Testing for quality in solar panels
A renewed emphasis on solar module quality raises the question of how quality is evaluated
If you look at the issue of solar panels, you quickly discover that there are numerous makes, models, and sources. With hundreds of manufacturers, many of them from China and other overseas markets, it’s probably not surprising that there can be a vast difference in panel quality – not only among various brands, but between specific factories and individual panels. This matters a lot, as investors are counting on panel performance expected to last 25 years or more, and making sizeable bets, in the hundreds of millions of dollars. It’s going to matter even more in the years to come: Navigant recently issued a report projecting global solar installations of 438,000 MW by 2020, with $134bn in annual revenues.
Industry players and analysts have recently noted that panel quality is becoming an issue, pointing out that executives with companies that inspect Chinese companies for panel quality have recently discovered even well-known companies either substituting cheaper, untested materials or subcontracting assembly to other manufacturers.
According to some manufacturers, the cost per watt metric that everybody uses is pretty flawed. But it’s still used because panel manufacturers are selling watts rather than watt-hours. And that’s because energy yields are different for various reasons. Factors include types of inverters used, whether you have tracking, and different levels of solar radiation. They all affect output. And panels are not created equal. They will perform slightly differently even in the same location.”
Historically, panels were tested in labs using light with a specific angle of incidence, which is very different from the real world, where conditions change. In modern labs, many different environments are created so that banks and others can model energy yield more accurately. Modern labs enable specific forecasting using actual characteristics of panels. Two specific areas they focus on are vendor quality and statistical batch testing on a project-by-project basis.
The vendor quality focus is geared for buyers and investors to understand whether they are investing in the right module for their system. Historically, investments have been based on brand and balance sheet. But the historical methodologies don’t make sense any more.
Testing companies now put modules through accelerated testing in environmental chambers and do performance forecasting. They see a pretty good spectrum of performance: some over-perform and some under-perform. Since they are all selling on a dollars-per-watt basis, it’s very difficult for buyers to differentiate between brand x and brand y.
Statistical batch testing then focuses on volume. If somebody buys 10,000 or 150,000 panels, they perform random spot checks and confirm with some statistical accuracy that the population has no defects. Financiers are starting to ask these questions about quality.
When you compare solar with other energy sources, PV is different in that it is a whole bunch of widgets that are supposed to be the same. So consistency really matters, especially for utility-scale procurement, whether it’s utility projects or the big residential suppliers buying thousands of panels.