How the two differ in kind
A photovoltaic array is constrained in two independent ways, and they are different kinds of constraint: one is a hard limit that destroys equipment when it is crossed, the other is an economic optimisation with no hard edge at all.
STRING SIZING is a VOLTAGE problem. Modules in series add their voltages, and the number that may be put in a string is bounded at both ends by the inverter's DC input window. The part that catches people is the direction of the temperature effect: a module's voltage RISES as its temperature FALLS. So the highest voltage a string will ever produce is not on a hot summer afternoon — it is at first light on the coldest morning of the year, with full irradiance on cold modules and no heat yet built up. If that voltage exceeds the inverter's maximum DC input, the inverter is damaged and the warranty does not cover it, so the check is performed at the site's record low temperature rather than at any typical value.
The other end of the window matters too. The lowest string voltage occurs on the hottest day with the modules at their maximum operating temperature, and if it drops below the inverter's MPPT minimum, the string falls out of the tracking range and produces little or nothing at exactly the time the sun is strongest.
INVERTER SIZING is a POWER and ENERGY problem, and it works on a completely different logic. An array reaches its nameplate rating only under test conditions that almost never occur together in the field, so an inverter matched exactly to the array's nameplate spends nearly every hour of its life running well below capacity. Sizing the array LARGER than the inverter — a DC-to-AC ratio above one — captures more energy through the long shoulders of the day and the many hours of partial irradiance, at the cost of CLIPPING the few peak hours where the array could have produced more than the inverter can pass.
That trade is usually strongly favourable, which is why an array larger than its inverter is normal, deliberate, and not the error it appears to be.
The factors that actually differ
Show
| String sizing | Inverter sizing | |
|---|---|---|
| What it constrains | VOLTAGE — how many modules may be put in series. | POWER — how much of the array's output can be converted at any instant. |
| The limiting condition | The COLDEST expected temperature, because module voltage rises as temperature falls. | The few peak-irradiance hours of the year, where output exceeds the inverter's capacity. |
| Kind of limit | A hard one. Exceeding the inverter's maximum DC input damages it and voids the warranty. | A soft one. Exceeding it merely clips, and the inverter protects itself. |
| The other end of the range | The hottest day — the string must stay above the MPPT minimum or it stops tracking. | An inverter far larger than the array runs inefficiently at low load and costs more. |
| The counter-intuitive part | Winter is the dangerous season, not summer. | The array is deliberately larger than the inverter, and that is correct. |
| What the design uses | The module's temperature coefficient of voltage and the site's record low and high temperatures. | A DC-to-AC ratio, chosen against the site's irradiance profile and the array's orientation. |
| Effect of orientation | None on the voltage limits. | Large. An east-west or shallow-pitched array has a flatter profile and tolerates more oversizing. |
| What failure looks like | A destroyed inverter on a cold clear morning, or a string that does nothing in the afternoon heat. | A flat-topped output curve on the sunniest days — energy lost, nothing broken. |
| Which is negotiable | Not at all. It is an equipment rating. | Entirely. It is an economic choice with a broad optimum. |
| Who checks it | Anyone designing the array, and inspection regimes commonly ask for the calculation. | Whoever is optimising yield against cost, which is often the same person and a separate exercise. |
Which one, and when
Choose string sizing when…
- Deciding how many modules go in a series string, which is the check that protects the inverter.
- Cold climates especially, where the record low temperature drives the maximum voltage hard.
- Where modules or inverters are being substituted, since both the coefficient and the input window change.
- Where a string is producing nothing on hot afternoons, which points at the MPPT minimum rather than a fault.
Choose inverter sizing when…
- Choosing the inverter's rating against the array, where a ratio above one is normal.
- Modelling annual yield, where the clipped energy has to be weighed against the energy gained elsewhere.
- East-west, shallow-pitch or diffuse-climate arrays, whose flatter output profile tolerates more oversizing.
- Where an export limit or a grid connection agreement caps the AC output independently.
Now run your own numbers
This page holds no prices on purpose — a national average is wrong for almost every real project. Quantify both options with your dimensions and your local quotes.
Frequently asked questions
- Why does cold weather raise the voltage?
- Because a photovoltaic cell's open-circuit voltage has a negative temperature coefficient: it increases as the cell gets colder. Current is driven largely by irradiance and changes comparatively little with temperature, but voltage moves measurably and predictably in the opposite direction to temperature, which is why manufacturers publish a coefficient for it. The consequence is that the worst case for the inverter's DC input is not the hottest day but the coldest — specifically a cold, clear morning when the sun is strong on modules that have not yet warmed up, which produces the highest voltage the array will ever deliver. Designers therefore calculate the maximum string voltage at the site's record low ambient temperature, using the module's coefficient, rather than at any typical or seasonal average.
