Solar Thermal Data Logging

Site tube flow meter, showing 1.6 gpm.

A temperature gauge well can also be used to place a temperature probe.

The ambient temperature probe hanging in the shade behind the collectors.

The only way to truly know how well your solar thermal system is working is to collect data. This can be as simple as writing down a few temperatures each day from thermometers installed in the collector loop, or as complex as monitoring every possible variable and logging them many times an hour. If you want to determine your collector and system efficiency, here’s the minimum amount of information you’ll need to gather:

• Collector loop flow rate (gpm)
• Fluid temperature entering collector (°F)
• Fluid temperature exiting collector (°F)
• Ambient temperature (°F)
• Solar irradiance (W/m2) 

Collector flow rate and fluid temperature in and out are the data needed to calculate the amount of power being delivered by the collector loop. Adding those values together over a period of time will give us the energy collected. Ambient temperature and solar irradiance, with the previous data, will provide enough information to determine the efficiency of the collector loop and to compare performance to SRCC test results. Plus, you’ll get a much better feel for how the system operates, and perhaps use the data to troubleshoot or improve the design.

Taking Measurements

Flow rate (gpm) in the loop you’re testing is often hard to estimate—it really needs to be measured. If your system uses an AC pump, the flow rate will be fairly constant—once you know it, it won’t need to be logged. With a PV-direct pump, this isn’t the case. In my system, the voltage to the PV pump is regulated by a linear current booster so that, in full-sun conditions, the pump runs about the same speed most of the day. I’m able to measure that flow with an in-line flow meter, but as a result of not being able to constantly log the flow from my PV-powered pump, my flow rate information is only applicable to sunny days. Flow meters for data loggers are available, but are more expensive than the other data collection sensors in the system.

Collector fluid temperatures are relatively easy to obtain. The data logger I used has standard temperature probes and you can choose to download the information in either degrees Fahrenheit or degrees Celsius. I placed the probe tips into thermal wells that usually house temperature gauges in the collector loop. Insulation stuffed into the well holds the probes in place and blocks the ambient temperature from influencing the measurement. The probe measuring ambient temperature should be positioned out of direct sunlight, but near the collectors.

The most complicated data to collect is solar irradiance, which requires using a pyranometer. Finding an affordable model for home use can be challenging, but I chose to use an Apogee PYR-P Class 2 pyranometer ($170), which uses a silicon PV cell to produce electrical current proportional to irradiance. Because the PV cell is not sensitive to all the wavelengths of sunlight, a calibration is required. The results are generally quite accurate (within 5%) as long as the pyranometers are exposed to natural, unobstructed daylight conditions and pointed skyward.

The data logger will only record voltage (in the 0.0 to 2.5 V range), which must be converted to irradiance. The output of the pyranometer is too small to be used directly, so it must be amplified. A handy amplifier is the AMP04EP integrated circuit chip (about $11) built by Analog Devices. It requires a single supply voltage of 5 to 15 VDC, which is a nice match for a standard 9 V battery. It consumes just over 0.7 mA of current, so a rechargeable nickel metal hydride (NiMH) cell (8.4 V, 180 mAh) will run it for about 10 days. Six AA rechargeable batteries in series would be enough to keep the amplifier running for several months, and most data loggers will run for much longer than that on their own internal batteries. The logic and calculations are applicable to a wide variety of data acquisition situations.

I used Onset Computer’s Hobo data logger to collect and store the four streams of data. Standard temperature probes were plugged into the first three ports, and the signal from the amplified irradiance meter is plugged into the fourth.

Number Crunching

For this round of testing, I chose a 5-minute time interval to sample and store all three temperatures and irradiance. This allows me to see the response to transient events such as system start-up and passing clouds. Downloading the data to a computer and exporting that information to a spreadsheet file is simple with the software supplied with the logger. Once in a spreadsheet, the data must be manipulated to calculate thermal power for both the sun (on the collector array) and the plumbing loop. To get the total solar power, multiply irradiance in watts per square meter (W/m2) times the surface area of the collectors (m2).

For the plumbing loop, multiply flow rate (gpm) by 60 (minutes per hour) by water density (8.33 pounds mass per gallon) by specific heat of water (1.0 Btu per pound mass per degree F) by the DT (°F) to get power in Btu per hour. Now divide that result by 3.412 to convert from Btu per hour to watts.

Then the power values can be multiplied by the time interval to get energy (WH), and added up over the day. On the temperature graphs, make sure the supply and return temperatures are very close before and after the test, so you’ll have confidence in the DT values. On the power graph, the system power should be substantially lower (at least 40%) than the solar power because of the inefficiency of the collectors and other losses. The system power curve should closely parallel solar power during a clear day. In both cases, toss out the transient portions of the data before using them to determine efficiency or energy collected.

Solar irradiance is measured with a pyranometer.	The data logger records three temperatures and irradiance.	A signal amplifier built on a bread-board adjusts the pyranometer signal for the Hobo.