumens and when you have it close by you can feel the heat. The glass on them “protects” you from UVB. I have been tempted to remove the glass and get that UVB, but I don’t know what level of UVB comes out of those. Is 100x more than is natural or is it a small amount that would be safe? I’m not sure. You need to exercise caution when dealing with these artificial light sources as they can have profound effects on your health.

It was also shown in a rat study that some types of LED lighting can permanently damage your retina. The study was unreasonably and placed the light source very close to the rats so it was disregarded as this didn’t show you “practical use”. So when you guys are blaring 100k lumens in your face, well like the study, that’s not practical use either and so I would be very concerned about that retinal damage.

On the flip side, if you want to stick with your CFL or LED lighting anyway and miss that sun like warmth, buy that double halogen work lamp. Use them with each other. As far as color/brightness/feel goes it literally feels like you are outside ;)

Double halogen work lamps

David Chapman 2017-02-03

Those do definitely work for me, too! I used them before I started experimenting with LEDs.

I agree that, subjectively, a feeling of warmth helps with SAD. I find being physically cold and feeling depressed are closely connected. I keep the house warmer in winter than many people would and/or wear more clothes.

I got annoyed with halogen worklamps because they burned out very quickly. Running them all day, I had to replace the bulbs every week or two—not a big deal—but the electrodes also lasted only a couple months, and weren’t replaceable. Then I had to throw the whole unit away and buy a new one. I did get the cheapest ones available, and maybe more expensive ones are designed to last longer?

They run very hot, which does give a sun-like feeling of warmth. I found the fire hazard a bit worrying, and also I couldn’t get as close to them as I would have liked, for enough brightness.

I guess I would recommend getting at least two dual-500W worklights if you go this way—partly for more brightness, and partly because one of them is likely to burn out, and then you’ll have at least one while you are waiting to get a replacement.

For SAD light therapy, doesn

Me 2017-03-11

For SAD light therapy, doesn’t the light fixture need to output a very specific wavelength? ~430 nm? 405 nm? Does anyone know what is ideal or am I getting basically the same result buying a generic high power 5600k fluorescent or LED light as compared to all of these light fixtures that are branded for SAD light therapy?

Does anyone know how SAD light therapy actually works? Is it due to the light of a specific wave length hitting the retina or does it also have to hit one’s skin? Is full body exposure necessary?

How it works

David Chapman 2017-03-12

I’ll do my best to give answers to your questions, but I’m “some guy on the internet,” so don’t take them too seriously.

Wavelengths in the low 400nm range are blue light. There is some evidence that blue light is particularly relevant for SAD, but the evidence looks weak to me. “Cool white” LEDs include a lot of blue light (“warm white” lights do not). 5600k is “cool.”

Some commercial products have a blue plastic filter so they output mainly blue light. Evidence that this is helpful appears weak or non-existent to me.

As far as I can tell, you are better off with generic cool white LED lamps than SAD-branded ones, which are much more expensive and not meaningfully different. I have a more recent post about this you might like to read.

As far as I can tell, no one knows for sure how phototherapy works. Based on my reading of the literature, probably what matters is just getting enough daylight-like light in your eyes. But most of the experiments one would like to see have not been done, so there’s a lot of uncertainty about this.

There are SAD therapy visors that shine lights into your eyes from very close up. I haven’t tried these. I’d be interested to hear about experience with them from anyone who has.

Sad Light Fixture

Jim 2017-03-12

Welcome to the forum ‘Me’! I’ll tell you what I have found.
Basically there are two lights on the market right now for SAD:

  1. 10,000 LUX light
  2. ‘Blue’ led lights

The 10,000 LUX lights are adopted by the medical establishment.
Some people claim they work. Others, not so much. They do nothing for me and it appears the same for many on this forum. That’s why we’re here!

The ‘Blue’ led lights on the market are a new type of light due to recently discovered ‘ganglion’ cells in the eye that researchers found respond to certain ‘blue/violet’ wavelength(s). I’m not sure that the American medical establishment is fully behind these lights as far as clinical trials go, but nevertheless the private sector has a ton of em on the market already claiming they work. Like most things, try it and find out.

What is ideal? Honestly - the sun. The sun is natural. We can’t really replicate the sun 100% that I know of at least. We’re all here basically trying to figure out what works for us to keep our mood and energy up when it’s dark and grey outside.

Does anyone know how SAD light therapy actually works? It’s complicated but here’s it in a nutshell. Now this is just from me… like anything, do your homework/research and decide for yourself.

The eye receives light which is transmitted as electrical pulses to the brain to control circadian rythm, hormone secretion, physiology and behavior. That’s it. But it’s way more complicated than that.

There are tons of theories on why light therapy works. Most revolve around brain neurotransmitters like serotonin and the ‘SERT’ gene, melatonin, cortisol, etc. etc. etc. Most trained doctors believe light stimulates serotonin to be produced in the brain which makes you feel good. It also turns off melatonin which makes you sleepy/groggy.

As far as light hitting one’s skin goes - there is research that light on the back of the knee can stimulate the circadian rythm which is pretty interesting. Also light hitting the skin produces vitamin D, which most people lack in the winter. So there’s definitely a correlation there.

There’s a lot to it and honestly it’s a lot of trial and error to find what works best for you since we’re all different. I’d start with a 10,000 lux light and use it in the morning for a month or two and go from there. You’ll know if it’s working or not based on your mood, thinking abilities, and energy levels. Hope this helps.


Sarah Hoppes 2017-07-02

Love reading all of this. I kinda am wondering about the UV comments.... Regardless of cancer stuff…who trusts the FDA these days anyways?… We are lacking the UV in these products. Wondered if this would work… Lol… A good ol’ reptile lamp.

50,000 Lumen Light - From an LED Car Spotlight

Bryan B 2017-11-11

Found this on amazon. It is an LED Bar with 50,000 Lumens, with a temperature of 6000-6500 K (for a truck spotlight). I wanted to ask if you think it would work with this color and intensity? I plan to mount it near the ceiling over the headboard of my bed.

With the 500w, DC to AC transformer, it would only cost about $120.

Added benefit, when you turn it on inside your room, people driving by think you are having a close encounter of the third kind (supposed to illumine up to 900 feet). Also, if it doesn’t work for depression, I can always mount it on my Acura to “inform” folks when they forget to dim their lights. :-)

Really appreciate your advice!

