decent stand alone UV sensor?

[ocb]

Hi,
This is my first time here - so hello everyone!

I just got myself a non-UV WeatherEye (aka FineOffset), but would very much like to have UV readings.

I'm not going to spend the money to get the Davis kit, but I wondered if anyone has tried the "hobby boards" 1-Wire sensor?
http://www.hobby-boards.com/store/produ ... ector.html

Any advice on getting a decent - but reasonably priced - UV setup would be appreciated!

[AllyCat]

Hi,

Welcome to the forum. Do you really mean "UV" (Ultra Violet) light or just "Solar" data?. The link you've posted specifies: "It detects light in the visible to infrared portion of the spectrum".

Personally, I don't see the advantage of "low cost" UV measurements because the true "UV Index" is a very difficult measurement, hence the cost of around £500 for the Davis sensor. If you're "concerned" about skin damage then wear some sunscreen.

If you actually just want "Solar" data, then even that can be quite complicated. How do you want to use the data? You will find various threads on this forum concerning solar measurements: "Lux" data (e.g. from Fine Offset Stations) cannot easily be converted to "Is the Sun shining" (there is also a very long thread on that topic), nor even to an accurate value for Watts/square metre. Silicon photosensors generally respond to both Visible and Infra Red light (as suggested by the specification above).

The "final" issue is that there is no easy way to add solar data to a non-solar FO station, nor, I believe, to combine it into Cumulus. However a forum search should find a few references to Homebuilt One-Wire systems, particularly from Charlie and Gina.

Cheers, Alan.

[ocb]

Hi,

Thanks for the quick reply!

Yes, I saw the Gina posts, and was a bit daunted by the prospect of going down that route, or even if it was going to give the results I wanted.

Yes again, although I wanted both solar data - and UV. Solar, I'm happy with "basic" info, more for indication and general "completeness" of meteo. For UV, ideally UV-A,B and C as well as the overall UV Index. Last time I looked into this, I was surprised how few products there were out there. I didn't realise it was so hard to do...I looked at the logic of the UV index, and it's not *that* complicated. As for the sensors - I'd no idea they were that hard to make!

I have young kids, and even yesterday - which was the first nice day in the garden we've had for nearly 7 months - we all got a tiny bit reddened. Checking the hourlies on a local weather site, indeed the index had been predicted to be 4 during the afternoon.

I want to keep an informed eye on their exposure. My mum went through 14 years of skin cancer...but perversely it might have been that she didn't get *enough* sunlight.

So - it looks like I'd be cheaper sticking a low end tablet PC - with the local weather site on it - next to my weather station ...

[Werk_AG]

With an Arduino and this module (UVM-30A UV Sensor Module) I believe you can do a decent UV-B sensor. The problem is how to make the data avaiable to Cumulus.

UVM-30A UV Sensor Module

Operating voltage: DC 3-5V
Output voltage: DC 0-1V
Test accuracy: 1 UV INDEX
Operating Current: typ 0.06mA maximum value of 0.1mA
Response wavelength: 200nm-370nm
Job stability: -20 C~85 C
1. Designed specifically for the needs of high reliability and precision of measuring UV Index (UVI) occasion;
2. Suitable for measuring the total amount of sunlight UV intensity;
3. Control the grading standards of the World Health Organizaion UV Index;
4. Detection UV wavelength: 200-370nm;

[Charlie]

I've used the Hobby Boards UVI sensor and it seems O.K. - at least my readings more or less agree with the local weather office. I think you'll find he is no longer selling the 1-wire version; only the stand alone unit. His solar radiation sensor is junk, however. There's a wide variation from device to device, the sensor devices fail regularly, and there is no help mapping whatever output you get to the real world values. Save your money - I bought mine before doing the research, and his other sensors (temperature, humidity) seemed to work well.

I've been tempted to "roll my own", and I have a design outline, but I'm a bit low on motivation. My house is in a clearing not much larger than the house, essentially in the middle of the forest. There is no location where I get full sun all day, so until I resolve that problem, the sensors are purely an academic exercise.

[AllyCat]

With ... UVM-30A UV Sensor Module I believe you can do a decent UV-B sensor. The problem is how to make the data avaiable to Cumulus.

