Welcome to the Cumulus Support forum.

Latest Cumulus MX V3 release 3.28.6 (build 3283) - 21 March 2024

Cumulus MX V4 beta test release 4.0.0 (build 4017) - 17 March 2024

Legacy Cumulus 1 release v1.9.4 (build 1099) - 28 November 2014 (a patch is available for 1.9.4 build 1099 that extends the date range of drop-down menus to 2030)

Download the Software (Cumulus MX / Cumulus 1 and other related items) from the Wiki

Homebrew Davis Solar sensor

For discussion of DIY weather equipment - sensors, accessories, improvements to existing kit etc
User avatar
mcrossley
Posts: 12685
Joined: Thu 07 Jan 2010 9:44 pm
Weather Station: Davis VP2/WLL
Operating System: Bullseye Lite rPi
Location: Wilmslow, Cheshire, UK
Contact:

Re: Homebrew Davis Solar sensor

Post by mcrossley »

A nearly perfectly clear day today, just contrails drifting around and dispersing.
The graph shows my sensor being shaded before ~07:00 and by trees after ~19:00
The green plot is UVI - the UVI scale on this graph is the same as watts/100.
perfect_day.png
You do not have the required permissions to view the files attached to this post.
vagyver
Posts: 13
Joined: Tue 25 Sep 2012 6:14 am
Weather Station: DIY
Operating System: XP
Location: Ελλάδα
Contact:

Re: Homebrew Davis Solar sensor

Post by vagyver »

I am also using the UVM-30 sensor as "Werk_AG" describes and my results seems accurate. I am also compare it with a ready made portable UV meter.
And here in Greece in summer the sun is really strong, so my tests have large dynamic scales.
So, i am really recommending it.
(The usual known circuit with a photodiode, an opamp and a resistor in MΩ category needs a lot of tests, opamp with special features, calculations of dark light current, offset etc.).


What i noticed is that if i place it vertically towards the sky and the sun in not at 12:00 then there is a small difference with placing with tilt directly towards the sun. So, i think it needs a cosine correction lens. Does anyone experience with this issue?

And what i would like also to ask is if aynone has placed outdoor. Will it get failed by rain water? Does aynone knows?

Thanks in advance!
guillemfreixanet
Posts: 1
Joined: Mon 19 Aug 2013 7:38 pm
Weather Station: Hand made
Operating System: Windows vista
Location: Barcelona

Re: Homebrew Davis Solar sensor

Post by guillemfreixanet »

Hello! I am also trying to build a self-made UVI device. I am a beginner in this world of UV sensors and I have started with the UVI-01 sensor from Reyax. I noticed you jave also done something with this sensor and I would like to ask you if you'va found with a similar problem than me. The problem is that when I connect the sensor as it is explained in the datasheet, the output voltage is always 0 even in the brightest days. However when I measure the voltage of the sensor without connecting the resistence, them I do measure tension. Do you know why? Did you find the same problem? And how did you solve it?

Thanks a lot for your help!
vagyver
Posts: 13
Joined: Tue 25 Sep 2012 6:14 am
Weather Station: DIY
Operating System: XP
Location: Ελλάδα
Contact:

Re: Homebrew Davis Solar sensor

Post by vagyver »

guillemfreixanet wrote:The problem is that when I connect the sensor as it is explained in the datasheet, the output voltage is always 0 even in the brightest days.
What is your connection? Do you use a resistor? Diode polarity?
vagyver
Posts: 13
Joined: Tue 25 Sep 2012 6:14 am
Weather Station: DIY
Operating System: XP
Location: Ελλάδα
Contact:

Re: Homebrew Davis Solar sensor

Post by vagyver »

vagyver wrote: What i noticed is that if i place it vertically towards the sky and the sun in not at 12:00 then there is a small difference with placing with tilt directly towards the sun. So, i think it needs a cosine correction lens. Does anyone experience with this issue?

And what i would like also to ask is if aynone has placed outdoor. Will it get failed by rain water? Does aynone knows?
I am coming back to this.
Does anyone know something about it?
AllyCat
Posts: 1124
Joined: Sat 26 Feb 2011 1:58 pm
Weather Station: Fine Offset 1080/1 & 3080
Operating System: Windows XP SP3
Location: SE London

Re: Homebrew Davis Solar sensor

Post by AllyCat »

vagyver wrote: What i noticed is that if i place it vertically towards the sky and the sun in not at 12:00 then there is a small difference with placing with tilt directly towards the sun.
Hi,

AFAIK that is what is supposed to happen. The sensor (and the general requirements for meteorology) should have a sine characteristic, i.e. the output is "100%" when the (sun) light strikes the surface of the sensor (or the Earth's surface) at 90 degrees, 50% at 30 degrees and 0% if parallel to the surface.

