Treat every PV design question as a set of interacting constraints, not a list of independent rules. Work each scenario end to end: temperature-correct the string voltage, check it against the inverter window, then re-examine how orientation and shading change the power profile before committing to an answer.
String sizing that survives cold mornings, not just STC labels
Size strings using temperature-corrected open-circuit voltage at the site's lowest plausible cell temperature, then confirm the upper and lower voltage limits of the inverter's MPPT window. A string correct at Standard Test Conditions can exceed the inverter maximum in cold weather.
Open-circuit voltage rises as cell temperature falls, because the temperature coefficient is negative. Panel datasheets quote Voc at 25°C cell temperature, so a design that counts panels against the inverter's maximum input voltage using the STC figure has skipped a step. The correction is straightforward: Voc at temperature T equals the STC Voc multiplied by one plus the coefficient times the difference between T and 25°C. Practise doing this arithmetic cleanly, showing the temperature you assumed, because the justification matters as much as the number.
Worked example (illustrative figures only): a module has Voc of 41.5 V at STC and a temperature coefficient of -0.28%/°C. For an assumed cell temperature of -10°C, the adjustment factor is 1 + (-0.0028 × (-10 - 25)) = 1.098, giving about 45.6 V per module. If the inverter's maximum MPPT voltage were 600 V in this hypothetical design, eleven modules per string would give roughly 501 V and pass, while a twelve-module string at about 547 V still passes at -10°C but leaves little margin for colder sites. Compare this against a same-question shortcut that used 41.5 V at STC: twelve modules would look comfortably safe when it is not.
Lower string voltage and the bottom of the MPPT window
The low-voltage end matters as much as the cold-morning maximum. On hot afternoons, maximum power voltage drops, and a short string may fall below the inverter's minimum MPPT voltage, forcing the inverter to stop or operate poorly.
The same temperature coefficient works against you at the other end of the day. On a roof cell temperature of 60°C, a module's maximum power voltage (Vmp) sits well below its STC value, and Vmp is always lower than Voc to begin with. A string sized only to stay under the inverter's ceiling can therefore end up too short: at the hottest part of a summer afternoon its operating voltage may dip under the inverter's minimum MPPT voltage, and generation collapses exactly when the array should be producing hardest.
Continuing the illustrative example: suppose the module Vmp is 34.0 V at STC with the same -0.28%/°C coefficient. At 60°C the factor is 1 + (-0.0028 × 35) ≈ 0.902, so Vmp falls to about 30.7 V. If the inverter's minimum MPPT voltage were 200 V in this hypothetical design, a seven-module string would sit near 215 V and only just clear it, while eight modules give about 245 V of comfortable margin. The lesson to rehearse: every string decision needs both a cold-morning Voc check and a hot-afternoon Vmp check before it is defensible.
Inverter sizing when clipping and cost pull in opposite directions
A larger inverter reduces clipping but costs more and may run less efficiently at part load; a smaller one clips on peak days but is cheaper. The better decision depends on the array's generation profile, especially its orientation spread.
Clipping occurs when the DC array could produce more power than the inverter's AC rating, so output is flattened at the ceiling. The plausible single-rule mistake is to treat any clipping as a design fault and upsize the inverter. In reality the annual energy lost to modest clipping is often small because the excess above the ceiling is shallow and brief: even on clear warm days, a staggered array may reach the ceiling only around midday and only by a small margin. Judging clipping therefore requires thinking about the shape and duration of the flattening, not the peak number alone.
Worked scenario (illustrative): a roof takes 8 kW of modules split evenly east and west, feeding a 6 kW inverter. A peak-style answer picks an 8 kW inverter to eliminate clipping. The better reasoning: with an east-west split, the two sub-arrays peak hours apart, so the combined DC output rarely rises far above 6 kW, and where it does the flattening is shallow and confined to a narrow midday window, costing only a small share of annual energy. The modest inverter also spends more hours operating nearer its efficient rated range. The decision matters because it changes system cost and payback without materially changing yield — and in an exam answer, stating that reasoning explicitly is what demonstrates design judgement.
Reading a shading assessment into an array layout decision
A shading assessment is only useful once it changes the layout: string grouping, module placement or the choice of a design that tolerates partial shading. Describe the shade's timing and cause, then connect it to a concrete configuration change.
Generic advice says avoid shade; exam-style answers need more. Useful questions are: what object casts the shade, at what times of year and day, and does it sweep across the array or sit on one zone? A winter-only shadow from a deciduous tree affects a different part of the yield calculation than a year-round flue shadow crossing the same row each morning. Series-connected modules are particularly sensitive, because the shaded module limits the current of the whole string, so a small sweep of shade can erase a large share of a string's output.
