WeatherDuino Forum

Hi,
I have had a fine offset station for 5 years now and thought I would upgrade. After looking at Davis and Ecowitt systems I decided to go down the weatherduino route
I ordered a TX pcb and Receiver plus a solar panel, which took just under a week to be delivered. I have ordered all the parts either from https://cpc.farnell.com/ or Ebay. Most of the parts have now arrived but I am still waiting for a couple of bits.

While I am waiting I decided to make an enclosure for the Reciever, made out of some 2"x1" wood and some 3mm thick pvc sheet. Hope you like what I have done,

[attachment=3018]
[attachment=3019]
[attachment=3021]
[attachment=3020]

As you can see I moved the temp, humidity, pressure sensor to the rear in it's own enclosure.

It's always a pleasure seeing WeatherDuino users creativity. The final result looks very nice..
Thanks for having shared the pictures.

Rgs
Werk_AG

To Continue.....

The rest of the components have now arrived, even the UV sensor from China only took a week

I have finished the Transmitter and tested it - all OK

I am now turning my attention to the sensors, for the outside Temp/Humidity I am using an SHT21. I have machined a holder for it out of some polypropylene sheet and covered it in an old fuel filter out of a petrol tank from a generator. This will go inside a home made radiation shield I made a few years ago.
[attachment=3022]
[attachment=3025]

For the UV/Solar sensors I have machined a housing out of polypropylene sheet, this is mounted on some 20mm conduit and will be attached to a post. I will cover the sensors with a piece of 3mm Perspex.
[attachment=3024]
[attachment=3023]

Thanks,

Hello,

Once again thanks for sharing pictures of your assembling, they can be very useful to others users, giving ideas...
Out of curiosity, do you have the specs or a link to a data sheet for the UV sensor you are using?
Thanks.

Rgs
Werk_AG

(06-03-2022, 21:00)Werk_AG Wrote: [ -> ]Hello,

Once again thanks for sharing pictures of your assembling, they can be very useful to others users, giving ideas...
Out of curiosity, do you have the specs or a link to a data sheet for the UV sensor you are using?
Thanks.

Rgs
Werk_AG

The item was from Ebay - https://www.ebay.co.uk/itm/223557457547

The only data is from the Ebay page, Product descripton:
Operating Voltage: DC 3-5V
Output voltage: DC 0-1V
Test accuracy: ±1UV INDEX
Current: 0.06mA(typ), 0.1mA(max).
Response wavelength: 200nm-370nm
Work temperature: -20℃~85℃
Detect UV wavelength:200-370nm
PCB size:28mm*12mm*10mm
Color;Black and Blue

Package included：
1PCS x UVM-30A UV Ultraviolet Detection Ray Sensor Module 3V 5V for Arduino

Thanks,

Hi,

I was concerned about the output voltage, which should be between 0 - 1.1V, so that module its ok.
Thanks

That case is fabulous!

Good job! Congratulations! 

Hi,
 

(06-03-2022, 19:41)SR123 Wrote: [ -> ]For the UV/Solar sensors ....... I will cover the sensors with a piece of 3mm Perspex.

Yes a beautiful construction, but the use of Perspex (aka Acrylic Sheet or Plexiglass) might not be an optimum choice.

Certainly those particular PV panels do need some protection because surprisingly they do not appear to be satisfactory for long term external use. However, an issue with any "transparent" panel is the reflection from both of its surfaces (about 10% in the graph below), which increases greatly at "glancing" angles (e.g. low elevation of the sun in winter). Also, the majority of Acrylic sheets block Ultra-Violet radiation, either by "accident" or by design, since they are often used to protect materials in Museums and Art Galleries against fading, etc.. Here's an example from This Web Link , and if I read another data sheet correctly, even a "UVI Transmissive" version (ACRYLITE OP-4) is only transparent to UVB in a thickness of less than 1 mm.

