Jay Johnson

Photovoltaic DC Arc Fault Detector testing at Sandia National Laboratories

The 2011 National Electrical Code® (NEC®) added Article 690.11 that requires photovoltaic (PV) systems on or penetrating a building to include a listed DC arc fault protection device. To fill this new market, manufacturers are developing new Arc Fault Circuit Interrupters (AFCIs). Comprehensive and challenging testing has been conducted using a wide range of PV technologies, system topologies, loads and noise sources. The Distributed Energy Technologies Laboratory (DETL) at Sandia National Laboratories (SNL) has used multiple reconfigurable arrays with a variety of module technologies, inverters, and balance of system (BOS) components to characterize new Photovoltaic (PV) DC AFCIs and Arc Fault Detectors (AFDs). The devices detection capabilities, characteristics and nuisance tripping avoidance were the primary purpose of the testing. SNL and Eaton Corporation collaborated to test an Eaton AFD prototype and quantify arc noise for a wide range of PV array configurations and the system responses. The tests were conducted by generating controlled, series PV arc faults between PV modules. Arc fault detection studies were performed on systems using aged modules, positive- and negative-grounded arrays, DC/DC converters, 3-phase inverters, and on strings with branch connectors. The tests were conducted to determine if nuisance trips would occur in systems using electrically noisy inverters, with series arc faults on parallel strings, and in systems with inverters performing anti-islanding and maximum power point tracking (MPPT) algorithms. The tests reported herein used the arc fault detection device to indicate when the trip signal was sent to the circuit interrupter. Results show significant noise is injected into the array from the inverter but AFCI functionality of the device was generally stable. The relative locations of the arc fault and detector had little influence on arc fault detection. Lastly, detection of certain frequency bands successfully differentiated normal operational noise from an arc fault signal.

A Comprehensive Review of Catastrophic Faults in PV Arrays: Types, Detection, and Mitigation Techniques

Three major catastrophic failures in photovoltaic (PV) arrays are ground faults, line-to-line faults, and arc faults. Although there have not been many such failures, recent fire events on April 5, 2009, in Bakersfield, CA, USA, and on April 16, 2011, in Mount Holly, NC, USA, suggest the need for improvements in present fault detection and mitigation techniques, as well as amendments to existing codes and standards to avoid such accidents. This review investigates the effect of faults on the operation of PV arrays and identifies limitations to existing detection and mitigation methods. A survey of state-of-the-art fault detection and mitigation technologies and commercially available products is also presented.

Differentiating series and parallel photovoltaic arc-faults

The 2011 National Electrical Code® requires PV DC series arc-fault protection but does not require parallel arc-fault protection. As a result, manufacturers are creating arc-fault circuit interrupters (AFCIs) which only safely de-energize the arcing circuit when a series arc-fault occurs. Since AFCI devices often use the broadband AC noise on the DC side of the PV system for detection and series and parallel arc-faults create similar frequency content, it is likely an AFCI device will open in the event of either arc-fault type. In the case of parallel arc-faults, opening the AFCI will not extinguish the arc and may make the arc worse, potentially creating a fire. Due to the fire risk from parallel arc-faults, Tigo Energy and Sandia National Laboratories studied series and parallel arc-faults and confirmed the noise signatures from the two arc-faults types are nearly identical. As a result, three alternative methods for differentiating parallel and series arc-faults are presented along with suggestions for arc-fault mitigation of each arc-fault type.

Photovoltaic prognostics and heath management using learning algorithms

A novel model-based prognostics and health management (PHM) system has been designed to monitor the health of a photovoltaic (PV) system, measure degradation, and indicate maintenance schedules. Current state-of-the-art PV monitoring systems require module and array topology details or extensive modeling of the PV system. We present a method using an artificial neural network (ANN) which eliminates the need for a priori information by teaching the algorithm “good” performance behavior based on the initial performance of the array. The PHM algorithm was tested on two PV systems under test at the Outdoor Test Facility (OTF) at the National Renewable Energy Laboratory (NREL). The PHM algorithm was trained using two months of AC power production. The model then predicted the output power of the system using irradiance, wind, and temperature data. Based on the deviation in measured AC power from the AC power predicted by the trained ANN model, system outages and other faults causing a reduction in power were detected. Had these been commercial installations, rather than research installations, an alert for maintenance could have been initiated. Further use of the PHM system may be able to indicate degradation, detect module or inverter failures, or detect excessive soiling.

Lab Tests: Verifying That Smart Grid Power Converters Are Truly Smart

During the last few years, many countries around the world have seen a massive deployment of distributed energy resources (DERs) in their distribution systems. In certain regions, penetration has reached levels that increasingly challenge traditional power system management, affecting the overall stability, reliability, and efficiency of grids. The uncoordinated response of large numbers of DERs may even put overall grid security at risk. This fact was clearly highlighted by the famous 50.2 Hz problem in Europe: it was discovered that the simultaneous tripping of several gigawatts of DERs due to a minor overfrequency event could potentially lead to an undersupply in the European power system so large that it could not be compensated for by using conventional reserve capacities.