Trace this example: a site report notes a vent pipe shading the lower corner of a proposed array between roughly 8 and 10 am. A weak answer records 'some shading' and proceeds. A stronger answer splits the array so the affected modules are grouped on strings that either face the shade directly or feed an inverter input chosen to tolerate it, keeping unshaded modules on their own full-length strings. State the expected consequence: grouping confines the current limitation to the few affected hours rather than letting one shaded module drag a mixed string down for the whole day. That chain — observation, timing, configuration, consequence — is the reasoning pattern to practise.
Choosing between string configurations: a decision table
Compare candidate configurations against the constraints rather than committing to the first workable option. Build a habit of tabulating each configuration — voltage window, shade behaviour, roof area and inverter inputs — so trade-offs become explicit instead of hidden.
A table forces every candidate configuration through the same checks, which is exactly the discipline the temperature and MPPT-window examples above reward. It also exposes trade-offs that a single-rule answer hides: one option may win on voltage margin but lose on shade resilience, and saying which constraint you prioritised is what turns a correct number into a design decision.
Use the table below as a rehearsal scaffold, not a substitute for calculation. With any real scenario, fill in the voltage figures from the datasheet and the temperature assumptions, then check each row against the specific inverter's window before ranking the options. Practise writing one sentence per row explaining the trade-off in your own words.
| Configuration option | Main advantage | Main risk to check | Best fit |
|---|---|---|---|
| One long series string per roof face | Fewest connections; simple monitoring | Cold-morning Voc may exceed inverter maximum | Unshaded faces with mild climate range |
| Two shorter parallel strings per face | Fits a narrower MPPT window; some redundancy | Needs correct overcurrent protection per string | Long faces that exceed single-string voltage limits |
| East-west faces on separate MPPT inputs | Each face tracked at its own peak; no cross-face penalty | Uses more inverter inputs; mismatch if faces are very unequal | Roofs with two productive orientations |
| East-west modules mixed in one string | String fills voltage window easily; flatter midday profile | Weaker module sets the string current for hours | Small shaded roofs where inputs are scarce |
| Sub-array fed by an inverter tolerant of partial shading | Shade confined to affected modules' contribution | Cost and complexity; verify the topology claim | Sites with fixed, recurring shade zones |
A practice exercise: shaded roof, two faces, one inverter
Run a complete paper design on one invented site: two roof faces, one shade object, one hypothetical inverter with a stated voltage window. Produce a configuration with justification, then score yourself against the rubric below.
Set the exercise: a hypothetical 400 W module with Voc 41.0 V and coefficient -0.30%/°C; a hypothetical inverter with an MPPT window of 200-550 V and two inputs; a north face of twelve possible positions and an east face of eight; and a flue shading the lowest two east-face positions for the first two hours after sunrise. Your task: propose a configuration, show the cold-morning Voc and hot-afternoon Vmp calculations, and explain what the shade does to whichever string it touches.
Self-check rubric — award yourself a point for each: (1) a stated lowest cell temperature and the corrected Voc arithmetic shown; (2) a stated hot cell temperature and the corrected Vmp check against the window floor; (3) shade confined to identifiable modules rather than spread across a mixed string; (4) each string's voltage and count written explicitly; (5) one sentence naming the trade-off you accepted and why. Five points indicates the full reasoning chain; below three, re-run the temperature calculations before changing anything else. When you finish, check your attempt specifically against point 2: if the hot-afternoon Vmp calculation is missing, re-run both temperature corrections together, because pairing the cold ceiling and hot floor checks in one pass is the habit this exercise is designed to build.
An adaptable preparation sequence and readiness checks
Prepare in three passes: concepts first, then calculation drills on invented systems, then full paper designs with justification. Readiness means you can produce a justified configuration for an unfamiliar site without consulting the rules for each step.
Pass one: for each design topic — string sizing, MPPT windows, clipping, shading behaviour, orientation trade-offs — write a two-sentence explanation of what the concept changes about a design and what evidence (datasheet line, site note) feeds it. Pass two: drill the temperature-correction arithmetic until the two checks (cold Voc ceiling, hot Vmp floor) are automatic, using invented modules and inverters with round-number windows. Pass three: build five or six complete paper sites with different constraint combinations, like the exercise above, and score each against a fixed rubric.
Concrete readiness checks before you sit: you can correct Voc and Vmp for temperature without looking up the method; given any hypothetical inverter window and module datasheet, you can propose a string count that passes both checks in under a few minutes; you can describe, in one paragraph each, what clipping, partial shading and orientation spread each do to daily output shape; and for your last practice site you scored at least four of the five rubric points. Treat these as learning milestones, not predictions of any result. Note that accreditation eligibility, exam logistics and current credential details sit with the Clean Energy Council, whose website is the administrative reference for those matters; this guide addresses design reasoning, not enrolment rules.
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.