[JT118]

Didcot Instruments Weather Station. I have acquired some of the bits for this station See attached, Would it be reasonable to fit an Op Amp to the DRS-5 to get the signal up so that Weatherduino could read it. My knowledge of electronics is a bit limited.

Didcot DRS-5 Calibration Certificate.pdf Size: 78,93 KB  Downloads: 297

Didcot Weather Station.pdf Size: 763,25 KB  Downloads: 301

Other bits I have acquired:
Anemometer DWR-205
Wind Vane DWD-205
Global Radiometer DRS-5
Net Radiometer DRN-301
Wet and dry PT100 thermometers DTS-5

DRS-5:

Silicon pV cell below polycarbonate diffuser, cosine corrected (see calibration Certificate) attached;

350-1100nm
14mV/kW/m2

   
   

Not sure how much of this can be made to work with Weatherduino.
The Wind vane and anemometer are easy the other bits less so; or maybe impossible.

[JT118]

DRN-301;

Global Net Radiometer; 300-80,000nm

1.4mV/kW/m2

   

The DRN 301 has a thermopile; facing ground (bare copper side)

   

and facing sun (black coated side).

   

Presumably long wavelength for the ground to sense radiation emitted by soil, and Short wavelength for the sun.

I think this used have a polythene "bubble" around either side of the thermopile which is kept inflated by a diaphragm pump

   

which pumps air through silica gel to dry it out to ensure bubble stays dry and clear.

[AllyCat]

Hi,

I don't know anything about that specific radiometer and the additional information that you've posted today, but:

The calibration certificate says that the solar sensor delivers 12.71 mV at 1000 watts/m2, which doesn't seem too far from the nominal specification of 14mV. I don't know what input level werk_ has used for the input calibration, but I see the "full scale" is (about) 1.1 volts and his maximum expected insolation is around 1400 Watts/m2. So I guess the required input level for 1000 W/m2 is about 700 mV. If you want to use the actual calibration of your sensor (assuming its calibration is still acurate after 15 years) then you will need to know (or be able to set) the exact value.

So basically you need to amplify the signal by about 700 / 12.71 , or around 55. If you are using a circuit as shown in post #52 here, then if the pot R3 is set to the top of its track (or better replaced by a 47k fixed resistor) then pot R4 needs to be set to about 850 ohms (takiing into account the 18 ohms source resistance). So a 1k or 2k2 pot might be better, or ideally an 820 ohm fixed resistor if you are able to use a calibration factor in the WeatherDuino software.

Of course the optimum calibration range will depend on your location (very different peak light levels between UK and Australia) and I don't know how accurate/stable is the "1.1 volt" reference in the Arduino.

Cheers, Alan.

[JT118]

Hi,

I don't know anything about that specific radiometer and the additional information that you've posted today, but:

The calibration certificate says that the solar sensor delivers 12.71 mV at 1000 watts/m2, which doesn't seem too far from the nominal specification of 14mV. I don't know what input level werk_ has used for the input calibration, but I see the "full scale" is (about) 1.1 volts and his maximum expected insolation is around 1400 Watts/m2. So I guess the required input level for 1000 W/m2 is about 700 mV. If you want to use the actual calibration of your sensor (assuming its calibration is still acurate after 15 years) then you will need to know (or be able to set) the exact value.

So basically you need to amplify the signal by about 700 / 12.71 , or around 55. If you are using a circuit as shown in post #52 here, then if the pot R3 is set to the top of its track (or better replaced by a 47k fixed resistor) then pot R4 needs to be set to about 850 ohms (takiing into account the 18 ohms source resistance). So a 1k or 2k2 pot might be better, or ideally an 820 ohm fixed resistor if you are able to use a calibration factor in the WeatherDuino software.

Of course the optimum calibration range will depend on your location (very different peak light levels between UK and Australia) and I don't know how accurate/stable is the "1.1 volt" reference in the Arduino.

Cheers, Alan.

