# Building a 'SAD' Light Therapy box

Status
Not open for further replies.

#### greenguitar28

Joined Dec 19, 2010
2
I am trying to design a Light Therapy Box for the treatment of Seasonal Affective Disorder. Some of the things I know about the ideal specs are as follows:

10,000 Lux, Full Spectrum Lighting, Low UV to prevent skin issues and such, As portable as possible, As inexpensive as possible without using cardboard, aluminum foil, ...etc.

I was trying to figure out if this project could be accomplished using super bright LEDs (because of their small size mostly) but I have not found any plans online that use this design. I have seen many commercial designs using blue or other colored lights but I am not sure if these actually meet the requirements or what... I honestly know very little about the ratings of lights, leds, and such and the information I have been able to find online doesn't seem to be clearing the fog very well.

I am hoping for some input on this project from some of you who may have more experience with this type of thing, or know about lux/lumens/color spectrums, etc...

Thanks!
Eric

#### SgtWookie

Joined Jul 17, 2007
22,221

#### greenguitar28

Joined Dec 19, 2010
2
Thank you for posting this. I see there is some good information in this article

#### Wendy

Joined Mar 24, 2008
22,575
If you decide to build one Wookie brings up good points. A simple light defuser would be needed for the LEDs (something as simple as paper or plastic) and I suspect some carpentry skills.

If you want to pursue this let us know.

Starter course on LEDs...

LEDs, 555s, Flashers, and Light Chasers

#### Markd77

Joined Sep 7, 2009
2,806
10,000 Lux is very bright, like direct sunlight. I think it will be hard to get that kind of level with LEDs. Even a 500W halogen light is only that bright at very close range. This might give you some ideas:
http://www.blue-room.org.uk/index.php?showtopic=36230

#### wayneh

Joined Sep 9, 2010
17,153
FWIW, a simple pane of glass will dramatically reduce any UV risk, if your source is creating UV. Most halogen bulbs are covered by glass for that reason.

I think it's also worth re-evaluating your original premise that an LED-based device would have any advantages. LEDs are fantastic for flashlights but, for general illumination, they're not much better than fluorescents in light-per-dollar-of-electricity economics. I'd be worried also that, while they produce light, they may be heavy at certain wavelengths and wimpy at others. Not knowing which wavelengths matter, I'd lean towards something with a broad spectrum and not all the energy concentrated in just a few wavelengths.

Last edited:

Joined Jul 7, 2009
1,583
Here's an engineering opinion, as I have a modicum of experience with somewhat related things. Note I said opinion and take the advice accordingly. I'm also giving numbers from memory, so don't take them as gospel -- take them as roughly order-of-magnitude estimates.

First, if you're to do a proper job of building this thing, you'll need a device to measure the optical output of your device. There's an expense for a light meter. Only with such a tool can you properly evaluate what you build and know if you're reaching the recommended design levels. The measurement of lux involves measuring the incoming radiation over a band and weighting the measurement with the photopic response of the human eye. Note that 10 klx is around the illumination in daylight outside, but not in the direct sun.

Now, as to LEDs: they are inherently narrow-band output devices. The typical LED will have a FWHM of 10 nm. That means, roughly, that if you measure the optical output of an LED as a function of wavelength, all the output will be roughly in a wavelength band of 40-50 nm, perhaps a bit less. A white LED works on the same principle as a fluorescent tube: a typically blue LED is used to cause a fluorescent material to fluoresce over a reasonably wide bandwidth. You'll still typically see the dominant excitation wavelength in the output, however. My guess is that, for the money, LEDs will be too underpowered.

That leaves incandescent and fluorescent sources (I rule out things like high pressure sodium or mercury bulbs). Incandescent generates too much heat. This results in picking fluorescent bulbs, which are the only designs I've seen for lights sold for SAD (but that was 10-20 years ago and no doubt there are people trying to capitalize on the market with LED sources , but my guess is they're woefully inadequate).

With a modicum of mechanical and electrical skill, one can build a box that contains a number of fluorescent tubes. This usually involves finding the starters, ballasts, and tombstones for the tubes, as well as the tubes themselves. Note the recommended tubes have roughly a daylight spectrum. To decide on the number of tubes needed, you'll have to measure the output of one tube and decide on a treatment distance, then estimate the number of tubes needed with a geometrical diagram, aided by your measurements of output as a function of radial distance from the center of the tube. Then you'll need to measure the attenuation of the diffuser material you choose to put between the user and the fluorescent tubes. Are you sure you want to go to all this trouble? If you put your time at worth, say, $20/hour, you'll probably spend much more on the whole project than a ready-made box will cost you (but I haven't priced them; I'm basing that statement on a box my sister had 15 years ago that cost her around$50).