Hi,

Thanks, yes that's an "interesting" device which certainly offers the potential for far better accuracy than (say) a FO 308x station. However, I am a little concerned that it was already "discontinued" at the first supplier I found, and the second (DealExtreme) have rather a reputation for selling components that are "cheap for a reason" (perhaps like FO 3080 stations from Ebay sellers in Australia).

If you want to build a "UV sensor" there's no need even for the complexity of an Arduino, a simple PICAXE microcontroller could convert the Analogue signal to a suitable format such as One-Wire or serial "RS232", etc. But then what to do with the data?

However, I'm not convinced that that sensor truly measures the (skin damaging) "UV Index". The sensor chip itself is described as a "UV C, B and A" sensor with a cutoff at 360 nm whilst the correct "UV Index" cutoff is around 300 - 320 nm. I believe that the Davis sensor uses an additional optical filter and their (very informative) data sheet may help to (partly) explain why their sensor is so expensive (individual calibration, etc.). I'm not sure where I downloaded the data sheet but a Google seach for "6490_Spec_Rev_F.pdf" may be rewarding.

But I should admit that personally, I have doubts about the "usefulness" of the UV Index in general. There are many other variables such as skin sensitivity, reflection of UV from sea/snow and of course the (correct) use of high factor sunscreens, etc. that are likely to be far more significant.

Cheers, Alan.

[Werk_AG]

Thanks, yes that's an "interesting" device which certainly offers the potential for far better accuracy than (say) a FO 308x station. However, I am a little concerned that it was already "discontinued" at the first supplier I found, and the second (DealExtreme) have rather a reputation for selling components that are "cheap for a reason" (perhaps like FO 3080 stations from Ebay sellers in Australia).

The module can be found on several supliers, the best price I found is at DealExtreme (never had any problem with produts from this suplier) where I had ordered one unit (ordered last week). As soon as it arrives I will try to make some test, though unfortunately i don't have reference equipment to compare the readings. I'll try to use online data online from near official stations.

If you want to build a "UV sensor" there's no need even for the complexity of an Arduino, a simple PICAXE microcontroller could convert the Analogue signal to a suitable format such as One-Wire or serial "RS232", etc. But then what to do with the data?

You're right. I mentioned the Arduino, because I'm using one to make a system with temperature / humidity sensor (DHT22 in a Radiation Shield), a light sensor (BH1750FVI), and the UVM-30A UV sensor. The same Arduino further includes an RF transmitter. The data is then received by the same receiver that I use to capture the data from my Auriol 13726A station.
The system is already operating with temperature humidity sensors, hopefully soon add the rest.
My Arduino software (for Auriol stations) can be downloaded here (apologies for the page is only in Portuguese language).
As you said, the problem now is, how to use the extra UV and Light Sensor data. Now I use Cumulus that colect data via easyweather.dat file (generated by Arduino), but this file don't have fields for this kind of data. Hope Steve some time will implement some type of open protocol to Cumulus.

However, I'm not convinced that that sensor truly measures the (skin damaging) "UV Index". The sensor chip itself is described as a "UV C, B and A" sensor with a cutoff at 360 nm whilst the correct "UV Index" cutoff is around 300 - 320 nm. I believe that the Davis sensor uses an additional optical filter and their (very informative) data sheet may help to (partly) explain why their sensor is so expensive (individual calibration, etc.). I'm not sure where I downloaded the data sheet but a Google seach for "6490_Spec_Rev_F.pdf" may be rewarding.

But I should admit that personally, I have doubts about the "usefulness" of the UV Index in general. There are many other variables such as skin sensitivity, reflection of UV from sea/snow and of course the (correct) use of high factor sunscreens, etc. that are likely to be far more significant.

I agree completely that the UV index information is of very relative usefulness. At the moment, for me, all of this is for the pleasure of playing with electronics and learning some C++.

Best Regards, and thank you

[jrtsan]

Hi Werk_Ag,

I went to your website and clicked on contact us but could not find an email address or phone. Can you PM me with your email as I have some off topic questions about your arduino setup with the UVM-30A UV Detection Sensor Module. Thanks!
I love your site below and found it very helpful
http://www.meteocercal.info/wxaurioltools.php