Cheers, Alan.
vagyver
Posts: 13
Joined: Tue 25 Sep 2012 6:14 am
Weather Station: DIY
Operating System: XP
Location: Ελλάδα
Contact:

Re: Homebrew Davis Solar sensor

Post by vagyver »

AllyCat wrote: The sensor (and the general requirements for meteorology) should have a sine characteristic, i.e. the output is "100%" when the (sun) light strikes the surface of the sensor (or the Earth's surface) at 90 degrees, 50% at 30 degrees and 0% if parallel to the surface.
Cheers, Alan.
Thank you Alan for your answer.
Theoretically, lets suppose that someone has the sensor in his hand and points it right towards the sun during a whole day (so the sun light strikes the surface of the sensor at 90 degrees during all day).

So, is that a (indirect) sine characteristic for the sensor?


If the answer is yes, then i just need (in my DIY UV sensor) a cosine correction lens to place it right above the sensor its like a Davis UV sensor?
AllyCat
Posts: 1124
Joined: Sat 26 Feb 2011 1:58 pm
Weather Station: Fine Offset 1080/1 & 3080
Operating System: Windows XP SP3
Location: SE London

Re: Homebrew Davis Solar sensor

Post by AllyCat »

vagyver wrote:.. points it right towards the sun during a whole day (so the sun light strikes the surface of the sensor at 90 degrees during all day). ..... So, is that a (indirect) sine characteristic for the sensor?
Hi,

I'm certainly not an expert, but no, I don't think so. AFAIK a fixed horizontal sensor will (theoretically) give a sine (elevation angle) response (cosine relative to the vertical axis). But a problem arises as soon as that sensor is mounted in any form of "protective enclosure". Then some of the light will be refracted and/or reflected at any "optical surfaces" (glass, plastic, etc.). So the cosine correction lens is basically correcting for its own presence. :?:

Don't forget that not all the "radiation" is coming directly from the sun but is also diffused from the sky. Also, that materials which are "transparent" (or not) to visible light do not necessarily behave the same way to UV and/or IR.

Cheers, Alan.
User avatar
Werk_AG
Posts: 198
Joined: Sun 13 Jan 2013 8:04 pm
Weather Station: WeatherDuino 4Pro
Operating System: Windows 7
Location: Cercal CDV - Portugal
Contact:

Re: Homebrew Davis Solar sensor

Post by Werk_AG »

vagyver wrote:I am also using the UVM-30 sensor as "Werk_AG" describes and my results seems accurate. I am also compare it with a ready made portable UV meter.
And here in Greece in summer the sun is really strong, so my tests have large dynamic scales.
So, i am really recommending it.
(The usual known circuit with a photodiode, an opamp and a resistor in MΩ category needs a lot of tests, opamp with special features, calculations of dark light current, offset etc.).


What i noticed is that if i place it vertically towards the sky and the sun in not at 12:00 then there is a small difference with placing with tilt directly towards the sun. So, i think it needs a cosine correction lens. Does anyone experience with this issue?

And what i would like also to ask is if aynone has placed outdoor. Will it get failed by rain water? Does aynone knows?

Thanks in advance!

I use mine horizontally, without any lens. It has a acrylic cover transparent to UV rays. On top is the Reyax UV-01, at botton is the UVM-30. Actually only use data from the UVM-30.

Image

Image

It's output is directly connected to an analog port of an Arduino. So, I read it use the Arduino ADC with a internal reference set to 1.1Volt.

To convert the output from Arduino ADC to UV Index values, I'm using this formula: IndiceUV2 = ((ADC4 * 22.5) / 1023.0)
I have found that with this values it gives me very consistent results when compared with other nearby weather stations, using Davis UV sensors.

You can se live data from my sensor at: http://www.meteocercal.info/recentGraphs.php
At the menu, chose Solar option. When you see the graphics click at UVI.
Or, if it's daytime: http://www.meteocercal.info/wxgraphs.php

Image
Last edited by Werk_AG on Sun 01 Sep 2013 9:01 pm, edited 1 time in total.
Image
The freedom to choose the weather instruments you want
vagyver
Posts: 13
Joined: Tue 25 Sep 2012 6:14 am
Weather Station: DIY
Operating System: XP
Location: Ελλάδα
Contact:

Re: Homebrew Davis Solar sensor

Post by vagyver »

I am very interested in your "acrylic cover transparent to UV rays" that you used.
What is that? It seems like a outdoor light. What is it exactly?
I am asking because right now i am looking for something to protect my UVM30A from weather but not make losses.