You don't have much to worry about the UV exposure to fluorescent tubes, as their output is not terribly strong. If you put a piece of polycarbonate diffuser (acrylic is more common), you'll attenuate virtually all of the mercury 365 nm line. Glass works too, but is breakable. Part of the reason for the diffuser is diffusion as well as it protects the user from the shattering glass of a broken fluorescent tube (a rare event in normal use).

So, my conclusion is that yes, it's possible to build such a light source, if you're willing to do the engineering. Most people aren't. BTW, I can back up my statements with measured data from a spectroradiometer, as a number of years ago I investigated the output from UV fluorescent tubes.

The other aspect of such work is: are you going to sit for one to two hours every day in front of such a light? Personally, I don't have the patience to do such a thing. It also appears the mechanisms aren't terribly well understood. Hypotheses have been made coupling SAD and the light with vitamin D production (something we northern-latitude persons tend to miss out on during winter), which is something humans normally get from activity outside in the sunlight. Instead of building the box, you could consider taking a vitamin D supplement. In fact, my doctor once glibly stated I probably had a touch of SAD; since I refuse to take antidepressants, I supplement my diet with vitamin D pills during the winter. For me, the effect is equivocal, although I would probably say there's a very slight benefit. But it's a complicated subject.

#### Benny D

Joined Nov 5, 2015
2
I have recently built a portable light therapy device to help with my Seasonal Affective Disorder. One thing I learnt was that wavelength is important. Blue and greens (450nm - 550nm) are good for treating SAD, but just using white light without knowing how much blue/green is in there is a big no no. Using blue or green LED's with an output of about 300 lux (measured at the eye, not the light source!) is the way to go.

A couple of good references here:
https://www.flinders.edu.au/sabs/psychology/research/labs/sleep/bas.cfm
https://re-timer.com/blog/2015/04/17/green-red-white-or-blue-light-therapy/

#### ian field

Joined Oct 27, 2012
6,539
I am trying to design a Light Therapy Box for the treatment of Seasonal Affective Disorder. Some of the things I know about the ideal specs are as follows:

10,000 Lux, Full Spectrum Lighting, Low UV to prevent skin issues and such, As portable as possible, As inexpensive as possible without using cardboard, aluminum foil, ...etc.

I was trying to figure out if this project could be accomplished using super bright LEDs (because of their small size mostly) but I have not found any plans online that use this design. I have seen many commercial designs using blue or other colored lights but I am not sure if these actually meet the requirements or what... I honestly know very little about the ratings of lights, leds, and such and the information I have been able to find online doesn't seem to be clearing the fog very well.

I am hoping for some input on this project from some of you who may have more experience with this type of thing, or know about lux/lumens/color spectrums, etc...

Thanks!
Eric
Pretty sure Elektor magazines published a project a couple of years ago.

AFAICR: it was pretty much a panel of bog standard blue LEDs.

#### recklessrog

Joined May 23, 2013
985
There was one that I saw on sale in France. It contained 4 Phillips "North light" bulbs as used by artists etc, these have a specific spectrum that matches the light from the northern sky. The diffuser looked like a sheet of frosted glass.
I built a U.V box for exposing photo etch pcb's, I used a piece of the plastic that sits in front of the flourescent tube or led's in lcd tv's as the diffuser with only a piece of 4mm clear glass on top. This attenuated the U.V less than frosted glass. So use thicker frosted, (acid etched) glass if u want to block the U.V.

#### ian field

Joined Oct 27, 2012
6,539
There was one that I saw on sale in France. It contained 4 Phillips "North light" bulbs as used by artists etc, these have a specific spectrum that matches the light from the northern sky. The diffuser looked like a sheet of frosted glass.
I built a U.V box for exposing photo etch pcb's, I used a piece of the plastic that sits in front of the flourescent tube or led's in lcd tv's as the diffuser with only a piece of 4mm clear glass on top. This attenuated the U.V less than frosted glass. So use thicker frosted, (acid etched) glass if u want to block the U.V.
I'd shy away from any kind of discharge tubes - even sunbed tubes require filter goggle eye protection.

You can now get UV LEDs - and they come with an eye safety warning!

#### recklessrog

Joined May 23, 2013
985
The SAD light box's sold by Maplins and other sources use a special fluorescent tube that they sell separately as a spare.

#### ian field

Joined Oct 27, 2012
6,539
The SAD light box's sold by Maplins and other sources use a special fluorescent tube that they sell separately as a spare.
Quite often; things sold by Maplin are not fit for purpose.

#### recklessrog

Joined May 23, 2013
985
Quite often; things sold by Maplin are not fit for purpose.
Their sad lights seem to be the same as a lot of other suppliers, reviews are mixed, with about 50/50 as to whether they actually do any good. Probably do better to take a short winter holiday to somewhere sunny!

Status
Not open for further replies.