Efficiently Locating Ground and Arc Faults in PV Arrays

Efficiently Locating Ground and Arc Faults in PV Arrays

A Technical White Paper Presented by Joel Robinson with Seaward Electronic

Executive Summary

Ground and arc faults remain two of the most common and persistent challenges in maintaining photovoltaic (PV) systems. These issues can lead to lost production, costly truck rolls, and in severe cases, thermal events or fires. This white paper outlines modern testing methodologies to efficiently locate both persistent and intermittent faults.  Through a recent case study with 8MSolar in Raleigh, NC, we saw first-hand how advanced diagnostic tools such as the Emazys Z300 and Seaward PV test equipment reduce time, cost, and risk while increasing system uptime.

Introduction

As solar projects mature and age, fault occurrence inevitably rises.  Despite modern inverters having inbuilt detection to help prevent the problem from escalating, the need to efficiently locate the cause of the issue remains. This paper explores fault behavior, standard troubleshooting methods, and modern solutions designed to simplify the diagnostic process and minimize downtime.

Understanding the Problem

To begin with, it’s important to be aware of the difference between arc and ground faults. A ground fault occurs when an unintended electrical connection forms between a current-carrying conductor (positive or negative) and ground. Arc faults are unintended discharges of current across a gap in a conductor or between conductors, often caused by loose/broken connections, damaged insulation, or mechanical/electrical stress.

In a solar array, DC arc faults are particularly hazardous because PV modules continue to generate current as long as they are illuminated, which can sustain the arc indefinitely. This sustained discharge can produce extremely high temperatures, leading to potential equipment damage and in worst case scenarios, even fire.

Persistent vs. Intermittent Faults

Both arc and ground faults can be persistent (continuously present) or intermittent (appearing and clearing spontaneously). Persistent faults are generally easier to identify and resolve, so we’ll focus on the more challenging intermittent faults. It’s important to understand that while these faults may disappear for some time, the underlying issue persists. Several factors can contribute to the sporadic nature of these fault types:

Loose, degrading, or mismatched connections: As conductors expand and contract with temperature changes (day/night cycles, irradiance fluctuations), gaps may open or close, causing arcs to start and stop.

Mechanical movement: Vibration (wind, thermal cycling, flexing of module leads) can temporarily make or break contact, re-igniting arcs.

Environmental factors: Moisture, dust, or corrosion can intermittently bridge or widen a gap, altering arc stability.

Current/irradiance variability: Since the PV array’s current output fluctuates with sunlight, the fault may only sustain an arc under certain irradiance/load conditions.

Best Practices to Locate an Intermittent Ground Fault

Experienced technicians are likely familiar with the practice of using a digital multimeter to measure voltages to ground on + and -, looking for differences and then calculating a rough location based on the delta. We’ll call this the DMM Method. While effective, it requires a “hard” ground fault to work. For persistent and egregious faults this methodology is sufficient.

When attempting to locate intermittent ground faults the use of more advanced test equipment and methodology is crucial to avoiding lost time and multiple visits. Investing in the right equipment will pay dividends through time savings and increase revenues both for the PV system owner and the party responsible for remediation.

Ground faults are directly associated with a low Insulation Resistance (Riso) value. Various organizations and standards reference 1 megaohm as the minimum passing threshold for Riso values, however a new installation will often see Riso values of 50 megaohms and higher.

It’s important to understand that Riso values are subjective. The Riso values collected when commissioning a new installation are an important reference benchmark for future comparison. These values will vary based on site conditions and ultimately will degrade over time. Thermal cycling and the presence of moisture are two factors that can impact the results of an insulation resistance test.

Running the Insulation Resistance Tests

The first step will be to collect Riso values for each of the strings connected to the faulting portion of the array (inverter, combiner, etc.) looking for the outlier(s). Which string(s) have noticeably lower Riso values?  These are the string(s) we will initially look to isolate to allow for additional troubleshooting.