I am very happy with my UVM30A sensor's measurement which i have also placed horizontally like you (even thought there is some issues to improve more) (i can send photo).
With the opportunity, do you have in mind that its output is not exactly converter to UVI? E.g. 0.5V is not 5 UVI. 0.5V is something more that 4 UVI. There is a table about it. I can send you the table if you want.
In my DYI design i am using a microchip microcontroller.
vagyver
Posts: 13
Joined: Tue 25 Sep 2012 6:14 am
Weather Station: DIY
Operating System: XP
Location: Ελλάδα
Contact:

Re: Homebrew Davis Solar sensor

Post by vagyver »

Well,Werk i made some calculation about your UVI formula.

When there is 10 UVI then sensor gives about 1083 mV, lets say 1.1V (according to the table i said at above post).
According to your voltage reference, at 1.1V voltage the ADC will give 1023 bits. Now according to your UVI formula, a value of ADC4=1023 will give a UVI=22.5 which is not normal for earth ground. It is 2,25 times more than expected UVI=10.

So, my conclusion is: since you calibrated the UVI output with a known Davis station,your acrylic cover is not absolute transparent to UV but makes a absortion (loss) about 2,25 times.

By the way, are your UVI measurenents always parallel to Davis' measurents during the whole day? Or you have a agreement (the maximum value displayed) ONLY at 12:00 noon?
I mean, what is the comparison at 09:00-10:00am or at 16:00-17:00 which the sun is not right above the horizontally placed sensor as it is at 12:00?

I am really interested in this issue.
User avatar
Werk_AG
Posts: 198
Joined: Sun 13 Jan 2013 8:04 pm
Weather Station: WeatherDuino 4Pro
Operating System: Windows 7
Location: Cercal CDV - Portugal
Contact:

Re: Homebrew Davis Solar sensor

Post by Werk_AG »

Hi...
I am very interested in your "acrylic cover transparent to UV rays" that you used.
What is that? It seems like a outdoor light. What is it exactly?
Yes, it's the case from an outdoor light. This kind of outdoor light are used on the ground.
It comes with a glass cover... almost glass types except cristal glass absorbes UV, so I ask a friend to make for me
a similar cover but in acrylic material. I have read on many places that acrylic is almost tranparent to UV.

Now, related to the formula...

Your calculations are absolutely correct...I think I also know the table you talk. But what I have found is that the sensor, when at horizontal position never reaches 1.1V, even at 12:00 noon in a clear day. It reaches that voltage when it faces the soon directly.
This is the reason why I have used in the formula a value of 22.5 (after some trial error testing). After conversion, the maximum UVI I get is around 11 or 12.
By the way, are your UVI measurenents always parallel to Davis' measurents during the whole day? Or you have a agreement (the maximum value displayed) ONLY at 12:00 noon?
I mean, what is the comparison at 09:00-10:00am or at 16:00-17:00 which the sun is not right above the horizontally placed sensor as it is at 12:00?
I compared the values ​​obtained with those reported by Davis stations at various hours of the day. They do not differ much throughout the day.

In any case I'm not absolutely satisfied with the actual acrylic cover. Last month I'm geting a strange shape between 13:00 and 15:00 hours.
When I have the time I will try to make a new case, and let top of the sensor face the sky without any protection, and cover all the rest with liquid epoxy to protect the electronics from moisture. Let's see how many time if will survive.
Image
The freedom to choose the weather instruments you want
vagyver
Posts: 13
Joined: Tue 25 Sep 2012 6:14 am
Weather Station: DIY
Operating System: XP
Location: Ελλάδα
Contact:

Re: Homebrew Davis Solar sensor

Post by vagyver »

Werk_AG wrote: After conversion, the maximum UVI I get is around 11 or 12.
Are you sure that 11 or 12 is a correct measurent? It looks very large value to happen often.
I live in Greece where the sun is pretty strong (of course in Portugal is strong too) and the maximum i got is UVI=8.5-9 while the temperature was about 35 Celcious and that large UVI only a few days.

The site is not ready yet because i am still working on it (note that it is hosted in hardware and not in software) but take a look at http://www.kairos-pireas.gr. Don't count the 28,29 and 30 August since that days i had power problems.