Thank you very much; I shall have a go and see what happens, given that this is Ayrshire Scotland, and overcast is the norm we have generally low light levels. it will be interesting to see how close I can get it to the one on my Davis Groweather (www.ayrshireweather.org/groweather ), (56W at the moment and raining) though this sensor is also probably over 15 years old.

As far as I can see 500w/m2 is the reference point from which they work the calibration.

Rgds Jeremy

[werk_ag]

Of course the optimum calibration range will depend on your location (very different peak light levels between UK and Australia) and I don't know how accurate/stable is the "1.1 volt" reference in the Arduino.

According to the ATMEGA328 datasheet the accuracy of the reference voltage is 1.1 ±0.1 V, which seems great (around 10%), but some sources refer that this accuracy is from chip to chip, and that this voltage (within the 10% margin) is stable for each chip, and almost nothing dependent of the temperature.


The measured reference voltage of the tested Arduino Nano was 1.089 V at an ambient temperature of 21 °C and the package temperature was 29 °C. When the package of the ATMEGA328 is cooled down with freeze spray to -18 °C the reference voltage measures 1.084 V. The temperature drift is therefore 100 ppm/°C.

[JT118]

Showing the flexibility of Weatherduino. Graphs from yesterday showing the Didcot DRS-5 and Davis Solar readings from wunderground graphs. There is an offset between the two, which increases as the insolation increases from a few watts /m2 at the lower end of the range to nearly twice at the upper end.  I must check against another pyranometer to see if it is the 17 year old Davis which has been in use for nearly a decade, or the new old stock Didcot needing a slight tweek on the 2K potentiometer (or calibrate the software). The DRS-5 is attached to the solar / UV board using the resistors detailed by Allycat above. The Davis is working through a 17 year old Davis Groweather system. (IMAYBOLE2 is Weatherduino; IAYRSHIR8 is Davis). They are about 1m apart.

Sunny Scotland!!

Weatherduino:

   

https://www.wunderground.com/personal-we...210/mdaily


Davis Groweather:

   

https://www.wunderground.com/personal-we...210/mdaily


Thermopile Next! Might take me another year!

[werk_ag]

Thermopile Next! Might take me another year!

Nothing comes from a day to another, especially when it involves research.
Congratulations for your work, and thank you.

[AllyCat]

Hi,

The difference in the amplitude of the two "spikes" in the afternoon are noteable.  It's probably a "sampling" issue: generally Cumulus (and other software) just takes a "snapshot" reading at a particular instant after each logging interval.  That makes comparison very difficult for values that are changing from minute to minute, or even second to second.  Of course Cumulus can maintain a running average for wind speed, but not AFAIK for the Solar measurements.

For Solar calibration you really need a cloudless sky, not just a clear view of the sun, because light can be reflected off clouds and increase the apparent insolation.  So Calibration is a lot easier for those in Portugal or Australia than here in UK.    

700 W/m2 seems too high for your location at this time of the year, if the sensor is horizontal.  The Cumulus "(Theoretical) Solar Max Rad" curve (select a graph) should give a fair indication during most of the day (I have some doubts about the level near sunrise/sunset at our latitudes).

Cheers,  Alan.

[JT118]

Thanks, 700w/m2 seemed high to me, I shall get a pyranometer we have at work and see how they compare. Nothing is calibrated though they are just used for teaching but if I get similar reading from it and the Groweather then at least I will have an idea we are near the mark. I just need that rare combination of time and good weather. A rare combination up here. I have acquired an Apogee pyranometer from a police ebay auction and shall try that when I get it up from Somerset at Easter.

http://www.apogeeinstruments.co.uk/sp-11...ranometer/

[werk_ag]

...  Of course Cumulus can maintain a running average for wind speed, but not AFAIK for the Solar measurements.

Hi Alan

I think that Cumulus, regarding the Solar Radiation doesn't make any running average, usually that is done by the weather station itself. If I'm not wrong the the solar radiation on the Davis weather stations is a 10 min rolling average. The WeatherDuino Pro2 uses a 2.5 minutes rolling average, which turns easier the visualization of spikes.

Regards
Werk_AG