What changes in the architecture
With a central inverter, many strings are grouped in combiner boxes and run through higher-current DC wiring to a high-power unit, normally installed in a shelter or on a skid. With string inverters, each unit serves a few strings and is located close to the array, so the long wiring runs are AC.
Maximum power tracking
A central inverter usually has few trackers for a very large array, so all the modules associated with the same tracker must share similar conditions. String inverters offer more independent inputs, and therefore tolerate differences in orientation, shading or age between sections of the array better.
Reliability and maintenance
- A central inverter failure takes a large portion of the plant out of operation until it is repaired; with strings, the loss per unit is smaller.
- Replacing a string inverter is a lighter task; a central unit usually requires switching operations, specialized spare parts and more highly trained personnel.
- With strings there are more units to inspect, and more components exposed outdoors, such as fans and communications.
Wiring and losses
The central architecture carries more direct current over long conductors, which requires suitable protection and cross-sections. Strings involve more distributed AC wiring and more protection equipment. Neither is better in every case; cost and losses depend on the geometry of the plant.
How to decide
In large plants on flat, uniform terrain, the central architecture has been used for its cost per unit of power. On roofs, irregular terrain or plants with differing sections, strings win on flexibility. In practice, grid requirements are also evaluated, such as reactive power control and the ability to operate through voltage sags, which depend on the model and not on the architecture.