Solar designers spend considerable time optimizing string layout, inverter selection, and shading analysis, and rightly so — these decisions drive system yield. But one design decision that gets specified almost as an afterthought, often defaulting to whatever the inverter manufacturer bundles or whatever the racking supplier recommends, is DC isolation. This is a mistake, because isolator selection affects three things that matter throughout the life of the project: technician safety, code compliance, and long-term maintenance cost.
A DC isolator's job sounds simple: disconnect part of the array so someone can work on it safely. In practice, safely interrupting direct current at array voltage is a harder engineering problem than interrupting AC at the same voltage, because DC has no natural zero-crossing to help an arc self-extinguish. A switch rated for AC service, or a DC switch rated for a lower voltage class than the array actually produces, can fail to fully interrupt the circuit — leaving a technician exposed to a live arc during what should be a routine disconnection.
This is why isolator specification should be treated as a genuine design decision tied to the array's electrical architecture, not a commodity purchase made late in procurement.
PV string voltage rises as cell temperature falls, which means the isolator's rated voltage needs to be checked against the array's temperature-corrected maximum open-circuit voltage, not its typical operating voltage on a mild day. A design team working across sites in different climate zones — a site in Arizona and a site in the northern US, for example — cannot use the same isolator spec for both without doing this calculation separately for each, since the colder site will push the array closer to a 600V, 1,000V, or 1,500V threshold sooner than the datasheet's standard test condition figures suggest.
There are typically several candidate locations for DC isolation in a commercial array: at the array itself, integrated into a combiner box, adjacent to the inverter, or some combination of these depending on jurisdiction and system size. A design that concentrates all isolation into a single combiner-integrated switch simplifies the bill of materials but can create a maintenance bottleneck — technicians working on one string may need to de-energize the entire combiner group rather than isolating just the section they need. For larger commercial arrays, this tradeoff between component count and operational flexibility deserves explicit discussion during design review rather than defaulting to whatever the racking or combiner vendor ships as standard.
Some jurisdictions require rapid shutdown functionality on rooftop arrays, and it's worth being explicit with clients and code officials that this is a distinct function from manual DC isolation, even though both interrupt current under certain conditions. Conflating the two in a design package is a common source of plan-review rejections.
Because voltage class, pole configuration, enclosure rating, and code requirements interact rather than existing independently, a structured selection process tends to produce better outcomes than working through a generic checklist. OmniSol's DC isolator guide for PV arrays sets out how to work through these variables in sequence — starting from array voltage and current calculations and ending with enclosure and certification verification — which is useful as a design review reference even for teams that already have their own internal specification process, simply as a way to catch the variables that get missed when isolator selection is handled quickly at the end of a project.
A misspecified isolator rarely causes a problem on day one. It shows up eighteen months later during a service call, when a technician discovers the installed switch isn't actually rated for the array's real-world voltage under cold-morning conditions, or discovers that isolating one string means shutting down six others. These are the kinds of findings that turn a routine maintenance visit into a warranty dispute. Getting isolator selection right at the design stage is considerably cheaper than fixing it in the field.