52” 4D LED Offroad Light Bar 500W Curved, AUTOSAVER88 Night Driving Lamp Lights 50000 Lumens Automotive Spot & Flood Combo Beam Work Light

LETOUR DC 12V Power Supply 40A 500W AC 96V-240V Converter DC 12Volt 40Amp 500Watt Adapter LED Power Supply for LED Lighting,LED Strip,CCTV

50,000 Lumens, 6000-6500k

David Chapman 2017-11-11

6000-6500k is the color temperature of sunlight, so that is good!

And 50,000 lumens is quite a lot! It definitely sounds like a plausible solution.

Please let us know how it works for you!

An Interesting find

Shannon 2018-01-17

I have recently begun digging through the internet to find an effective sunlight lamp. I don’t have SAD but my room is very dark and I can tell a difference in how I work in the room as well as how difficult it is to wake. I happened across a company whose goal it is to make the most efficient lamp for a hobbyist, ie: “correct color temperature, correct light intensity, minimal glare and heat” as well as “increasing productivity by recreating the ‘daylight effect’ “.
As you mentioned, the SAD therapy lights rarely mentioned the numbers associated with the lamp. But in finding ‘hobbyist’ lamps, I am seeing almost all the information one would generally ask about. A lot of them are giving the lux at a specific distance and also giving the effective range from the lamp.
I don’t know if you have seen these sites but the lamp I’m looking at is 2,090 lux at 12 inches. Some of the bigger lamps are more powerful, still. I don’t know if this helps some of your readers who may need to find something in a pinch but I wanted to at least let you and your followers know what I found. The site I am purchasing from is called Daylight.

Softbox studio lightning

Toon Alfrink 2018-01-29

Who needs to simulate daylight? Photographers!


That’s 120 dollars for about 36k lumen. Not as cheap as the led bars, but a lot less trouble to set up.

Any updates? It's that time of year

Kirk 2018-09-07

I found this sad light box on Amazon for $35.
Claims to emit 5000 lux.

Thinking about putting it on a wifi outlet timer turning on when I wake up.

Side note found this about running & SAD.

Gonna try running, diet, enough sleep, & light box this winter.

Lumens vs lux

David Chapman 2018-09-08

The 5,000 lux claim is meaningless; how many lux you get depends on how far you are from the lamp. (See the “Lumens vs lux” section of the post.) What matters is lumens, and it doesn’t say how many it produces.

Based on size and price, I don’t think this one can produce much light (but there’s no info in the Amazon page that lets one know).

Some people apparently respond well to relatively weak lights, so it may do the trick for you.

Exercise, diet, and sleep definitely also help me!

What are you using this winter?

Kirk 2018-09-08

Curious as to what you plan to use this winter?

This winter

David Chapman 2018-09-08

It depends on where I am. The lamp described on this page is somewhat portable; I take it to different locations around the house, and sometimes on road trips.

I have another page describing fixed overhead lights that I use when working in that room.

LED Floodlights

Ben M 2018-10-17

LED floods!

David Chapman 2018-10-17

Those look very good! They weren’t available when I was last shopping. (New higher-wattage LED light types come out every year, which completely changes what I would recommend each time.)

This looks like it may be the best solution so far—they run on regular 110V AC electricity, so a separate power supply is not necessary.

The only caveat I’d suggest is taking care to mount them in a way that takes into account the fact that they’ll get really hot, to avoid fire risk.

LEDs run roughly 100 lumens/watt, so the 100k lumen light will produce 1000 watts of heat—nearly as much as an electric space heater, hair dryer, or toaster oven. (All those are typically 1500W.)

Flood light looks enticing, but...

Liz 2019-02-28

Hi, I’m from the Midwest and have severe winter lethargy, anhedonia, and brain fog as thick as a cloud, despite standard light box therapy, exercise, a fishy diet, getting outside as much as possible, a Luminette 2 light visor, a blue light that I made into a visor, a reptile bulb, and a vitamin D lamp. None of this stuff feels bright enough/effective enough. These surely help, don’t get me wrong.

I have been considering getting the 100,000 lumen LED flood light that Ben posted from Amazon. (I tried to create a link to the Amazon page, but the comment wouldn’t post, sorry.) It looks like it would be around 82,000 lux at two feet away. It’s “Getseason” brand.

However, I’m wary of artificial light being so intense. Maybe there’s a valid reason the typical light box is 10,000 lux at a couple feet away. Even small amounts of artificial light exposure can give me a headache or eye strain, but natural light ten times brighter than light box therapy is my anodyne.

Does anyone have any thoughts on Dr. Alexander Wunsch? I tried to create a link to the Mercola page about him. (I take Mercola with a grain of salt, but I have an open mind.) Wunsch thinks LEDs are dangerous to human health and recommends incandescent lighting over any other artificial light.

So, the thought of having 80,000 or so lux beaming down on me right now is very, very alluring. But I don’t want to damage my health in the process.

Any thoughts would be appreciated. I can always ask an ophthalmologist about this, but, “Hey, just calling to check if being a couple feet away from a stadium flood light every day for a couple hours would cause any health concerns!” doesn’t sound like I’d get a call back.


LEDs and health

David Chapman 2019-02-28

Sorry to hear about your SAD symptoms!

The Mercola Wunsch article is, in my opinion, mostly nonsense. Most of the many different theories in it about why LEDs might be bad for you are biologically or physically implausible.

That said, we can’t be absolutely sure there are no bad effects of LED lighting. Not enough research has been done to rule out the possibility. However, it’s unlikely that enough research will ever be done, on this or most other low-probability health risks, so one has to weigh likely benefits vs risks.

For me, the benefits of bright LEDs seem large, and the risks seem small, so it’s an easy choice, but everyone has to make the choice for themselves.

There’s one point in the article that is pretty well-supported by evidence, which is that blue light in the evening interferes with sleep. I use and recommend software that cuts blue wavelengths on computer and phone screens after 8pm.

The least implausible risk the Mercola article mentions is that blue light (during the day) might lead to macular degeneration. As far as I can tell, the evidence for this is slim to non-existent. However, if you are concerned about it, you can go with lower-color-temperature LEDs, which emit much less blue.

bright LEDs damaging retina

Pen Helm 2019-05-21

The French government has issued a report saying bright LEDs may damage your retina! … summary

ANSES statement on LED lights

David Chapman 2019-05-21

Here’s their official statement:

They say:

There is currently little data on exposure to lighting, irrespective of whether the systems are based on LED or other light sources. As a result, ANSES was able to present risk assessments with supportive numerical data only for cases of exposure to blue light.

There’s no specifics about this. They also talk about lighting systems that are 1000 times as bright as normal ones, and the fact that they are uncomfortable. It certainly seems that light that bright, or that is uncomfortable, may present a risk. It’s also possible that blue-only light may present a risk (although they don’t say why they think that).

