Solar String Sizing — Cold Voc, Hot Vmp

As many as fit under the inverter's maximum input voltage at the coldest temperature the site ever sees. Open-circuit voltage rises as cells get colder, so the limit is set on a freezing clear morning, not on a hot afternoon. With a 49.5 V module at −0.27%/°C and a 600 V inverter, a −10°C record low gives 54.18 V per module and 11 modules maximum. Drop the design temperature to −25°C and the same string falls to 10.

Silicon's open-circuit voltage has a negative temperature coefficient — typically around −0.27%/°C for Voc. Below the 25°C standard test condition the voltage climbs above nameplate, and it climbs most on exactly the kind of cold clear morning that also produces full irradiance. NEC 690.7 therefore requires maximum system voltage be calculated at the lowest expected ambient temperature. Get it wrong and the inverter's input stage fails — not on commissioning day, but on the coldest sunny morning of its first winter.

Heat. Maximum-power voltage falls as cells warm, and if the string drops below the bottom of the inverter's MPPT window the inverter stops tracking and array output collapses. Cell temperature — not ambient — drives it, and a roof-mounted module runs 25 to 35°C above ambient. At a 65°C cell temperature a 41.5 V module at −0.35%/°C delivers 35.69 V, so an 80 V MPPT floor needs a minimum of 3 modules.

The module's own coefficient, whenever the datasheet gives you one — and it always does. Table 690.7(A) is a fallback for crystalline silicon when the manufacturer's figure is unavailable, and it is deliberately conservative. At −10°C the table applies a flat ×1.14, giving 56.43 V against the 54.18 V the real coefficient produces. That difference costs a module: 10 instead of 11. On a 20-string commercial array, using the table instead of the datasheet throws away 20 modules of capacity for no safety benefit.

The lowest expected ambient at the site, which is a record low rather than an average winter minimum. The usual source is the ASHRAE Extreme Annual Mean Minimum Design Dry Bulb Temperature for the nearest station — the same dataset load calculations use. Do not use the average January low; the string has to survive the worst morning in the design life, not a typical one. If the record low is genuinely uncertain, one module fewer costs a few percent of production and removes the risk entirely.

No — string length moves voltage, not power. Ten modules in series and two strings of five produce the same watts. What longer strings buy is lower current for the same power, which means smaller conductors, less voltage drop, and fewer combiner terminations. That is why designers push toward the maximum. The cost of pushing too far is an inverter, so the sensible practice is to design one module below the calculated ceiling unless the record-low figure is solid.

156% of the module's short-circuit current, arrived at in two steps that people frequently apply only once. NEC 690.8(A) sets the circuit current at 125% of rated Isc — a correction for irradiance above 1000 W/m², not a continuous-load factor. Then 690.8(B) applies the ordinary 125% continuous-load factor on top. The two compound: 1.25 × 1.25 = 1.5625.

Not in length, but they must not share an MPPT input unless they do. Strings of different lengths, orientations, or shading on the same MPPT are dragged to a single operating point and the weaker one drags down the stronger. Different faces belong on separate MPPT inputs, or on module-level electronics. Where strings do share an input, keep length, azimuth, and tilt identical.