While Riso testing can be gathered using a standalone Insulation Resistance tester, it can be beneficial to use a PV specific device like the Seaward PV150, PV210, or PV:1525. Doing so allows the technician to connect both the positive and negative sides of the string, gathering Riso values for the entire string (+ & -) with only one test.

Once a suspected string has been identified and isolated, further division of the string using a jumper and repetitive insulation resistance testing can help the technician narrow down the specific fault.

It may be helpful to (carefully) introduce moisture around the suspected string/module. Using a handheld or backpack sprayer to mist the modules and wiring of the suspected string before running additional insulation resistance tests can help to reduce Riso values and perhaps even recreate the previously detected ground fault.

It’s worth considering that some O&M companies have begun strategically conducting insulation resistance testing immediately following a module cleaning service, effectively collecting worst-case scenario Riso values.

Advanced Equipment = Time Savings

While standard equipment like digital multimeters and insulation resistance testers can be helpful and will always have their place, better options have come along. The Emazys Z300 is one such example.

Billed as a PV Troubleshooter, the Z300 provides extremely accurate Riso values, but also identifies the specific location of the ground fault in scenarios where the Riso value is below 3 megaohms, avoiding the additional time required to further divide up suspected problematic strings.

Best Practices to Locate an Intermittent Arc Fault

Because arc faults are located within the positive or negative conductors and not present on ground, these types of faults are often even more challenging to locate. That’s where we can employ a similar methodology with a different type of test and result - Series Resistance (Rs).

Frequently, arc faults are associated with elevated Rs values. In an inverse fashion to Riso values, Rs measurements for a new installation should be quite low, measured in ohms rather than megaohms. Just as Riso values decrease with system age, we can expect for Rs to increase over time.

Specific Rs values will vary based on several additional factors including string and wire lengths, module types, inclusion of RSDs and more. It’s important to note that Rs values are impacted by irradiance. Low irradiance can increase Rs while high irradiance will decrease it. For this reason, it’s important to compare Rs values from various strings with similar irradiance conditions. Strings of similar composition and age should have similar Rs values when tested under similar irradiance conditions.

Let’s imagine that we have a similar scenario to the above - one inverter has ceased production for a short time due to a detected arc fault. The technician will approach the diagnostic process with the same technique, testing each of the strings and looking for the outlier, however in this case we’re evaluating for an elevated Rs value.

Once a suspected string has been identified, we can then create a jumper to break the string in half, testing each half for the presence of this elevated Rs value and through process of elimination, identify the specific DC component causing the spike in Rs and likely the associated arc fault.

Case Study: 8MSolar Rooftop System

To translate this to the real world, I’ll illustrate specific steps taken during a recent site visit for a system owned and operated by 8MSolar out of Raleigh, NC.  This system was commissioned in Q1 of 2025 and is a rooftop installation that utilizes an RSD solution by APSystems along with SMA Core One inverters.

Inverter 2 recently ceased production with error logs showing a detected arc fault on String Group (SG) 2. String 5 was tested twice to confirm that it was the outlier. See data collected while on site:

Test data highlights elevated Rs values on String 5. To confirm identification of this issue the technician swapped SG2 with SG1 on Inverter 2. As a result, the arc fault error moved to SG1. Disconnecting the String 5 from SG1 then cleared the error allowing for the inverter to resume production and confirming the conclusions drawn from the test data.

Remaining steps include utilization of the Z300 and a jumper to divide the faulty string and further isolate the specific cause of the arc fault and associated elevated Rs value.

Conclusion

As PV assets age, efficient troubleshooting becomes essential for maintaining ROI.  Employing Riso and Rs analysis enabled by modern diagnostic instruments reduces site time, avoids repeat visits, and enhances operational reliability.  As more data is collected and evaluated, predictive maintenance is becoming a real possibility.

Contact

Joel Robinson
National Business Development Manager, Seaward Electronic
Email: jrobinson@seaward.com
Phone: (980) 867-0755