The role of solar photovoltaic (PV) systems is growing in the energy industry, where its application ranges from telecommunication towers to C&I areas. In ESI Africa Issue 4 2019, we explored part one of how damage to the solar modules can occur and the prevention techniques. Read on for part two of this instalment.
The attractiveness of solar PV technology is in its minimum system maintenance, reduced operating and maintenance costs, and affording access to pollution-free power. As such, this power market is growing at a rapid pace. Demand for PV is also attributed to the rising demand for energy, support for the technology extended by governments, and the growing concern regarding the environment.
According to research company MarketsandMarkets, the global PV market is expected to grow at a CAGR of 18.30% between 2014 and 2020; and Its overall worth is estimated to grow from $89.52 billion in 2013 to $345.59 billion by 2020. Recently, the solar PV industry has also experienced several new technological advances, such as the introduction of glass-glass modules, dual/ half-cells, floating PV, and bi-facial PV. Currently, the vast majority of all existing and new module sales consist of crystalline silicon cell-based modules, either poly/ multi- or mono-crystalline. The rest of the market consists mainly of various thin-film PV technologies.
This article continues from part one and presents a further four case studies of solar PV module damage and how these can be prevented in future
Case study: Printing
Setting the scene: During the manufacturing of solar cells, conductive silver paste is screen printed on the front side of the solar cell to create the busbars (the two vertical lines) and fingers, also called tracks (the thinner horizontal lines), as in Figure 1. As with conventional (paper and ink) printing, bad prints also occur when manufacturing the busbars and tracks of solar cells.
When the tracks, which serve the purpose of conducting electricity, are not well printed, it can lead to dead areas within the cell. If the dead areas are significant enough, the poorly printed cell will produce less electricity than the other cells in the module. This miss-match will lead to the poorly printed cell heating up.
In some instances, the cells are slightly bowed during manufacturing. When the cells are laminated, which is one of the manufacturing processes, they are flattened. This defect could lead to a crack that runs all along the busbar and leads to what is commonly called broken fingers (see Figure 2).
After the lamination process, the module is tested on the production line. An electroluminescence (EL) image to assess cell cracks and poor printing (see more on EL in Issue 4) and a flash test (to determine the maximum power of the module) is performed. Broken fingers are often not picked up by these tests if the crack has not yet caused an electrical disconnect of the fingers. After manufacturing the module is transported from the factory to the site where it is installed. Over time the cracks may develop into an electrical disconnect of the fingers.
There have been cases on large South African plants where broken fingers have been found in the majority of the installed modules. This defect is typically only identified a few years after the plant has been commissioned and is flagged when the plant underperforms.
One method to mitigate this is to perform module testing before the end of the defects liability period, which is typically two years after commissioning of the plant. It is advised that power and EL testing be performed. These tests are best performed on-site, but lab testing will also suffice. It is also recommended to perform an aerial thermographic inspection of the entire plant with a drone at the same time.
Case study: Thunder
Setting the scene: Since PV modules are installed on roofs or in open fields, lightning damage will inevitably occur. Lightning may cause a current or voltage surge within a string of PV modules. This surge can cause the bypass diodes to short-circuit, or it can result in a disconnected ribbon – both of which are located within the junction box.
A single short-circuited bypass diode will lead to a loss of 1/3 of the module’s generation capacity. In some cases only one bypass diode fails, while in others two or all three diodes fail – which results in a loss of 2/3 or the entire module’s generating capacity.
Since modules are typically connected in strings of 20 modules (1,000V configuration) or 30 modules (1,500V configuration), the loss in production is not picked up on the SCADA system, especially since most plants do not have string level monitoring. However, in most cases, the combined output of two strings is monitored. The production loss of 1/3 of a module out of 40 or 60 modules is too small to detect.
One of the most effective ways to detect defective modules is with aerial thermographic inspections utilising a multi-rotor copter – an infra-red camera mounted on a drone. During such an investigation, different module thermal signatures are identified and classified. Short-circuited bypass diodes lead to a checker-board (also referred to as a patchwork) pattern. The temperature difference between the hottest cell and module average temperature could reach 30°C.
In almost all cases the cells are not damaged. The pattern is caused due to the cells not all being completely similar. This mismatch is amplified given that the cells are operating in short-circuit conditions. The cells that produce less current than the other cells will heat up. The modules can be repaired by replacing the diodes or, in some cases, the entire junction box.
Disconnected ribbons have a homogeneous thermal pattern on the affected area (module sub-string) of the PV module. The temperature difference is typically in the 3–5°C range, which makes it challenging to identify using a hand-held thermal camera (refer Figure 3). In some cases, the disconnected ribbons are not caused by lightning but by a dry or loose connection. The repair is achieved by tightening the connection (in case of springloaded or screw connectors) or resoldering the connection (in cases with soldered connections).
Case study: Hand-me-downs
Setting the scene: Developing countries often report that they are being supplied with low-quality goods. With regard to PV modules, the quality could be the power rating that is incorrectly labelled (or on purpose) or even due to the modules being second hand. In a recent LinkedIn post regarding a large PV plant in Germany that was hit by a hailstorm, a company specialising in on-site testing performed EL and UV inspections and posted some of the results/images of the damaged modules (Figure 5). The CEO of a German-based company replied on the post with the following message: “We are happy to purchase all modules with microcracks and cell breakage. We have customers who build “new” plants with the modules. Mostly from Eastern Europe or Africa. If you have any used modules feel free to contact me.”
The dumping of damaged modules is a reality for developing countries, and the only way to prevent this is to inspect the product before installation. Used (modules that have been in the sun for at least six months) can easily be identified by examining them with a UV light. Figure 6 shows the UV Fluorescence signature of a used (4-year-old) module (left) and new module (right). This method will also indicate cracked cells (refer to part one of this article in ESI Africa Issue 4).
Case study: Downhill
Setting the scene: Once installed, PV modules will degrade over time, and the output decreases each year. Most module manufacturers warrant that the module will not deteriorate by more than 20% over 25 years (Figure 7). There have been cases with large PV projects where the modules degraded faster than the expected rate.
The immediate action is to assess the performance, and specifically the performance ratio (PR), of the plant. If a module degrades faster than the warranty, the owner can lodge a warranty claim against the manufacturer. The manufacturer will be required to conduct third-party tests in an accredited laboratory. The challenge here is in the transportation of the modules to the lab.
This challenge is being addressed worldwide by utilising mobile PV test laboratories to perform the required test on-site. In May 2019, the first such mobile test lab was commissioned in South Africa (Figure 8). This lab is being used locally but also in neighbouring countries. The lab consists of an A+A+A+ LED flasher (to perform a flash rest and determine degradation), high-resolution Electroluminescence (EL) cameras (to assess cell damage), an electrical insulation tester and infrared camera.
Given the pressure to achieve a low-cost and high-volume production of PV modules, low quality goods from some suppliers are inevitable. This can be mitigated through on-site testing and inspection of products but even better is to conduct in-factory inspections. Several consulting companies offer the service to oversee the production of a client’s modules in the production facility. Most developers of large PV projects make use of such services. Some consultants, such as SolarBuyer (now part of PI Berlin), offer historic inspection data. This allows developers to completely avoid engaging with module suppliers with a bad record on quality. ESI
About the author
Riaan Meyer is the managing director of GeoSUN Africa – a Stellenbosch University spin-off company established in 2012 and active throughout the African continent. His work focuses on solar resource assessment and PV module testing and inspection. GeoSUN collaborates with a spin-off company from the Nelson Mandela University in Port Elizabeth, PVinsight, to provide PV test-related services to its clients.