Werk_AG wrote: When I have the time I will try to make a new case, and let top of the sensor face the sky without any protection, and cover all the rest with liquid epoxy to protect the electronics from moisture. Let's see how many time if will survive.
I had exactly the same idea. To let outside ONLY the face and place the WHOLE body inside a waterproof electrological box and seal it with epoxy glue.
But, i send a email to the company (roithner-laser.com) that makes that sensor (guva-s12sd) to ask if this would create any problem to the sensor. They answered me that this is NOT ok. That's why i looking for an other way to protect the sensor.
User avatar
Werk_AG
Posts: 198
Joined: Sun 13 Jan 2013 8:04 pm
Weather Station: WeatherDuino 4Pro
Operating System: Windows 7
Location: Cercal CDV - Portugal
Contact:

Re: Homebrew Davis Solar sensor

Post by Werk_AG »

Code: Select all

 Are you sure that 11 or 12 is a correct measurent? It looks very large value to happen often.
I did not say I got these values ​​very often. I said that this is the maximum that has been recorded in a few days in July 2013. In those days, even the official forecasts indicated maximum levels of 11.

Sometimes it is even possible to record peaks high enough in very bright days with high clouds. This is a phenomenon I've found described on the Internet, but I did not save the page where it was described.

This last week, here in Portugal, we had a very hot week, with clear sky, and, how to see in the charts on my page, recorded UV values ​​have varied only between 7.2 and 8.7 (today).
This is an oficial forecast for maximum UV for today (02-09-2013): http://www.meteocercal.info/wxuvforecast_pt.php
For my region the maximum forecast for today is 8. My sensor was registered a maximum of 8.8

Are you sure that in Greece, maximum values for UV are 8,5-9. Perhaps in Setember, but what about July and August? What say your official forecast services?
The site is not ready yet because i am still working on it (note that it is hosted in hardware and not in software) but take a look at http://www.kairos-pireas.gr. Don't count the 28,29 and 30 August since that days i had power problems.
Interesting your site ... I suppose for what you say that the web server runs on a microcontroller or a similar platform. Like to know more about it.
But, i send a email to the company (roithner-laser.com) that makes that sensor (guva-s12sd) to ask if this would create any problem to the sensor. They answered me that this is NOT ok. That's why i looking for an other way to protect the sensor.
Thanks for that info. I had no idea who manufactured the sensor.
I saw there several sensors that seemed interesting, and with quartz coverage...

So the idea of exposing the sensor directly seems not to be recommended ... if I have the time, before ending the period of long days of sunshine, I still try this solution, although not recommended.

Regards to Greece from Portugal
Image
The freedom to choose the weather instruments you want
vagyver
Posts: 13
Joined: Tue 25 Sep 2012 6:14 am
Weather Station: DIY
Operating System: XP
Location: Ελλάδα
Contact:

Re: Homebrew Davis Solar sensor

Post by vagyver »

Werk_AG wrote:
I did not say I got these values ​​very often. I said that this is the maximum that has been recorded in a few days in July 2013. In those days, even the official forecasts indicated maximum levels of 11.

Are you sure that in Greece, maximum values for UV are 8,5-9. Perhaps in Setember, but what about July and August? What say your official forecast services?
Sorry, i misurderstood what you said. Now you explained.

Yes, there are days with values almost 11.
And those days were before installing me sensor.
So, i didn't get such high values in my sensor because i placed it in the current position AFTER the days with such high values.


Werk_AG wrote:
Interesting your site ... I suppose for what you say that the web server runs on a microcontroller or a similar platform. Like to know more about it.
In my design there is a microchip microcontroller at the terrace with the outdoor sensors (SHT15 for temperature/humidity and the rest sensors) and a RFM22B (rf trasceiver module) sending the values from the outdoor sensors.
Inside house there are three other boards with 240x128 lcds (in living room, bedroom1, bedroom2) and RFM22B transceivers modules displaying all each values and outdoor values from terrace.
And one more board with a WEB intarface. It is a Tibbo EM1206 board connecting via simple serial port. The whole web page is hosting in EM1206 which takes values from the other board.
In your case, if you want to do the same, you just need a arduino web intarface with W5100 (about 10-15 euro), a domain name and a dynamic dns service (10 eyro a year).
I have also tried the arduino web intarface with my boards and works ok using SPI connection.


Finally, i am attached a graph of my UVM30A sensor. I am investigating its linearity and also the posibility for a cosine corrector.
UV_graph.jpg
You do not have the required permissions to view the files attached to this post.
Post Reply