[attachment=3049]

Personally, I would use a thin sheet of PTFE (aka Teflon) sheet, which is known for its excellent optical diffusing (and reflection) characteristics, for all wavelengths between about 200nm (UVC) through UVB, UVA and Visible light to the Near Infra Reds beyond 1000nm. It also has excellent chemical and mechanical resistance properties and its diffusing characteristics can give the required "Cosine" optical response.  It actually reflects much of the light; Another Web Link indicates that a 1mm sheet will transmit (i.e. allow to pass through) only 10% of the light (and 2mm pass 5%), but this shouldn't be a problem: The PV panel can easily still deliver enough current and another interesting web link suggests that the "Guva-S12SD" UVI sensor (which I suspect might use the same chip) overloads in bright sunlight.

Not specifically relevant to this thread, but I'm very "suspicious" about the specification of the UVM-30A (and most others) as a "UV Index" sensor. I can't find any data sheet for the sensor itself (only for the complete PCB "module"), but the "Real UV Index" should strongly block wavelengths above 315 nm (i.e. UVA), whilst the sensor is said to cover UVA+UVB (200 - 370 nm). IMHO the "real" UV Index requires a specialist "Erythemal" Optical Filter (as used by Davis) or a true UVB sensor such as the VEML6075 (which has sadly got quite expensive now).

Cheers, Alan.

Alan,

Thanks for the interesting insight

I am still in the 'testing' phase of my weatherduino project so the sensors aren't in their final positions yet. The solar sensor is just leant against a wall at the moment! Yesterday was suprisingly sunny here in Lancashire and this is what was recorded,
 
[attachment=3050]

I am quite happy with that   The met office said a high of 1 for UV.
I will keep monitoring the results but seeing that the solar panel cost about £5 and the UV sensor was also about £5 I am not expecting super accuracy

Thanks Steve

(13-03-2022, 13:01)AllyCat Wrote: [ -> ]Hi,
 

(06-03-2022, 19:41)SR123 Wrote: [ -> ]For the UV/Solar sensors ....... I will cover the sensors with a piece of 3mm Perspex.

Yes a beautiful construction, but the use of Perspex (aka Acrylic Sheet or Plexiglass) might not be an optimum choice.

Certainly those particular PV panels do need some protection because surprisingly they do not appear to be satisfactory for long term external use. However, an issue with any "transparent" panel is the reflection from both of its surfaces (about 10% in the graph below), which increases greatly at "glancing" angles (e.g. low elevation of the sun in winter). Also, the majority of Acrylic sheets block Ultra-Violet radiation, either by "accident" or by design, since they are often used to protect materials in Museums and Art Galleries against fading, etc.. Here's an example from This Web Link , and if I read another data sheet correctly, even a "UVI Transmissive" version (ACRYLITE OP-4) is only transparent to UVB in a thickness of less than 1 mm.



Personally, I would use a thin sheet of PTFE (aka Teflon) sheet, which is known for its excellent optical diffusing (and reflection) characteristics, for all wavelengths between about 200nm (UVC) through UVB, UVA and Visible light to the Near Infra Reds beyond 1000nm. It also has excellent chemical and mechanical resistance properties and its diffusing characteristics can give the required "Cosine" optical response.  It actually reflects much of the light; Another Web Link indicates that a 1mm sheet will transmit (i.e. allow to pass through) only 10% of the light (and 2mm pass 5%), but this shouldn't be a problem: The PV panel can easily still deliver enough current and another interesting web link suggests that the "Guva-S12SD" UVI sensor (which I suspect might use the same chip) overloads in bright sunlight.

Not specifically relevant to this thread, but I'm very "suspicious" about the specification of the UVM-30A (and most others) as a "UV Index" sensor. I can't find any data sheet for the sensor itself (only for the complete PCB "module"), but the "Real UV Index" should strongly block wavelengths above 315 nm (i.e. UVA), whilst the sensor is said to cover UVA+UVB (200 - 370 nm). IMHO the "real" UV Index requires a specialist "Erythemal" Optical Filter (as used by Davis) or a true UVB sensor such as the VEML6075 (which has sadly got quite expensive now).

Cheers, Alan.