[nmcclana] wrote out this very detailed instructible on building Mr. Burns, a sun burn alarm. Enter your skin type, sunscreen type, and UV levels for the day and Mr. Burns will let you know when it is time to go seek shelter or re-apply that sunscreen. Built on a Propeller platform, he’s using a blue LED as the UV sensor. He mentions that the device is fairly accurate, however people tend to put sunscreen on too thin and that will throw off the readings. There is a video of it in action on the instructible.
Now if we can only get this in a wrist watch format for portability.
Wrist watch would give poor readings due to unpredictable orientation. A hat might fare a little better? Nice idea, though.
It seems to me this is pretty simple (still quite brilliant though) and should be easy enough to make into a wristwatch – maybe using one of those TI MSP430s that the launchpad comes with?
@Derrick – get some nice tan lines, too!
Thanks, Hackaday, for putting up my project! I don’t have a pure UV source to test the sensor, but I tested it using a UV filter. Then I took readings different light sources. I also compared to reported UV readings at solar noon to my measurements & it came out very close to the reported level of UV radiation.
@Derrick – yeah, this is a straightforward build, mostly it was UI work & entering the WHO data. One trick, though, was how I read the LED without an op amp – I got the idea for this Mitsubishi whitepaper (http://www.merl.com/papers/docs/TR2003-35.pdf).
I’m cautious to dynamically re-calc exposure because I don’t think skin is equally sensitive throughout an exposure, but I couldn’t find any research on what the exposure curve is. It would be pretty cool, though, and not hard to program.
As mentioned the last time one of these “sun exposure” sensor projects came out, THE SENSORS DO NOT COVER ALL UV BANDS. There are THREE UV bands and some of them are very wide.
Nevermind that risks are based on melatonin levels, sunscreen protection, etc. None of which can be estimated by an electronic device. That’s why they don’t exist commercially- they’re useless!
This looks better:
Oregon Scientific EB612 Personal UV Monitor with Exposure Timer
it would be cool to make a PC frontend that took your location and gave you weather predictions and automatically updated the UV levels
An LED cannot be used as UV sensor, the acrylic is not UV transparent. Your just measuring (blue) light. Propeller is far overpowered, too. 1 UV-photo diode, an LED and 1 mikrocontroller plus a lithium battery can do pretty much the same. Not even need a resistor.
what about UV leds? if the acrylic isn’t uv transparent then why do they make things glow?
I’m rather disappointed that the first post wasn’t “Excellent”
otherwise very cool but a UVA + UVB sensor would probably work better i’m assuming?
If you use UV Leds, they won’t detect much. An LED is only sensitive to SHORTER wavelenghts. In result you can’t capture short UV with a standard long UV (blacklight) LED, as the epoxy blocks short UV. The only possibility there would be, to use one that has a UV transparent dome/plate, like some high power Nichia UV Leds. But then you might as well get a real sensor.
I have to agree with Steve that using the Propeller for this is overkill. A simple PIC would’ve done the trick.
Cool idea for the project, but I wouldn’t trust those readings coming from a single blue LED if I’m going to be frying under the summer sun ;)