Jay A. Kratochvil

Photovoltaic module performance and durability following long-term field exposure

Our investigations of both new and field-aged photovoltaic modules have indicated that, in general, todays commercially available modules area highly reliable product. However, by using new test procedures, subtle failure mechanisms have also been identified that must be addressed in order to achieve 30-year module lifetimes. This paper summarizes diagnostic test procedures, results, and implications of in-depth investigations of the performance and durability characteristics of commercial modules after long-term field exposure. A collaborative effort with U.S. module manufacturers aimed at achieving 30-year module lifetimes is also described.

Analysis of factors influencing the annual energy production of photovoltaic systems

The most relevant basis for designing photovoltaic systems is their annual energy production, which is also the best metric for monitoring their long-term performance. An accurate array performance model based on established testing procedures is required to confidently predict energy available from the array. This model, coupled with the performance characteristics of other balance-of-system components, provides the tool necessary to calculate expected system performance and to compare actual versus expected energy production. Using such a tool, this paper quantifies the effect of the primary factors influencing the DC-energy available from different photovoltaic module technologies, and contrasts these influences with other system-level factors that often result in significantly less AC-energy delivered to the load than the array is capable of providing. Annual as well as seasonal energy production is discussed in the context of both grid-tied and stand-alone photovoltaic systems.

Dark current-voltage measurements on photovoltaic modules as a diagnostic or manufacturing tool

Dark current-voltage (dark I-V) measurements are commonly used to analyze the electrical characteristics of solar cells, providing an effective way to determine fundamental performance parameters without the need for a solar simulator. The dark I-V measurement procedure does not provide information regarding short-circuit current, but is more sensitive than light I-V measurements in determining the other parameters (series resistance, shunt resistance, diode factor and diode saturation currents) that dictate the electrical performance of a photovoltaic device. The work documented here extends the use of dark I-V measurements to photovoltaic modules, illustrates their use in diagnosing module performance losses and proposes their use for process monitoring during manufacturing.

Stabilization and performance characteristics of commercial amorphous-silicon PV modules

The successful commercialization of any new photovoltaic technology is difficult. Understanding the products performance and aging characteristics is a prerequisite for the manufacturer. Amorphous-silicon thin-film modules are now in commercial production, and their market penetration is being limited to some degree by a lack of understanding of environmental influences that impact system design and operation. This paper summarizes our detailed performance characterization of multiple modules from four different manufacturers over several years of continuous outdoor exposure in Albuquerque, NM. Common stabilization characteristics have been observed for both tandem and triple-junction modules, and the influences of solar spectrum and seasonal (thermal) annealing have been clearly identified. Implications for system performance modelling are presented.

Applications for infrared imaging equipment in photovoltaic cell, module, and system testing

Anomalous temperature distributions are often an indication of atypical behavior in a device under investigation. Portable infrared (IR) imaging systems (cameras) now provide a convenient method for measuring both absolute and relative temperature distributions on small and large components with a high degree of temperature and spatial resolution. This diagnostic tool can be applied during the development, production, monitoring, and repair of photovoltaic cells, modules, and systems. Planar objects with nearly uniform material composition are ideally suited for analysis using IR imaging. This paper illustrates investigations of localized shunting in cells, resistive solder bonds in field-aged modules, module bypass diode functionality, reverse-bias (hot spot) heating in modules, temperature distributions in flat-plate and concentrator modules, batteries during charging, and electronic component temperature in power processing equipment.

Long-term performance and reliability assessment of 8 PV arrays at Sandia National Laboratories

In the last decade, c-Si module degradation rates of ≪1%/year have been reported [1–3]. It is unclear if this degradation rate extends directly to the string level and what is the expected statistical spread of degradation rates. Nine photovoltaic (PV) arrays totaling nearly 100 kW at Standard Reporting Conditions are currently being used at Sandia National Laboratories (SNL) primarily for inverter testing. The measured power degradation of these arrays at the string level varied from no change over three years within measurement error to greater than 25% in three years. This paper outlines the methodology used to test the DC output, outlines analysis techniques used to evaluate the array performance, provides a current reliability assessment, presents the comparative data for up to five years of use and exposure, and discusses the methods used to track down the causes of unexpected string-level degradation.

Comparison of module performance characterization methods

The rating and modeling of photovoltaic (PV) module performance has been of concern to manufacturers and system designers for over 20 years. Both the National Renewable Energy Laboratory (NREL) and Sandia National Laboratories (SNL) have developed methodologies to predict module and array performance under actual operating conditions. This paper compares the two methods of determining the performance of PV modules. The methods translate module performance to actual or reference conditions using slightly different approaches. The accuracy of both methods is compared for both hourly, daily, and annual energy production over a year of data recorded at NREL in Golden, CO, USA. The comparison of the two methods is presented for five different PV module technologies.

Diagnostic analysis of silicon photovoltaic modules after 20-year field exposure

The objective of this study was to investigate the technology used by Spectrolab Inc. to manufacture photovoltaic modules that have provided twenty years of reliable service at Natural Bridges National Monument in southeastern Utah. A field survey, system performance tests, and a series of module and materials tests have confirmed the durability of the modules in the array. The combination of manufacturing processes, materials, and quality controls used by Spectrolab resulted in modules that have maintained a performance level close to the original specifications for twenty years. Specific contributors to the durability of the modules included polyvinyl-butyral (PVB) encapsulant, expanded metal interconnects, silicon oxide anti-reflective coating, and excellent solder/substrate solderability.

Experimental optimization of the performance and reliability of stand-alone photovoltaic systems

Stand-alone photovoltaic systems are deceptively complex. Optimizing the performance and reliability of these systems requires a complete understanding of their behavior as a function of site-dependent environmental conditions. Individual component specifications provide useful design information. However, to fully understand the interactions between components, it is necessary to simultaneously characterize the performance of the system and its separate components under actual operating conditions. This paper describes how a new 30-day outdoor testing procedure was coupled with array performance modeling to accomplish this objective. The procedure measures battery capacity, determines appropriate set-points for charging, and based on daily intervals quantifies DC-energy available from the array, charge-controller efficiency, battery efficiency, inverter efficiency, overall system efficiency, days of autonomy, and AC-energy available by month.

Field investigation of the relationship between battery size and PV system performance

Four photovoltaic-powered lighting systems were installed in a National Forest Service campground in June of 1991. These systems have identical solar cell arrays, loads and charge controllers. The only difference was in the rated capacity of the battery bank for each system. The battery banks all use the same basic battery as a building block with the four systems utilizing either one battery, two batteries, three batteries or four batteries. The purpose of the experiment is to examine the effect of the various battery sizes on the ability of the system to charge the battery, energy available to the load, and battery lifetime. Results show an important trend in system performance concerning the impact of charge controllers on the relation between array size and battery size which results in an inability to achieve the days of battery storage originally designed for.<<ETX>>