Solarimeters prototyping

What follows is a document of historical significance on our first Sunmeter solarmeter.

Of course we have made progress, yet we find it useful to keep this document available online, to show that the approach made to shape our tools is started from '"everyday learning" what tools and companies that make long day of dealing weather detection.
We know that our obsessive quest for accuracy has pointed out and improve aspects of operational weather center, but what we learned were much bigger.

Date: Sett.2009

In order to test the measure's reliability in the various weather conditions of our digital solarmeter SUNMETER, was asked the collaboration of the center of Teolo ARPAV weather.

The data obtained with our pyranometer were compared with a first class pyranometer ARPAV. The comparison process has improved gradually the structure and the software of SUNMETER improving the accuracy to obtain an output equal to that of the pyranometer within experimental uncertainties. Several months of tests and measures have provided excellent results for Sunmeter. It should be cleared now that the spectrum measured by the first class pyranometer is much broader than the spectrum measured by a silicon-cell solarmeter, however, the situation observed during the months of sampling led us to say that the only differences notables turned out to be only a few days during the first hours of dawn and sunset.

DATA PROCESSING AND UPGRADE

Positioning of SUNMETER solarimeter (in two different versions) and pyranometer reference at the weather center of ARPAV.

During the testing process has been used as a reference a first class pyranometer and two different versions of the digital solarmeter SUNMETER: the one at the bottom of the image has a support bracket with a closed structure around the silicon cell, while the one in the top has an open structure around the cell.

Our first measurements were obtained using two different models of our solarmeter.

Data were acquired with sampling intervals of a minute.

At this stage, was chosen the best support structure to be used for SUNMETER.

First results of the data acquisition process. The curves represent, respectively: the temperature (blue curve), the output data from SUNMETER, which was not made any correction, with the structure of engagement "closed" (orange curve) and the structure "open "(yellow curve), the data output from the pyranometer ARPAV (pink curve), the output of SUNMETER solarimeter as it would be under the same conditions of measurement, at the end of the testing process and improvement in the same atmospheric conditions.
In a first step we have used two different types of structure to fix the SUNMETER. One can observe how, in the afternoon, the yellow curve (SUNMETER with an "open"structure) was higher than orange (SUNMETER structure with a "closed" structure around the silicon cell). This effect is due to the fact that an open structure allows the silicon cell to detect a higher amount of light, capturing even the common components. By contrast, a structure "closed" blocks the optimum passage of light

You can see how the output of SUNMETER needed to be amplified. With the comparison of pink and green curves, we see that despite the amplification in presence of high values of temperature, SUNMETER shows greater output than the pyranometer, anyway the difference remained lower than the measurement uncertainty.

At this point it was clear that our solarmeter was underestimating the incident radiation compared to the measurements made with the ARPAV's pyranometer (pink curve). For this reason, the SUNMETER's signal was amplified

In a second step, we worked to identify an appropriate amplification factor of the output signal from SUNMETER to make it as close as possible to that of the reference pyranometer. This was done by comparing the curves of the digital ouput of the comparison solarmeter and pyranometer SUNMETER.
After identifying the correct amplification factor with which to fix the output of the solarmeter, our SUNMETER was appropriately reprogrammed.

Final results of the acquisition process. The green curve represents the output of SUNMETER as result of these fixes, the pink curve the output of the pyranometer ARPAV, the blue curve represent the temperature.

In these figures we see that the output of the SUNMETER almost superimposes the output of the pyranometer ARPAV.

Finally, it is interesting to note the differences in the shapes of the curves presented here, differences attributable to different weather conditions. It is also evident as our solarmeter has an output closer to that of the pyranometer when the values of solar radiation does not fluctuate very rapidly. However, even under these conditions, the values are good on an average of more than 5 minutes.