They say that some white LEDs might also present a risk but that they have no way of determining that.

The brightest white LED system you might reasonably encounter will be less bright than sunlight, and no bluer, so I’m not persuaded that this is worth paying attention to. (Unless you are an installer of ultrabright outdoor lamps maybe.)

This is just my opinion, and I’m not qualified to give any advice about this. One could consult an opthamologist.

I’d be interested to read technical information about the risks of blue light if there is any. When I last did a literature search, there seemed to be “concern” but no experimental data that looked significant to me.

Possible health risks with blue light?

Chris Moore 2019-07-08

Hi David,
My understanding is that blue light (in the 400 - 490 nm wavelength range) is (or may be) damaging to the retina. Here’s one summary (from 8/2018) but others can be found:
Are you aware of this risk, & - if so - is it a concern? Is there a way to mitigate the risk (of retinal damage) using blue light therapy (for circadian rhythm problem, in my case)?
Chris Moore

Blue light

David Chapman 2019-07-08

I am not an expert on this. My lay reading of what I’ve seen of the evidence so far is that it does not warrant significant concern unless there are special factors involved. Further studies may find a basis for real concern (or may find that current precautionary warnings are groundless). But, c0nsult a professional instead if you want an informed opinion.

My amateur take-away so far: Don’t look into lights that are bright enough that it’s unpleasant to do so. In the evening after summer sundown time (i.e. ~8pm year-round) turn off the SAD light. Use “warm” lights in the evening, and use a program like f.lux to shift your screens to warmer (yellower/redder) colors. If you are approaching the age where macular degeneration becomes an issue, consult a professional.

HPS and MH - also looking into it

Mateusz 2020-02-03

This is a very useful article, thanks for writing it. I’ll be honest, I suffer SAD as well, recently purchased a 100 GBP Beurer TL100 and come on, it’s just an LED lamp which is bright, but not like I’d like it to be so I also decided to venture and try MH lighting, currently, I’m trying the GE CMH150/T/UVC/U/830/G12 with a warm 3000K colour temperature, love the light, it’s the kind of low sunsety one, perfect for me. I am also evaluating CMH150/T/UVC/U/942/G12 which is the 4200K lamp. Both are supposed to be within an enclosure but currently I run it in a corner without one, I hope the bulb wouldn’t explode brand new.

I’ve been researching the topic of UV on those blulbs and I am incredibly confused, so confused that I feel like I should reach out to GE themselves. As far as I understand, they come with the UV filter, this is the outer glass. If that cracks, the lamp with radiate UV like crazy, however inside an enclosure, at home, I’m sure you can take a quick look at it every day.

In the official specs document, the product data says “UV radiance: Exempt” according to IEC 60598-1 I think. It also talks about blue light hazard, but that can be avoided by not staring into the lamp.

I am also exploring the 250 and 400 W variants, all UV Exempt, the GE CMH250/T/U/830/E40 and CMH400/TT/UVC/U/830/E40. Those are for the 3000K colour temp.

For a cooler one, 5500K Osram HQI-BT 400 W/D PRO provides “Output of up to 400 W High efficiency Long lifetime E40 screw base for simple lamp handling UV values significantly below
the maximum permitted thresholds to IEC 61167 thanks to UV filter” and so does the HQI-T 250 W/D PRO.

Well, I bought a commercial floodlight from ebay for 20 GBP and I’d be planning to stick one of those bulbs in it with the matching ballast. The floodlight bears the IEC 60598-1 compliance mark. and a document I found titled “LE_Photobiological_Safety_Feb2013” mentions
As already stated in the introduction, the UV safety risk is already covered by the existing standard IEC 60598-1 under clause 4.24 and annex P.”

As for HPS, I read that the new bulbs emit almost no UV, but then I found a contradicting document saying they’re very dangerous. I’m going to assume that those dangerous bulbs are the non UV filter equipped ones, at least in MH, but to be honest, I haven’t seen any HPS UV enabled bulb and yet they’re used all over the place. Perhaps their radiation is negligible.

So, what do you guys think?

Do you have any suggestions as to who I could ask to get a reliable answer?

Want to try 100,000 lumen

Andy 2020-02-05

I’m considering installation of the 100,000 lumen Amazon light that Ben linked to in comments. I’m slightly concerned that it will be just too much light coming from a single source close to my head… It will be suspended from a drop ceiling (from wooden supports I will build) that is only 7’ 8’‘. I wonder if it will give enough spread vs. the “jeep light” setup that worked for you. I guess at 100,000 lumen brightness, it’s going to be bouncing tons of light off the walls in my small breakfast nook… hopefully enough to produce extremely bright ambient light, but not blinding glare!

Heat is also a potential issue. My electrician was satisfied by the fact that the Amazon listing mentions using the light for indoor projects. I can also set up a fan close to ceiling height, blowing across the room. If you were to do this again, would you go for one really bright 100,000 lumen light, or multiple lights with fewer lumen?

Does the warmth of the generated light matter in how it might address SAD symptoms? The options are warm (2,800-3,000k) and cool (6,000-6,500k). I’m leaning towards the cool, just based on the look of the fixture.

Thank you for sharing your invaluable experience!

100,000 lumens

David Chapman 2020-02-05

Well, 100,000 lumens is a lot! relative to sunlight, it ought not to be too much, but I have no experience with that much.

For most questions in this area, there’s no good evidence, so all we have to go on is intuition. My guess is that simulating sunlight as closely as possible is the best bet, in which case a single source is better. But this is purely a guess.

6,000 kelvin is closer to sunlight than 3000 is; I would be inclined to go with that.


Rowan 2020-10-13

Thanks for this, it was relieving when your website finally came up on google for me. I have narcolepsy and grew frustrated with the inefficiency of SAD lamps on the market. I wonder if the “something missing” feeling is just the lack of receiving all factors at once like we could from the sun…bright light that illuminates everything, diffuse blue light, UVA, UVB, warmth. It almost feels oddly dystopian to be trying to recreate the experience of being in the sun! I do get green light from the trees outside my window, but direct light (sunbeam) only a few times a year, and I need more. I’m in an apartment, so I don’t think my landlord would appreciate it if I hang up truck lights!

I try to respect my lower energy levels in the winter, after all, modern office life goes against billions of years of evolution. I wish there was a better solution, because my body seems to know when I am trying to trick it. I have bought clear umbrella, so on rainy days I can still go out and get maximum light. Tanning beds used to work great for me, the only thing that ever felt like “real” sunlight, and did wonders for my energy, pain levels, and mood. But I have stopped that due to health concerns.

