PV Commissioning Tips and Best Practices: Page 4 of 5
Inside this Article
Measure voltage. PV system commissioning is not necessarily a one-size-fits-all endeavor. When commissioning a residential project with a multi-MPPT string inverter, you may be able to verify proper system operation—or at least rule out major issues— simply by scrolling through the inverter’s display screen. What are the Vmp values for each string? Do these values make sense based on the ambient conditions and the number of modules per source circuit? If inverter-output power also checks out, this quick-and-dirty performance analysis may be adequate as a system execution test within the context of LEED commissioning and verification activities.
A more common and reliable commissioning practice is to independently measure the Voc of each string. By isolating each string, you can tell whether a source circuit has the correct polarity and the proper number of modules in series. With consistent test conditions, a spreadsheet and an accurate multimeter, you can even identify module-level issues such as a failed bypass diode. When checking voltage in an inverter with multiple strings on a common bus, it is important to isolate each source circuit. In a larger inverter, you can usually open and close fuseholders under no-load conditions to take these measurements. Where two or three strings are paralleled without series fusing, as is often the case on multi-MPPT string inverters, you may need a tool to isolate individual source-circuit conductors.
When isolating source circuits in inverters or combiner boxes, it is important to use a dc clamp meter to check for current before opening any fuseholders or lifting any wires. If a voltage mismatch exists between strings paralleled at a common busbar or MPPT channel—perhaps due to an installation error or equipment issues—then the higher-voltage strings can dump current into the lower-voltage string even if the dc disconnect is turned off. While the 15 A or 20 A string fusing prevents module damage, opening a circuit with current flowing in it can pull an arc capable of damaging equipment or starting a fire.
Check for ground faults. To check for ground faults while taking Voc measurements, simply measure the voltage to ground from both the PV positive lead and the PV negative lead. Unless one of the poles of the array is intentionally connected to ground, the expected voltage reading is 0 Vdc, so it is easy to tell if a string has a ground fault.
The exception occurs where PV systems are deployed with rapid-shutdown boxes. This practice is most common in string inverter systems subject to the array-level rapid-shutdown requirements that first appeared in NEC 2014. Because the rapid-shutdown box will skew ground-fault measurements, you need to lift the strings from the box to accurately measure voltage to ground. Fortunately, it is seldom necessary to test individual strings for ground faults in this scenario, as the inverter itself is equipped with advanced ground-fault detection technology.
Perform IR thermography. Thermal scans performed with an IR camera are useful for identifying a variety of common issues within the array—including cracked cells, hot spots, defective diodes, failed modules and nonoperational strings—as well as high-resistance electrical connections and thermally stressed overcurrent-protection devices. One advantage of a walk-through thermal scan is that a secondary visual inspection happens by default. If an IR scan produces an odd or unexpected thermal signature, technicians can take a closer look to identify any obvious issues or causes.
When doing thermal scans, you must have the right equipment and know how to use it. PV inspections are most effective performed with an IR camera operating within a specific thermal sensitivity range. An IR camera’s thermal sensitivity is a function of its noise equivalent temperature difference (NETD) rating, expressed in milli-Kelvin (mK). While the IR camera manufacturer FLIR recommends an NETD rating of less than 80 mK for testing fielded PV systems, our experience is that a 100 mK camera works just fine. It is not necessary to have an IR camera with a high pixel count and high image resolution, but these features can speed up the scanning process by letting you capture images farther from the array.
Before adding an IR camera to your test kit, you need to get training and do some test runs. Since IR cameras pick up reflections in glass such as the front of a solar module, they are sensitive to false positives. In some cases, it is helpful to take an IR scan of the back of a module to correct for this. You also need to make sure you have set the camera’s temperature range properly based on the operating temperature of the component you are testing. There is a science behind interpreting IR images to determine acceptable variations between readings, identify problematic outliers or understand how environmental factors play a role. While an IR camera is a powerful tool for the commissioning toolkit, it is not one for beginners.