Crosstalk nuisance trip testing of photovoltaic DC arc-fault detectors

To improve fire safety in PV systems, Article 690.11 of the 2011 National Electrical Code (NEC) requires photovoltaic (PV) systems above 80 V on or penetrating a building to include a listed arc-fault protection device. Many arc-fault circuit interrupter (AFCI) devices are currently being listed and entering the market. Depending on the manufacturer, AFCIs are being deployed at the module-level, string-level, or array-level. Each arc-fault protection scheme has a different cost and arc-fault isolation capability. Module-level and string-level AFCI devices tout the ability to isolate the fault, identify the failed PV component, and minimize the power loss by selectively de-energizing a portion of the array. However, these benefits are negated if the arcing noise-typically used for arc-fault detection-propagates to parallel, unfaulted strings and cause additional AFCI devices on the PV array to trip. If the arcing signature “crosstalks” from the string with the arc-fault via conduction or RF electromagnetic coupling, the location of the arc-fault cannot be easily determined and safe PV generators will be disconnected. Sandia National Laboratories collaborated with Texas Instruments to perform a series of nuisance trip scenarios with different PV configurations. Experimental results on a 2-string array showed arc detection on the faulted string occurred an average of 19.5 ms before unfaulted string-but in some cases the AFCI on both strings would trip.

Arc fault signal detection - Fourier transformation vs. wavelet decomposition techniques using synthesized data

Arc faults are a significant reliability and safety concern for photovoltaic (PV) systems and can cause intermittent operation, system failure, electrical shock hazard, and even fire. Further, arc faults in deployed systems are seemingly random and challenging to faithfully create experimentally in the laboratory, which makes the study of arc fault signature detection difficult. While it may seem trivial to simply record arcing signatures from real-world system, an obstacle in capturing these arc signals is that arc faults in the PV systems do not happen predictably, and depending on the location of the sensors relative to the arc location, may contribute a negligible portion to the magnitude of the sensed current or voltage waveform. The high-frequency content of the arc requires fast sampling, long memory, and fast processing to acquire, store, and analyze the waveforms; this adds substantial balance-of-system cost when considering widespread deployment of arc fault detectors in PV applications. In this paper, we study the performance of the fast Fourier transform arc detection method compared to the wavelet decomposition method by using synthetic waveforms. These waveforms are created by combining measured waveforms of normal background noise from inverters in DC PV arrays along with waveforms of arcing events. Using this technique allows the ratio of amplitudes are varied. Combining these separate waveforms in various amplitude proportions enables creation of test signals for the study of detection algorithm efficacy. It will be shown that the wavelet transformation technique produce more easily recognized detection results and can perform this detection using a much lower sampling rate than what is required for the fast Fourier transform.

PV ground-fault detection using spread spectrum time domain reflectometry (SSTDR)

A PV ground-fault detection technique using spread spectrum time domain reflectometry (SSTDR) method has been introduced in this paper. SSTDR is a reflectometry method that has been commercially used for detecting aircraft wire faults. Unlike other fault detection schemes for a PV system, ground fault detection using SSTDR does not depend on the amplitude of fault-current and highly immune to noise signals. Therefore, SSTDR can be used in the absence of the solar irradiation as well. The proposed PV ground fault detection technique has been tested in a real-world PV system and it has been observed that PV ground fault can be detected confidently by comparing autocorrelation values generated using SSTDR. The difference in the autocorrelation peaks before and after a ground-fault in the PV system are significantly higher than the threshold set for ground-fault detection.

Electrical simulations of series and parallel PV arc-faults

Arcing in PV systems has caused multiple residential and commercial rooftop fires. The National Electrical Code® (NEC) added section 690.11 to mitigate this danger by requiring arc-fault circuit interrupters (AFCI). Currently, the requirement is only for series arc-faults, but to fully protect PV installations from arc-fault-generated fires, parallel arc-faults must also be mitigated effectively. In order to de-energize a parallel arc-fault without module-level disconnects, the type of arc-fault must be identified so that proper action can be taken (e.g., opening the array for a series arc-fault and shorting for a parallel arc-fault). In this work, we investigate the electrical behavior of the PV system during series and parallel arc-faults to (a) understand the arcing power available from different faults, (b) identify electrical characteristics that differentiate the two fault types, and (c) determine the location of the fault based on current or voltage of the faulted array. This information can be used to improve arc-fault detector speed and functionality.

PV faults: Overview, modeling, prevention and detection techniques

Recent PV faults and subsequent fire-hazards on April 5, 2009, in Bakersfield, California, and April 16, 2011, in Mount Holly, North Carolina provide evidence of a lack of knowledge among PV system manufacturers and installers about different PV faults. The conducted survey within the scope of this paper describes various faults in a PV plant, and explains the limitations of existing detection and suppression techniques. Different fault detection techniques proposed in literatures have been discussed and it was concluded that there is no universal fault detection technique that can detect and classify all faults in a PV system. Moreover, this digest proposes a transmission line model for PV panels that can be useful for interpreting faults in PV using different refelectomery methods.