Have you tried using mirrors to increase the area of the light source? Maybe not on the other side of the room, unless you wanted to increase overall light in the area, but angled out on either side of the light source. Paying attention to finishes and colors around the light is important as well in my opinion… ask anyone who’s ever gotten sunburned because the snow reflected too much light! I’m not sure how much light is technically really reflected from a mirror though, or if it’s exactly the same.

Prioritizing work spots right by the window has also helped significantly in the past, now that I’m working from home due to Corona, it’s been much better than office fluorescents. It’s almost like hearing the sounds outside, feeling the breeze, etc is all part of the experience my body is pissed at me for not getting. Apparently an hour long walk a day is not enough. It has improved my focus and energy significantly even on rainy days, if I can get some diffuse green light and breeze on one side.

For anyone interested… this is now a thing (for filthy rich people)
I’d love to cover my whole ceiling in these! It would only be a couple million dollars…

Is there a way to offer

Rowan 2020-10-13

Is there a way to offer myself up as a study subject for these sad light studies? I’d be a good one…! I’m left-handed though so they might not want me :(

I defeated SAD this year

Pen Helm 2020-10-13

I felt SAD coming on this year. I prayed until it went away. It worked. I had to be persistent, but I was determined to not let it win.

I found a cheaper, and easier alternative.

Michael Watts 2020-11-15

So I just spent about 2 hours looking up research papers and reading what you said. I don’t want to build a set up so I found an alternatives that gives enough light/wattage etc with a single item. It’s meant for garages and large spaces but it’s something you can simply install above your head. Please check it out and let me know (250W version) 5000k

How to power the lights in your build?

Faiz 2021-12-07

Hi there,
I’m very keen to replicate your build, given that corn lights have gotten cheaper. I’m wondering how you powered the lightbulbs. Do I need a special power supply for them? Or do I just directly connect the socket into the wall?


Corn bulbs: no power supply required

David Chapman 2021-12-07

Yes, most (probably all) corn bulbs run on regular wall power. (You could check specific models about this.)

Something to look out for is that the larger ones usually require a non-standard extra-large socket (“Mogul” type), so they won’t fit a regular lamp. But you can connect those sockets to a regular electrical cord and plug them into the wall.


David Wu 2021-12-12

you want the heat from the halogen? LEDs don’t have infrared, which you need to balance the spectrum and not just blast your retinas


Nikita 2021-12-13

Hi! You’ve made me curious and I did some research. Seems like it’s not clear if you need infrared light, which obviously has some biological effects, but you’d need a powerful halogen lamp to get a significant quantity of infrared.

Here’s an article discussing that:

Infrared and ignorance

David Chapman 2021-12-13

That was really interesting, thanks Nikita!

When I read the medical literature on lighting effects, my take-away conclusion is “nobody knows anything.” The evidence that bright light treats SAD seems pretty good, but beyond that everything is speculation based on too-small experiments that also don’t test what they were meant to. (Studies of the possible good or bad effects of blue light, which I’d particularly like to know about, are especially confused, as far as I can tell.)

This is true of medical science in general, unfortunately. Just a few days ago, the results of a massive, high-quality effort to replicate cancer biology findings were released. The upshot, simplifying slightly, was that most journal articles about cancer are not true.

This problem is something society has to get to grips with and force changes in.


Nikita 2021-12-13

Thanks! I completely agree with you, David.

I wrote to you about psychiatry in some other comments. Everyone who’d pay proper attention to available data would understand that antidepressants are doing more harm than good, but nobody wants to acknowledge that and 15% of people in some Western countries are taking them.

I love the subject of ignorance. There is a nice book of essays called “Agnotology: The Making and Unmaking of Ignorance”. They’ve invented a special term for the subject! The essay from the book called “M a n u f a c t u r e d Uncertainty: Contested Science and the Protection of the Public’s H e a l t h and Environment” deals specifically with medical research.

Here is also a nice article discussing the methods used to obtain results that you want:

Conduct a trial of your drug against a treatment known to be inferior.
Trial your drugs against too low a dose of a competitor drug.
Conduct a trial of your drug against too high a dose of a competitor drug (making your drug seem less toxic).
Conduct trials that are too small to show differences from competitor drugs.
Use multiple end points in the trial and select for publication those that give favourable results.
Do multicentre trials and select for publication results from centres that are favourable.
Conduct subgroup analyses and select for publication those that are favourable.
Present results that are most likely to impress – for example, reduction in relative rather than absolute risk.


Nikita 2021-12-13

Once I wrote to Robert Sapolsky about the claim in his book regarding the ratio of glial cells to neurons ( He said that he didn’t know that the ratio may be actually 1:1, not 10:1. The fact that we are still overestimating the number of glial cells 10 times shows just how little we still know about ourselves.

Best commercial lamp for 'amateurs'?

Chris 2022-04-03

Hi, I was wondering if you could suggest what is the best commercial lamp that doesn’t need any ‘set up’. I can’t build anything in my room, I know it would be more effective but I can’t because of my parents. What is the best alternative that I can just buy and put on use instantly? Thank you ^^

Marine aquarium Led for a SAD therapy light

Peter Fulko 2022-09-21

I hope this finds you well. I’m new to SAD and light therapy. I do however grow coral successfully. Many of these corals come from shallow water and bask in the midday sun. These lights replicate the sun. I set up one of the smaller “radion” lights on its side, dialed in 6000k and turned the uv channel down low and ran the light at 50% via the phone app. You dare not look at it as it’s like looking at the sun. I’m going to trial this for the foreseeable future. Happy to send a photo.
Thanks, Pete.

Can I concluce the Lux levels

Joel 2022-11-22

I am beginner with bright light therapy and I wanted to buy the Alaska Aurora lightpad mini ( ) bc it has 10.000 Lux at 25 inches. But now I found a construction lights which costs 10 times less and the technical datas are even better, that means a lot more watts, hz, lumens, also 6400K. I only have no infos about the lux at 25inches, but I thought it should have even more or at least also 10.000 lux at 25inches ?!
Thats the device ( Besides the 10.000 Lux I ask me if the low Unified Glare Rating (UGR) is bad for bright lighty therapy. Do you know more about the role of UGR in bright light therapy?
Thanks in advance!
Best wishes

The 3 x corn bulb project

Markus 2022-11-29

Im interested in building it but i would need a more in depth guide or atleast all the parts needed including cables etc.

Would be much appreciated.
Kind regards

Video Loghts

Carl 2022-12-03

Hi guys. Just bought this video light for €200:
Aputure Amaran 100d
LED. 36,000 lux at 1 meter. 5600K. CRI95+.
They have even more powerful lights.
Might be something to consider.

