Technical Note 1:
Solar Panel Faults
and Reduced Lifespan of solar installations.
Operational faults in PV (photovoltaic) arrays have always been one of the critical factors affecting the PV system’s power-generation efficiency and life cycle duration. Solar panel reliability and durability in terms of performance can vary as a function of materials, manufacture, design, transport and installation. In turn, power production levels can vary as a function of solar irradiation levels; temperature; spectral response of panels; light conditions including incidence angle; Light Induced Degradation; and various extremes of weather, climate, and soiling.
Nonetheless, the 'traditional wisdom' across the solar industry today is that solar panels should be wired in series in what is termed String Series, especially for those housed within large-scale Solar Energy Facilities (SEF’s). Several strings can in turn be connected in series via a Combiner Box and several combiner Boxes again in series to one or more Central Inverter(s). Therein, when panel faults occur all panels within a given String Series fall-back to the lowest common denominator, i.e. the panel with the lowest level of power production. Because such faults go undetected over lengthy periods, they often only detected as a result of what are termed ‘catastrophic failures’, i.e. once function ceases all together. Earlier detection strategies would allow for failing performance to be detected and / or predicted possibly many years in advances of such problems. In the severest cases, the latter can result in fire and shutdown of an entire solar plant comprised of many tens or hundreds of thousands of solar panels.
In large-scale SEF's, the 'traditional wisdom' states that economies of scale can afford the overall system losses that go undetected or impact only slightly the overall performance of one or a few solar panels in amongst hundreds and sometimes thousands and hundreds of thousands of 'String Series' of solar panels. An individual String Series can vary from 4 to 72 panels and occasionally more, but occur most frequently as Strings of 12, 20, 24 or 36 solar panels. In turn, multiple strings can be combined in a String Connector / String Combiner box to the extent that it is not uncommon for a few to several hundred solar panels to be interconnected in series; and this along with commonly 60, 72 or 90 solar cells per module again connected in series. Even with improving levels of solar module performance, this constitutes considerable potential for fault occurrence over two or three decades.
The actual location of faults or impending faults with respect to a particular solar panel in these agglomerations remains a major technical challenge. The difficulty of accurately locating solar panels with faults or impending faults is further exacerbated when multiple faults occur in parallel within the same series or within the same large-scale SEF’s. As the size of the latter increases, the likelihood of a co-occurrence of a multiplicity of under-performing solar panels increases dramatically, i.e. it approaches a probability of occurrence = 1. This multiplicity of faults currently defies all methods of detection and location within large-scale SEF’s.
The contrary scenario is the major negative impact caused by the loss of optimal functionality provoked by one or two panels in a small domestic rooftop setting of perhaps ten panels, for example, whereby underperformance to the tune 10% or 20% of system performance can result, i.e. if panels are linked in series. Thus, to date, micro-inverters have been deployed mostly in domestic roof-top settings and some larger industrial roofs and carpark shading systems. Micro-inverters are mounted in parallel and work to maintain maximal performance of individual solar panels independently of all other.
Lost performance in one or two panels and therefore one or two String Series in amongst hundreds or thousands of solar panel strings is currently perceived as not warranting the additional expense of wiring solar panels in parallel so as to monitor panel performance individually by the use of micro-inverters. This industry view has not changed since the inception of large-scale SEF’s over several decades. However, one should ask in the Age of the “Internet of Things” why this view should remain constant in the case of large-scale SEF’s and not be better attuned with higher-resolution continuous systems monitoring that increasingly dominate other industrial sectors so as to ensure predicted levels of power production, longer term systems durability and low levels of investor uncertainty.
In the case of SEF’s, financiers, asset owners and developers continue to assume that solar panels will last for 25-30 years and thus they decide NOT to underwrite the additional cost of micro-inverter installation at outset, which from time to time can act to their detriment (See following text).
‘In systems engineering across all industrial sectors, until a phenomenon is measured over time, efficiencies in terms of economy, production and quality of output cannot be delivered with confidence’.
In addition to factors giving rise to differential settlement in Solar Covered Landfill (see LanneSolaire Technical note 2), here the issues surrounding individual solar panel and solar cell reliability and durability shall be examined, namely, a long list of causes that impact negatively solar energy production of SEF’s. To these must be added cabling and central inverter malfunction. Central inverter faults can be at the base of major system shutdown and significant losses in power production until repaired by a technician or replaced. In large-scale and domestic SEF’s, inverter failure is by far the most commonly-recorded fault type. In addition, central inverters must be replaced regularly (every 5-10 years) at high cost relative to other operations and maintenance expenses during systems lifespans. These expenses at SEF installation and the replacement costs of one or more central inverters is obviated all together by more generalised deployment of micro-inverters at SEF installation. Here, the issues surrounding transformers, charge controllers or overall PV-array availability shall not be addressed.
