How cold solder joints turn solar modules into warranty cases

The anonymous investor in this case had planned and commissioned the construction of PV plants in Europe over three years with a total generation capacity of 50 MW.

However, initial testing during final acceptance at the end of the engineering, procurement and construction (EPC) warranty period revealed a mess. As is often the case, an operations and maintenance (O&M) service provider was supposed to carry out this inspection. In this case, it was the same company that built the plants as EPC provider.

During visual inspection, spots in the modules looked as though they had already heated up during operation and thermographic images showed anomalies.

Anika Giller, responsible for business development in the Europe, Middle East and Africa and Asia-Pacific regions at Clean Energy Associates, says the O&M service provider spent months negotiating with the module manufacturer’s European representatives over how to proceed, “without success”. The company, CEA for short, is a partner in pv magazine’s module tests and, as a global technical advisory business for PV and battery storage projects, carries out defect analyses and due diligence for module buyers.

Contractual gray areas

In principle a warranty claim would, of course, be justified. However, if there is no contractual agreement on how to prove a warranty claim, the claimant is even more dependent on the good grace of the manufacturer.

“Our experience shows that even in cases of obvious product defects, the manufacturers will push back, and this is even worse when the defects are hidden,” says Giller. “In some cases, we have found that recognizing such a claim would have set a precedent for similar claims without any way to estimate the number of modules affected. That is a very big risk for manufacturers.”

Double-checking each module in the laboratory and confirming the fault using all technically available means would be prohibitively expensive. That is the point at which the O&M company called in CEA. “Our advantage is that we had direct access and [a] relationship to the decision makers in China,” says Giller.

Consider the manufacturer's mindset

The consultants managed to agree with the manufacturer in advance to a procedure whereby CEA surveyed 85% of the 50 MW, or around 180,000 solar panels, with electroluminescence at night, says Giller. What made such an agreement possible was accepting the mindset of the manufacturer, so that solutions could be proposed that were both acceptable and meaningful to all of the parties. To ensure that these measurements could be carried out effectively, CEA worked with the manufacturer to determine the image resolution, which was just good enough to detect the faulty solder joints and convince the manufacturer to accept them as proof of a warranty claim.

“We used a tripod that allowed us to walk quickly through the rows and set the camera at the right distance,” says Giller. In total, it took only two months. It showed that cold solder joints were present in 5% of the modules. Nobody had any idea that the situation was so serious. Cold solder joints occur when the temperature during the soldering process is not high enough, and they are difficult to see with the naked eye.

“It turned out to be important that we had agreed in advance with the manufacturer on the criteria for replacing modules,” says CEA's Director of Technology and Quality, George Touloupas. This made it possible to have discussions on a technical level without the distraction of the value at stake. It is also important that the contract not only states this in English, but also in Chinese.

“In our experience, what ends up in China is sometimes different from what was agreed in Europe,” he says. “What counts in the end is that the decision-makers and the technical team in China understand everything rather than relying on words written in English.”

In this case, it worked out well. The manufacturer has agreed to replace more than 5,000 modules with a current market value of almost half a million euros, according to Giller. What helped was that it could easily be demonstrated that this was a production defect.

Comparison with production audit

“If we do a production audit when we purchase the modules, we check a number of parameters including the soldering temperature. Another test we do very frequently, several times a day, is the peel-off force between the ribbon and the solar cells,” adds Giller.

That makes for a due diligence process which, in addition to a production audit, includes clearly defining quality criteria and providing advice on contracts. In this particular case, that check would have cost around one-fifth of what the field testing ended up costing.

“In another warranty claim case, proper upstream due diligence would have cost only a tenth compared to the downstream testing that had to be carried out to detect faulty modules, but this depends on the type and magnitude of the on site inspection,” says Giller, emphasizing that the added value of an upstream audit is not just in cost savings but in the increased level of confidence a buyer is receiving.

Who foots the bill?

In this case, the additional costs were borne by the EPC and O&M service provider. Added to that were the costs for additional installation work and the personnel expenses of assisting with the warranty claim. “The investor is often stuck with the bill for these,” warns Giller. That depends on the contract the investor has with the EPC and, according to Giller, such contracts routinely fail to define criteria that address problems such as cold solder joints.

If 5% of modules have 20% lower output, the total capacity reduction depends on the distribution of faulty modules in a power plant. If the total lost yield is only 1%, it may go unnoticed even though returns decrease. The question also arises as to whether non-replacement of the modules with the cold solder joints could lead to the development of hot spots that could continue to adversely affect output.