High intensity, high-CRI LEDs

David Chapman 2022-12-03

Thanks, that looks good!

It’s only in the last year or so that high intensity, high-CRI LEDs have become available. I’m looking forward to trying some out, to see if the CRI makes a difference for me!

trying the Carex Day-Light Lamp

Sue 2022-12-31

Model DL93011 from Amazon. I might screw around making indoor growing lamps for lettuce etc, but this is so easy to just try out first for my long standing middle of night awake for 3-4 hours insomnia.
Will let y’all know.

3 off the shelf 10000 Lumen boxes?

Sumo 2023-01-30

Great post. Would 3x off the shelf 10000 lumen SAD light units do the same job? Or even 6x 5000 Lumens and so on? I’m not a very DIY kinda guy and would love to have a more convenient solution. I’d be willing to build this if it was the only way. But buying multiple units off the shelf would be grand. I’ve already wasted this winter away with a fried brain. Its been the worst one for me yet and I suspect it will only get worse if I don’t do something.

Flood lights vs ufo bay lights vs photography lights

Tricia 2023-09-08

I have sad in seattle and have various sad lamps which really don’t do anything for me. I grew up in Las Vegas and I think my brain just wants sun and blue skies. But I read that sweden has started these ligjt therapy rooms where the goal is to make the entire room super bright like the sun and you sit there for an hour to get energy.

I’m curious if you think any of the above combos or alone would work in a small room? And which do you think would be most effective. I am not a diyer so I’m looking for plug and play in my apartment. The technology has changed so much since your original post I’m wondering if you have an updated recommendations. Thank you!

Flood lights vs ufo bay lights vs photography lights

Tricia 2023-09-08

I have sad in seattle and have various sad lamps which really don’t do anything for me. I grew up in Las Vegas and I think my brain just wants sun and blue skies. But I read that sweden has started these ligjt therapy rooms where the goal is to make the entire room super bright like the sun and you sit there for an hour to get energy.

I’m curious if you think any of the above combos or alone would work in a small room? And which do you think would be most effective. I am not a diyer so I’m looking for plug and play in my apartment. The technology has changed so much since your original post I’m wondering if you have an updated recommendations. Thank you!

An inexpensive, low-effort option

David Chapman 2023-09-08

I haven’t tried any of the ones you listed… they all seem plausible.

An inexpensive, low-effort option is a high-wattage “corn bulb” in a standard torchiere base. I have several of these, with 200W bulbs (around $100 each last time I bought one) in a $29 Home Depot base. You could start with one of those and see if it helps any. If it feels good, you might want several. Or you could try something else that has more watts in a single fixture.

Corn bulbs

Tricia 2023-09-09

Couldn’t figure out how to reply directly …
Something like this?

Limited-time deal: 2-Pack 200W LED Corn Light Bulb, E39 Mogul Base 5000K Daylight 28,000LM Lamp, Replacement 1000Watt CFL HPS Metal Halide Bulb, Large Area Lamp for Street, Garage, Warehouse High Bay Lighting

Corn bulbs

David Chapman 2023-09-09

Wow, apparently prices have come down a lot since I last looked (two years ago)!

Yes, those look plausible. The one thing to note is that “E39 Mogul Base” means the bit at the end which you stick into a socket is a larger size than is standard in home light fixtures (“E26”). You’ll need a “E26 to E39 adaptor” (like That doesn’t look great and the bulb may be slightly wobbly in the socket, but corn bulbs aren’t extremely good looking anyway :)

More on corn bulbs

David Chapman 2023-09-09

I see that those same ones you found are available in up to 400 watts! That’s new in the past couple years, too.

Price per watt is better for the higher-wattage ones. You might find 400W in a single source too much, though. The package of two 250W is plausible if you have room for two torchieres.

I’ve uploaded a photo of a torchiere, adapter, and corn bulb at

60.000 melanopsin (OPN4) lux lightdevice for SAD

Duncan Winters 2023-10-16

Hi David,

As a severe SAD sufferer i came across this light device:

“This is the highest fidelity white light you’ll find outside of a lab setting, with high color fidelity and extended reds so colors are rendered more naturally. Includes an adjustable 480nm sky blue channel for targeting maximum circadian regulation. [..] The Sky Portal is similar to a high powered SAD light, able to deliver well over 60,000 melanopic lux, with 480nm sky blue, blend-able with our extended red HiFi White.”

They also give some explaination about the cyan gap:

“What is a “Cyan Gap” & Why Do I Care?
Great question. In short, because of how white LEDs are built, there is a harsh gap in the 480nm sky blue range. This range is the exact tip of human melanopic sensitivity. This means that on top of it already being 100x darker inside than out, there’s also a massive gap in the light that responsible for the regulation of circadian rhythm, witch tells your brain that you’re awake.”

What do you think of it?

getchroma / melanopsin

David Chapman 2023-10-16

The short answer is that there’s some interesting science here; I don’t know how important/relevant it is; and there are less expensive alternatives that (as far as I can tell) should work similarly.

I corresponded a bunch with the guy who makes it when he was designing the rather different first version several years ago. There’s an interesting scientific hypothesis behind it. It’s based on biological research, and it’s plausible that it works the way he hopes, but I’m fairly skeptical that it’s cost-effective.

Now the long answer…

There’s two pieces to the science. One is the relatively recent discovery of melanopsin sensitivity in the eye and its likely role in SAD. There are light-sensing cells, separate from those we use to see with, whose only job is to tell the brain whether it’s day or night. The brain apparently relies heavily or exclusively on those, rather than vision, to regulate sleep.

Melanopsin is most sensitive to sky-blue light, which makes evolutionary sense. Therefore, our circadian rhythm, and presumably SAD, is most sensitive to sky-blue light, which has a roughly 480nm wavelength.

Standard white LED bulbs contain three types of LEDs, red, green, and blue. Their spectrum is very different from sunlight, which includes all wavelengths. We see both as “white,” however, because we have only red, green, and blue cone cells, and can’t tell the difference. If SAD depended only on the visual system, the difference couldn’t matter, but because of melanopsin, it may.

The blue LEDs in standard white bulbs mainly radiate royal blue, roughly 450nm, rather than 480nm sky blue. So the hypothesis is that bulbs that include LEDs that produce 480nm light would be much more effective at preventing SAD. They might not need to be nearly as bright overall as the massive amount of standard white LED light it seems we need, because maybe only the 480nm sky blue light matters.