- What happens if the maximum voltage is exceeded?
- The inverter is damaged, and it is the kind of failure a warranty excludes because it is a design error rather than a defect. The maximum DC input voltage is an absolute rating of the equipment's internal components, not a threshold above which performance merely degrades, and exceeding it even briefly on one cold morning can destroy the unit. Because the condition arises rarely — a handful of hours a year, at a time when nobody is watching output — an array with one module too many in the string can run apparently normally for months and fail on the first properly cold clear morning. That is why the calculation is done at a record low rather than a design low, why it is a common inspection item, and why substituting a different module into an existing design requires it to be redone.
- Why is the array deliberately larger than the inverter?
- Because an array almost never produces its nameplate output, so an inverter matched to that nameplate is oversized for nearly every hour of its life. Rated output assumes a standard irradiance, a standard cell temperature and a standard spectrum simultaneously — a combination that real conditions rarely deliver, and real cells run hotter and therefore less efficiently than the test condition. Sizing the array above the inverter raises output through the long shoulders of every day and through the many partially cloudy hours, which is where most annual energy actually comes from. The cost is clipping at the top of the sunniest days, and the clipped energy is typically a small share of the annual total. The ratio is an economic optimum rather than a rule, and it has a broad and forgiving peak.
- How much clipping is acceptable?
- As much as the economics justify, which depends on the site and on what the energy is worth. Clipping is not a fault and it does not stress the inverter — the inverter simply holds its maximum output and the array operates away from its maximum power point — so the question is purely whether the energy gained across the rest of the year exceeds the energy lost at the peaks. A site with a peaky irradiance profile, a due-south array at an optimum pitch in a sunny climate, hits its ceiling often and tolerates less oversizing. An east-west array, a shallow pitch, a diffuse northern climate or a site with regular morning cloud has a flatter profile and can be oversized considerably further. Where export is capped by the connection agreement, oversizing is often the only sensible response.
- What is the MPPT window and why does the minimum matter?
- The range of DC voltage within which the inverter can find and hold the array's maximum power point. Above the window it cannot operate and is at risk; below the window's lower bound it cannot track, so the string produces little or nothing regardless of how much sun is falling on it. The minimum string voltage occurs at the maximum expected cell temperature — a hot, still, sunny afternoon, with the modules considerably hotter than the air — which is precisely when the array should be producing most. A string sized only against the cold-weather maximum, with too few modules, can therefore drop out of tracking on exactly the best days of the year, which presents as an intermittent production fault and is a sizing error.
- Does shading change the string design?
- Substantially, because modules in series carry the same current, so the weakest module in a string constrains the whole string. Partial shading on one module — a flue, a parapet, a tree, a neighbouring roof — therefore costs far more than its share of the array's area, and bypass diodes limit the damage without eliminating it. The design responses are to keep shaded and unshaded modules on separate strings and separate trackers so that one string's loss does not propagate, or to use module-level electronics — optimisers or microinverters — which decouple each module and remove the series constraint at a cost per module. Where shading is unavoidable, that choice usually matters more to the annual yield than any refinement of the string length.
- Does a longer string or a shorter one produce more?
- A longer string is generally better within the window, for a reason that has nothing to do with the modules: higher voltage means lower current for the same power, and resistive losses in the DC cabling scale with the square of the current. So a design that runs at the upper end of the permitted string length wastes less in the wiring, needs less copper for the same loss, and generally converts more efficiently at the inverter. That is the pressure pushing string lengths upward, and the cold-weather maximum voltage is exactly what stops them — which is why the two calculations are done together, and why the answer often sits one module short of the theoretical maximum as a margin against an unusually cold morning.
- What changes when modules or inverters are substituted?
- Everything in the string calculation, and it is the most common way a compliant design becomes a non-compliant installation. A substituted module has its own open-circuit voltage and its own temperature coefficient, so the maximum string voltage changes even when the wattage matches; a substituted inverter has its own maximum DC input and its own MPPT window. Either substitution can push a string length that was correct into a length that is not, and the failure will not appear until a cold morning. The same applies to adding modules to an existing array, extending a string, or reconfiguring after a repair. Any of those is a reason to redo the calculation at the site's record low temperature rather than to assume that a like-for-like swap preserves the design.