In utility-scale solar operations, the choice between centralized and string inverters is no longer just an engineering preference but an O&M strategy with major implications for reliability, downtime, and long-term performance. Centralized inverters give you high capacity, simplified AC integration, and lower upfront costs, which is attractive for large, uniform terrains. But they also create a major operational risk: when a central unit trips, an entire block can drop off the grid at once, leading to large energy losses and requiring heavy-equipment support for repairs. String inverters flip the equation. Their distributed architecture improves fault isolation, reduces downtime, and gives operators granular visibility into module- or string-level performance. In plants where shading, soiling patterns, or uneven module ageing are challenges, string inverters often deliver better energy yield and faster troubleshooting. Still, they come with more components, more distributed maintenance, and potentially higher replacement frequency. So how do we decide? The answer often lies in site conditions, workforce capability, reliability targets, and O&M philosophy. Centralized systems may suit teams with strong heavy-equipment maintenance capacity and stable grid conditions, while string architectures shine where uptime, diagnostics, and agile maintenance matter most. Some developers now use hybrid designs, combining the robustness of centralized units with the intelligence of string-level monitoring. Do string inverters give you better diagnostics, or do their numbers overwhelm your maintenance plan? which architecture has proven more reliable? and If you had to design a new plant today, which would you choose and why?

Hello Guys I’m currently dealing with an issue where one of our solar parks is causing continuous power fluctuations to the grid, which has led the grid operator to shut down the plant when this occurs. The plant has a total installed capacity of 14 MW. At the time of commissioning, the grid infrastructure was not fully ready, so the plant was limited to 50% capacity. This limitation was applied directly at inverter level, where all Sungrow SG250HX inverters were capped at 50% output. In addition, the plant is controlled by a PPC (Power Plant Controller), which performs curtailment based on market signals (e.g., negative pricing). After approximately one year, maintenance was completed on the grid export infrastructure, and the plant was allowed to operate at 100% capacity. Following this change, we started to observe serious instability whenever the plant was actively curtailed by the PPC. We noticed that the inverters were generating large amounts of reactive power, and the cause was initially unclear. This issue persisted for several months. In summary, the inverters receive curtailment and control signals from the PPC, and we strongly suspect that the PPC control parameters (active and reactive power setpoints, control gains, ramps, or droop settings) were tuned when the plant was operating at 50% and were never updated after the plant increased to full capacity. This appears to be causing unstable inverter behavior and oscillations seen by the grid. A key observation is that when we disable the PPC/loggers and turn off both active and reactive power control, the fluctuations stop immediately. This strongly suggests the issue is related to PPC control logic and configuration rather than the inverters or the grid itself. I’d appreciate your thoughts on this.

Is there a simple way to weather adjust data (roughly) to be used for 1,000s of Residential assets clustered in a very small Region? We have monthly performance reviews and the same question comes up? Why was last year batter than this year? Etc… and the answer to all these questions is likely a column with some basic weather adjustment/month. Wanted to get thoughts as this would change my life! Cheers!

Don´t you have the impression that we are overwhelmed with Tier1 products all over?! Despite of existing Bloomberg definition of Tier1, I have the impression that many of the market players have self-declared their Tier1 status. Somehow similar to star-rating in the hotel industry. On the other hand, it can be that some of the "ordinary" module makers might be even better quality than many Tier1. So the question is: Does "Tier1" have any specific meaning for you? Is that the decision-driver?

It happened to me several times. Talking to a project owner. Some lousy products are on the BOM. I advise the owner to re-consider replacing some of the components. Reaction? Nah, I don´t want to add more money into the project. If something goes wrong, I am covered with insurance. My question is obvious: Are the insurance companies aware of the risk which are they taking? And what they do to minimize it? Does anyone here have any practical experience with insurance companies? And how do the insurance companies react for the exponentially growing risk?