So maybe for SAD purposes, all you need is a not-bright 480nm sky blue LED bulb. In fact, the first version of the getchroma product was that. This was more scientifically interesting than their current product. They don’t seem to offer it any more. Probably people found it too weird? It made everything look a sickly cyan. Or maybe it didn’t work very well? (It’s quite difficult to find a real 480nm LED, by the way. Most supposedly-480 ones actually are a 450 one plus a green one that come out looking sky blue together.)

OK, so why am I somewhat skeptical, and have I not tried this myself?

First, there’s an empirical fact that many people report very bright standard cool-white LEDs are effective against SAD. If 480nm were critical, and the bulbs didn’t produce that, we’d all have to be fooling ourselves. That’s possible, but I’d bet against it.

In fact, if you look at the spectrum sensitivity of melanopsin, it’s quite broad. It’s true that it peaks around 480nm, but it’s about 40% as sensitive at 450nm. It’s also about 50% as sensitive around 550nm: green light that’s produced profusely by “cool white” LEDs.

It’s also true that, although standard cool white LEDs have a maximum blue output around 450nm, and have a big dip in output around 480nm, they still produce some 480nm light, and quite a bit at adjacent wavelengths which melanopsin is sensitive to. You can see that in a spectrum here, for example.

(You can also see in that diagram that warm white LEDs produce nearly no light around 480nm, which is why I, and many others, switch to them in the evening. Subjectively, that seems to make a big difference.)

Together, the relatively broad sensitivity of melanopsin, together with the relatively broad output of standard LED bulbs, suggest you may only need around twice as much cool white LED light to get the same melanopsin effect as an ideal white. Standard cool white LEDs are now very inexpensive. “Very high CRI” LEDs, with a more even spectrum, are still quite expensive, unfortunately. The price difference is much more than two-fold.

I keep intending to try high-CRI cool white bulbs, which many people say they like better than standard LEDs, but the cost has put me off. The getchroma product is an example.

There are other very-high-CRI alternatives that are less expensive per brightness. Some are sold for video studio lighting; others for high-end retail: they make expensive clothes look extra good. I can’t recommend any product in particular because I haven’t tried them at all.

Thanks David

Duncan Winters 2023-10-16

Thanks for your comment. I will post my experience with the device in a couple of weeks. I also tried TUO lightbulb but that didn’t work for me.

TUO, and a relevant 2022 study I hadn't seen before

David Chapman 2023-10-16

For what it’s worth, my belief is that what mostly matters is getting a LOT of light. This belief has not been confirmed by rigorous scientific studies, but is consistent with a lot of anecdotal evidence, plus a priori reasoning: winter light doesn’t lack particular wavelengths, it’s just overall weak.

The TUO bulb is 250 lux, which is very weak. They have an unusual theory for why they think it works, plus one study supporting it. I am skeptical of the theory (and your N=1 experience is evidence against it).

The getchroma product doesn’t say how bright it is. However, because they suggest using it one foot from your face, I assume it is also very dim.

TUO’s page of science stuff links this 2022 article which is very interesting and relevant to the melanopsin-only theory that underlies the getchroma approach. They found that, contrary to that melanopsin-only theory, normal cones (regular color vision cells in the retina) do much of the work of circadian phase-setting (which is apparently what goes wrong in SAD).

The “S cones,” which are the blue light ones, are especially important, according to their results. The S cones are most sensitive at around 450nm, which is what standard LEDs put out the most of. That suggests that fancy, expensive LEDs may work no better for the same amount of light output.

If the getchroma device doesn’t work for you, I suggest getting some very bright cool white LEDs, and trying that as an alternative. You could consider doing that now, and doing a self-study test by using the two methods alternately.

In any case, please let us know what your experience is; it would be great to know whether weak high-CRI works as well or better than strong normal LEDs!


Duncan Winters 2023-10-16

Hi David,

The getchroma i ordered has 60.000 melanopsin lux. It uses 45watt energy. At this moment i am using a 62000 lux videolight from Smallrig, above my head. I need more light than those commercial SAD lightboxes give, they are toys when you have severe SAD. I also think i need more light surround me than heavy light on one place. I feel very sick 8 months a year due to those severe SAD.


David Chapman 2023-10-16

I’m sorry it’s so bad for you!

60,000 melanopsin lux

This may be misleading. It’s based on a theoretical calculation of how much effect it would have if the melanopsin-only theory is true. If the light is 45W, it won’t be very bright overall. However, it’s possible that its special 480nm LED will make a significant difference.

I need more light than those commercial SAD lightboxes give, they are toys when you have severe SAD.

Yes. They are borderline fraudulent IMO.

I also think i need more light surround me than heavy light on one place.

Yes, I have found that illuminating the entire room makes a big difference. That takes several hundred watts of standard cool white LEDs. Depending on the room, I’d suggest 500-1000W total, in several lamps spread around.


Duncan 2023-11-05

Hi David, i am using the get chroma skylight for three weeks, and for the first time in my life with SAD i am seeing some relief. My Energy is better and the jetlagfeeling is also 50% better. There is also a side effect. If i use the light too long on cct 50% i get severe insomnia and am awake the whole night. For now i use the light on cct 50% for 2,5 hours a day and will add 15 minutes a day. When using the light on cct 100% on 8am for 1 hour it keeps me awake at night. I have the same problem with the Blue light from Philips. I am a very sensitvive person, in the Summertime too much light also keep me awake. Will post some updates soon. I am also testing nir/Red light.

Cheapest reliable option for me

Kamil 2023-11-22

First of all I want to thank you for your inspirational work on research the science and the available lamps for SAD.
If I had 100 euros I’d buy a UFO led lamp, 200W, dimmable at 4000k (their only other options are 3000 or 6500k, and 6500k is just too white for me). But I only had 25 euros so I got a 100w led floodlight at 4000k. It works!
I use it every morning when I wake up during breakfast. It’s installed just above my table so the light hits the bottom part of my retina without me having the glare in my direct field of vision. The lamp itself is 60cm from my eyes so I can take my time and don’t have to worry that I’m getting too much light for too long (did that for a week and ended up having trouble sleeping). I sometimes also use it for 10-15mn at the end of the afternoon, when days are darker than usual with lots of rain.
I also have a 3000k model but indeed it doesn’t work as well as the 4000k.

It works!

David Chapman 2023-11-22

That’s great to hear, thank you!

Suggestion, for the sake of low-spoon readers

Herschel 2023-11-22

Thanks a lot for writing this, also thanks for the email about it–I had intended to implement this months ago and simply lost motivation.

I think this post could be improved with a clear and brief section at the top that says something to the effect of “these are the simple options: a. $n for one room and m hours of work, b. $k and j hours etc”.