PV arrays perform below optimum output power levels due to faults in modules, wiring, inverter, and so forth. Today, most of these faults remain undetected for extended periods and result in power production losses. Technicians need to be sent into the field to locate and repair faults within a SEF ‘String Series’. The need to make time consuming field measurements within one or more ‘String Series’ can delay fault rectification. Today, most module-related and other fault types are detected, at best, belatedly. It is noteworthy that: “Systems durability is the product of its reliability over lifetime and is key to investment returns”. PV array faults left unrectified (noncleared) not only cause losses in power production, but also may lead to safety issues and fire hazards of varying proportions.
figure 1: Cascading failures of a PV module
In extreme cases, excessive heat production can provoke fire in a domestic, industrial or large-scale SEF. These fires can be prevented through by the introduction of an « automatic shutdown » capacity incorporated at the level of individual solar panels and not as at the level of string or central converters, which shutdown after a catastrophic fault in the form of fire. The former is a statutory obligation in domestic settings in the USA, but not yet in France or Europe, for example.
Figures 4: Fires associated with overheating of PV modules in domestic and industrial contexts
Historically, financiers, asset owners and solar developers have been content to assume solar panel longevity as a given at 25 to 30 years. A major caveat here is the lack of a significant body of evidence as to longevity of solar modules in the field to the extent of 20-30 years of durability. This is also underlined by the impossibility of obtaining such data for technologies developed since the year 2000, yet currently placed in large-scale SEF’s. This view of an assumed reality is reflected in what is termed the ‘bathtub curve’. In the first 1-3 years following installations defects become rapidly evident as reliability issues related to catastrophic failure and thereafter a steady rate of degradation-related failure issues prior to end-of-life module wear-out. In all cases, these faults must engender ‘Catastrophic Faults’ to be detected rapidly within ‘String Series’. Other electrical faults responsible for lower levels of power production may remain ‘uncleared’ or undetected over extended periods.
figure 3: Evolution of the failure rate over time
Until very recently, the best available data for field-based module failure is summarized in the following figure (After: TamizhMani, G.S. & Kuitche, J. 2013, Accelerated Lifetime Testing of Photovoltaic Modules, pp 1 - 106. Solar America Board for Codes and Standards) and based on datasets available in 2011.
Figure 4 : Reliability (failures) and durability (degradative) issues of photovoltaic modules
(FF = Field Faults)
Revelations from 2020: The abovementioned historical wisdom / industry view may start to be brought into question by recent revelations. The chemical company, Dupont, published in 2020 their findings on some 551 different SEF’s and 53 different solar module manufacturers (cf. Dupont global PV reliability – 2020 field analysis). Much to the surprise of industry specialists, faults were detected in 30% of nine million solar panels inspected during this study. This figure corresponded to serious “backsheet” faults in 16% of cases and cell-to-cell connection faults in the remaining 14% of cases. Some affected asset owners were reported to be in ‘panic mode’ today, if indeed their own SEF’s were affected. Furthermore, “backsheet” defects were heavily influenced by their chemical composition. In the case of PVDF (Polyvinylidene fluoride), “backsheet” cracking defects were observed in 23% of solar modules after just nine years in the field, while PET (Polyethylene terephthalate ) performed still worse with some 90% of modules presenting cracking defects after 15 years – all well in advanced of the assumed lifespan of 25-30 years for module reliability. These defects result chronologically from “backsheet” failure (Outer air-side layer) giving rise to 1) moisture ingress; 2) cracking, delamination and yellowing; 3) corrosion, ‘Hot Spots’, ‘Snail trails’, burn marks, short circuits; and finally, 4) Inverter tripping, PV array ‘shutdown’ (Zero power production) due to advanced catastrophic fault and safety risks, including fire.
Thus, more high-resolution continuous systems monitoring of SEF’s may be seen in a more positive and cost-effective light than previously, i.e. so as to detect and / or predict such occurrences much earlier, while concomitantly having faulty solar modules replaced at the manufacturer’s expense under warranty.
The following Table is an attempt to capture the origins and underlying causes of the most frequently encountered solar module faults and SEF under-performance - Table : Solar Panel Faults, Inverter faults, SEF underperformance.
Further details pertaining to PV array faults and an introduction to existing fault detection strategies can be found in the following references:
TamizhMani, G.S. & Kuitche, J. 2013, Accelerated Lifetime Testing of Photovoltaic Modules , pp 1 - 106. Solar America Board for Codes and Standards
Alam, A., Johnson, J., Khan, F. & Flicker, J. 2015. A Comprehensive Review of Catastrophic Faults in PV Arrays: Types, Detection, and Mitigation Techniques. IEEE J. Photovoltaics, doi: 10.1109/JPHOTOV.2015.2397599
Arani, M.S. & Hejazi, M.A. 2016. The Comprehensive Study of Electrical Faults in PV Arrays. J. Electric. Computer Eng. Pp. 1-10, doi : 10.1155./2016/8712960.
Extra reading :