Also, I assume you’re aware that Yudkowsky also came up with the same ideas and same intervention for his ex-spouse, he wrote a lengthy section about it in Inadequate Equilibria. I wonder how long it will take before someone calls you two “shills for Big Lumen”…

Low spoons, and history

David Chapman 2023-11-23

Yes—Before I started to really get a handle on this, there were many years in which I thought “this year I need to get a bright light solution worked out ahead of time, like in August, because by late October when I need it, I won’t be functional enough to figure out how to make it happen”! And sometimes I did manage to get the project under way in August or September, and had a solution when I needed it; but sometimes it got to November and I didn’t, and then I spent the winter not functioning. That was many years before there was a good solution, and the half-baked things I came up with might not last. Halogen lamps, in particular, usually burned out after a few weeks to months of heavy use.

Which is why the post says “The most important thing is to do something,” and suggests simple things that require little executive function.

I think this post could be improved with a clear and brief section at the top that says something to the effect of “these are the simple options: a. $n for one room and m hours of work, b. $k and j hours etc”.

I thought it has that? Maybe not quite close enough to the top? Near the top, it says:

The first half of the page, “Ways to get much more light,” begins with relatively inexpensive options that require no effort to set up. You can just order them off the internet and put them on your desk.

And it has a link to that. And then “This half of the page discusses four approaches,” and it lists them with costs.

I don’t think I can give hour estimates for anything other than the two “just buy this thing off Amazon” ones, because installation requires some improvisation.

I assume you’re aware that Yudkowsky also came up with the same ideas and same intervention for his ex-spouse

Yes, I haven’t actually read that, but I know he did. He wrote about it within a month or so of the time I published a first web page about it myself. I assume he used LEDs, so the timing isn’t coincidental; that was the year LED lighting first became practical. I know that, because I was experimenting with LED lighting for SAD the previous two years. Before corn bulbs came in, the available formats were light strips and COBs. I got some of each of those, but the reality at the time was making them work would take more mechanical engineering than I was up for.

Both our publications were preceded by a post from Luke Skaff in 2014, using metal halide bulbs. That’s a really interesting discussion, covering many of the same topics I’ve written about almost a decade later. As he noted, “LED technology is changing quickly and with time LEDs will likely overtake ceramic metal halide as the go-to choice.” It did only a couple years later. I hadn’t seen his post when I first wrote about SAD lights in 2015—but I did include a section on metal halide bulbs as a plausible alternative. They stopped being plausible by 2016.

But Skaff wasn’t first, by any means, either. There’s a history of this going back at least as far as the 1960s. People with SAD often figure out for themselves that light is what they need, and have been improvising solutions with whatever technology is available for at least that long.

When I was drafting the “LUX” post, I had a history section, which I deleted because I didn’t think many people would be interested. Maybe that was a mistake!

Loud 200 watt corn bulb alternatives

Chris 2023-11-23

So I bit the bullet and purchased a couple torchieres and a pair of 200 watt corn bulbs on amazon. They arrived and work exactly as i want them to. I don’t think I can go back at this point. However they are quite loud with their active cooling fan. At this point I’d be willing to pay a premium to get quieter light… what are your suggestions?

Silent corn bulbs

David Chapman 2023-11-23

Oh, great, I’m glad they work for you!

I didn’t know they still made noisy corn bulbs! The first ones I bought, in 2015, are noisy, but all the subsequent ones have been silent.

I’d recommend the 240W DragonLight ones, which are passively cooled (no fan), yet don’t get very hot (the aluminum housing acts as a heat sink, and the whole thing is flow-through for chimney effect). I’ve had three of them, and they’ve been reliable.

240W dragonlights

Chris 2023-11-23

Thanks for the recommendation! I initially didn’t want to spend $200 on the lights, so I opted for a cheaper pair that was $100 for a pair of 200 watt bulbs. Once I turned them on it was game over, though. I ordered a pair of those dragon lights, and I’m excited to see how they do! I’ll post again once I get them to give you my impression. Thanks for writing all this up, I’ve been pretty annoyed at the misleading marketing of SAD lights for a while.

In case you were curious, I am lighting a small office of around 120 square feet. I have my lights on torchieres on either end of my desk so they are just barely in the top corners of my peripheral vision. I’m not sure I’ll go beyond two, but we’ll see. I’m already going from 56,000 lumens to 74,800 lumens by upgrading to the 240 watt dragon lights.

Another thing I was curious about is if you have done any normalization of your desired “100,000 lumen for a small office” down to “lumens per square foot”. A cursory google search says the sun has about 10,000 lumens per square foot, which is insane - that would mean I would need 1,200,000 lumens in my office to emulate the sun. That’s 37 dragon lights in my 10’x12’ office, clocking in at 8,880 watts!! Compare that to what I’m going to be putting in my office which is 74,800 lumens, which is about 6% as much light as I would get if I just didn’t have a roof.

This is coming from having a single 4 watt bulb in my ceiling fan, combined with three 6 watt bulbs in a 3-bulb lamp post. I think I was getting around 2200 lumens before discovering high-watt LED bulbs.

All that to say that perhaps our desire to see indoor lighting increased by a factor of 10 (or more) is not actually crazy given the exponential range our eyes are built to see in, not to mention the neural circuitry inextricably linked to existing in spaces with lots of light.

And to continue rambling, if you are considering doing something similar to all the fine folks on this page, I would highly recommend considering your monitor’s peak brightness. I have an Aorus FI32U, which has a fairly high peak brightness, and I have to turn it all the way up just to be able to use it when these lights are on. You will want to make sure that you will still be able to actually work in the space after you light it up if this is for an office.

Up from 22 watts

David Chapman 2023-11-23

I have my lights on torchieres on either end of my desk so they are just barely in the top corners of my peripheral vision.

That sounds very good!

“100,000 lumen for a small office”

Actually, I said 100,000 for a medium-sized room. I think the two 240s might well be enough for me in a small office. Note also that we don’t (yet) have a good sense of how variable people’s light needs are; so what’s good for me might be overkill for you, or insufficient.

the sun has about 10,000 lumens per square foot, which is insane

Yes, this can’t be right. Setting aside the actual number, the units are wrong.

Your point about monitor brightness is a good one. That’s usually measured in “nits,” or candelas/m^2, which are the same thing. I looked yours up, and it’s 350 nits. Mine is 375. With all the lights turned on, it’s fully readable and not a problem, but it does look somewhat washed-out, and I’m wishing for a brighter one. It will be a criterion when I buy my next one. Generally it seems brighter ones are more expensive, but not always enormously so.

a tiny intuition to add to the 2nd footnote

Malcolm Ocean 2023-11-24

I would add to the footnote about Kelvin a remark about how you can think of it intuitively via noticing how it’s the same as how embers glow red but when you really get a fire going it becomes both hotter and puts off a much whiter light.

This post changed my life (and a few of my friends' lives, too)

vgm 2023-12-03

This post changed my life! One January, about six years ago, I simply couldn’t function in my west-facing room. I just felt nauseous. I ended up driving fifteen minutes every morning to sit on an east-facing park bench for an hour or two. (It occurs to me now that, before then, my habit for years had been to sit on an east-facing stairwell for an hour or two.)

Then I came across your site and bought a corn bulb and problem solved!

Last July, I was in a dark apartment and after a week, didn’t have enough executive function to figure out how to un-stuck myself.

Then I bought a UFO light and problem solved!

I’ve recommended this solution to a few people and their response has been very gratifying.

I’ve since found that, in addition to a sufficiently bright light, going for a walk first thing in the morning, even for a few minutes, really helps me. The later I wake, the more I viscerally crave being outdoors. Also, if I don’t go for a walk when the sun is low in the sky, I feel a little disregulated come sleep time.

Finally, I find my heart rate nadir is very consistently 2:00 - 2:30 AM. I wonder if the timing (mean and variance) could serve as a useful proxy for individual differences in sensitivity. Then again, the interventions here are so cheap that this might be more interesting than useful.

thanks for your post

Tricia 2023-12-03

What ufo light did you get?

UK/EU suppliers

Andrew 2023-12-06

If anybody has had any luck finding 100W rated lamp bases in the UK or the EU… corn lights I can find a plenty but bases to fit them into… drawing a total blank.

Any further thoughts on light diffusers?

Brian 2023-12-06

I tried out the basic 230W corn light setup in my small office and it’s great! I’m already feeling a difference; thanks for all the info here.

However, I do find the glare to be pretty intense. I have to place it in a spot on the floor where I can’t see it directly from my desk, and also have to be careful not to glance over when I get up. You mentioned the possibility of setting up a light diffuser in your post but didn’t go into much detail, and there didn’t seem to be much discussion in the comments either. Any thoughts on best practices?

I was thinking e.g. getting a light diffuser sheet like this one:

and lining the inside of a plastic or glass tube with, and placing the lamp inside the tube. I’d try to find a tube with a generous width and an open top to allow heat from the 230W lamp to escape.

I suppose my main concerns are (1) that it’s safe, (2) that it would be effective at reducing the glare, (3) that it wouldn’t be so effective that it would significantly reduce the benefits of the strong lighting. The above is my first stab at a way of trying to do that, but it’d be nice to have some informed opinions on whether that’s a good approach or not, or if there are better ways to do it.


David Chapman 2023-12-06

Hi, glad it’s working for you (except for the glare).

I don’t have any experience with diffusion materials. Maybe someone else reading this does, and can help?

I guess the questions would be (1) how much of the light does it block, vs diffuse? and (2) will it stand the heat?

FWIW, seems like wrapping it around the outside of the tube would be better. It says it’s heat-resistant, but not how much; the outside would be cooler.

If you can arrange the tube so there’s an air gap at the bottom, you’d get a nice chimney effect, and probably it wouldn’t get hot much at all.

All that said, my experience (and apparently that of many others) is that if you get the light up near the ceiling, it doesn’t seem glare-y. Is there a high shelf you could put it on? Or else maybe you could get a tall compact corner shelving unit and put it at the top of that.

(Also, yes, you definitely don’t want to look directly at it, even briefly. Not dangerous, unless maybe you stare for long periods, but quite unpleasant!)

Pendant light cord

i_am_lamp 2023-12-06

FYI, there’s also this pendant light cord version of the lamp base, which would be more convenient for most people:

Diffuser follow-up

Brian 2023-12-08

Thanks for the thoughts. I placed my 230W corn light on top of a tall bookcase in my office and that’s made things a lot better. There’s still a very bright spot off in the upper corner of my vision that I feel like I can’t look towards, but it’s much less of a nuisance than when the light is lower down. It’s kind of like having a mini-sun hanging low on the horizon. I felt a subtle dull ache in my eyes the first couple of days of using this light, but had it on again this morning with no discomfort.

I might still consider setting up a diffuser, but it’s less of a priority now. Some surface research on youtube suggests that cheap materials like a bed sheet or shower curtain would work just about as well as more professional grade materials at a substantial discount. If I do try this, I’ll just erect a rectangular sheet of the material across the top of the bookcase rather than my more complicated initial idea of wrapping diffuser material around a clear cylinder. That would be sufficient to block the light from direct view and there’s enough space and open air up there that heat ventilation shouldn’t be a problem.


Duncan Winters 2023-12-14

I’ve bought many Dragonlights bulbs but they are not providing what the seller is saying. The 100w versions are using 65watt and the 50watt versions are using 28watt. So that is 35% to over 40% less bright than advertised :(

Ufo light?

Tricia 2023-12-14

someone up above commented on UFO lights. my ceiling in my living room is quite low (8 feet, i was surprised to find out). I have no overhead lighting in the main part of the room. for the kitchen/office area i got two FEIT bulbs that are super bright. but it just serves to make the rest of the room very dark. Would a ufo light be too much? Or is there a better way. the main living area is about 15 x 12.

Bases for the 200W corn bulbs in UK

Lovkush 2024-03-10

First, thank you so much for the informative and practical guide.

These two bases are recommended in the original post:

However, they both have maximum rating of 60W. Has anybody found any bases suitable for the 200W bulbs?

Amazon changes links outside the US :(

David Chapman 2024-03-10

I’m really sorry, this is due to an Amazon misfeature. Those aren’t the bases recommended in the post! When you look at them from outside the US, if a product isn’t available in your country, it automatically substitutes something else it hopes is vaguely equivalent. And these aren’t, at all :(

I’m not sure what I can do about this (other than maybe putting a long wordy explanation into the post).

Update: Amazon link changing disabled

David Chapman 2024-03-11

I contacted Amazon about this problem. There was a bug involved, and it seems to be fixed now. It shouldn’t recur.

Sorry for the mix-up!

Reporting back on LED panel

Michael Lee 2024-05-20

Just wanted to report back. I bought the cheaper LED panel in the 800 W flavor that you linked, but mentioned that you hadn’t tried. I love it and have no complaints. Super bright lights, comes with a plug, no sound, low heat.

Might pick up a couple of corn bulbs to balance out the light in my room. Thank you for this write up!

Cheaper, super bright, comes with a plug, no sound, low heat

David Chapman 2024-05-20

Thank you very much! Glad to hear it!