Knowledge

When a tree branch, chimney, or a single passing cloud casts even a small shadow across your rooftop array, the way you wired your modules determines how much power you actually lose. Understanding series vs parallel solar panels is not an academic exercise; it is the single most important decision you make after selecting your panels, and shading is the reason why. Two identical arrays can sit on the same roof, receive the same shadow at 3 p.m., and produce wildly different output purely because one was wired in series and the other in parallel. In this guide, we explain exactly how partial shading behaves in each topology, what bypass diodes do and do not fix, and how to use a simple 4x220W array as a model for your own installation.

• Why Shading Is the Biggest Factor in Series vs Parallel Solar Panels
• How a Single Shaded Panel Impacts Series vs Parallel Solar Panels
• Bypass Diodes and Their Role in Series vs Parallel Solar Panels Shading
• Series vs Parallel Solar Panels: Performance Under Partial Shading
• Designing Your Array Around Shading in Series vs Parallel Solar Panels
• Series vs Parallel Solar Panels: Real Shading Scenarios Compared
• Final Recommendations for Series vs Parallel Solar Panels in Shaded Areas

Why Shading Is the Biggest Factor in Series vs Parallel Solar Panels

Most installers and DIY builders spend hours comparing panel efficiency, temperature coefficients, and inverter specs, yet the largest single source of real-world energy loss is almost always something far simpler: a shadow. A surprisingly small obstruction can cut output by far more than its physical footprint suggests, because solar cells within a module are connected internally in series, and modules within a string are connected in series again. That nested series structure means a loss in one place tends to propagate through the whole chain. Choosing between series vs parallel solar panels is really a choice about how you want that chain to behave when part of it goes dark.

The core electrical difference is easy to state and hard to overstate. In a series connection, voltage adds while current stays constant, so every module in the string must carry exactly the same current. In a parallel connection, current adds while voltage stays constant, so every module operates at the same voltage and each one contributes whatever current it can. These two rules, applied to a partially shaded array, produce very different outcomes. The series string behaves like a chain that is only as strong as its weakest link, while the parallel bank behaves more like a team in which a single slow runner slows only itself.

Shading matters so much precisely because it is not a binary condition. A panel is rarely fully covered or fully exposed; it usually sits somewhere in between, with one corner under a branch, a stripe of shadow from a vent pipe, or a soft haze from thin cloud. That gradient of illumination translates directly into a gradient of current, and current is the quantity that series wiring cannot tolerate variation in. This is why the shading question should come before the wire-gauge question, before the charge-controller question, and often even before the panel-count question when you are deciding how to configure series vs parallel solar panels.

The Chain Versus the Team: A Simple Analogy

Think of a series string as a single bicycle chain running through four riders in a line. If one rider stops pedaling, the whole chain stalls because the motion of every link depends on every other link. A parallel bank, by contrast, is four riders each turning their own wheel; one tired rider slows only his own contribution while the other three keep producing at full speed. The analogy is not perfect, because real arrays also have diodes and power-point trackers that complicate the picture, but it captures the fundamental difference in how failure and shade propagate through the two topologies.

Why Partial Shade Is Worse Than Full Shade for Series

It sounds counterintuitive, but a half-shaded panel in a series string can sometimes be worse than a fully shaded one, especially on older systems without proper diode protection. A fully shaded panel simply drops its contribution and can be bypassed cleanly, whereas a partially shaded panel can become a bottleneck that constrains current across the entire string while still absorbing power and heating up. This is the mechanism behind hot spots, and it is the reason shading, and the choice between series vs parallel solar panels, is a genuine safety consideration as well as a performance one.

The Role of Orientation and Tilt

Shading sensitivity is also a function of how your panels are oriented and tilted, which is why a fixed rule of thumb can mislead you. A panel facing directly toward the equator with a steep tilt tends to catch shorter, sharper shadows from nearby objects, while a flatter, more horizontal panel on a shallow roof sees longer, softer shadows that sweep across more of the array at once. In both cases the series-versus-parallel calculus changes, because the size, hardness, and duration of the shadow determine whether a bypass diode can do its job. Understanding these interactions is what turns the decision between series vs parallel solar panels from a guess into an engineering choice.

Shading Is a Moving Target

Unlike a fixed design parameter such as panel tilt or orientation, shade moves. It sweeps across the array as the sun tracks across the sky, it changes with the seasons as the sun's altitude shifts, and it grows or shrinks as trees leaf out in spring or drop leaves in autumn. Any series-versus-parallel decision has to be made against this moving target, which is why the most robust answer is often a hybrid arrangement that we will describe later. The key takeaway for now is that shading is not an edge case; it is the normal operating condition for most residential arrays, and it is the single biggest factor that separates a mediocre design from an excellent one.

How a Single Shaded Panel Impacts Series vs Parallel Solar Panels

To see the difference clearly, we need a concrete example. Imagine a modest array built from four 220W panels. A typical 220W module of the 60-cell type produces about 30V at its maximum power point and about 7.3A, which is 220W. Wire all four in series and you get roughly 120V at 7.3A, still 880W. Wire all four in parallel and you get roughly 30V at 29A, again 880W. On a perfectly clear day with no shadows, these two configurations produce essentially identical power, which is why so many beginners assume the wiring choice does not matter. The moment you add shade, the two configurations diverge dramatically.

Now place a single shadow across one panel so that its available current drops from 7.3A to 3A, a fairly typical result when a branch covers a corner of a module. In a series string, the entire string is forced to operate near the current of the weakest panel. The string cannot carry 7.3A anymore because the shaded panel cannot pass 7.3A, so the whole 880W array collapses toward roughly 3A times the full 120V, or about 360W before losses, a staggering 60% reduction caused by shading a single module. This is the single most important fact to internalize about series vs parallel solar panels: one shaded panel in series drags every other panel down with it.

In a parallel bank, the same single shaded panel behaves completely differently. The three unshaded panels continue to operate at 30V and 7.3A each, delivering about 660W between them, while the shaded panel contributes its reduced 3A at 30V, or about 90W, for a total of roughly 750W. That is still a loss, but it is a loss of only about 15% instead of 60%. The parallel bank localizes the damage, and that is the entire argument for parallel wiring in shaded environments in one clean example.

The Weakest-Link Effect in Detail

The series failure mode comes from a basic rule of physics: components wired in series must carry identical current. A solar cell is fundamentally a current source, and its available current is almost perfectly proportional to the amount of light hitting it. Cut the light in half and you roughly cut the current in half. When a single cell in a series chain has its light cut, it cannot source the same current as its neighbors, so the whole chain's current is pulled down toward the weakest cell's level. With four panels in series and 60 cells per panel, you effectively have a chain of 240 cells, and any one dim cell throttles all 240.

Quantifying the Loss With the 4x220W Array

Let us put concrete numbers to the two cases. With all four 220W panels in series and one panel shaded to 3A, a simple string without bypass diodes would collapse toward 3A at roughly 120V, about 360W, or roughly 41% of the 880W nameplate. With the same panels in parallel, the array holds near 750W, or roughly 85% of nameplate. The gap between 41% and 85% is the cost of series wiring in a shaded spot, and it is why the parallel-versus-series question is really a shading question in disguise.

What Happens to Voltage in Each Topology

It is worth being precise about the electrical quantities at play, because they explain why the two topologies fail differently. In a series string, the shaded panel's available current collapses, and because current is shared, the entire string's current collapses with it. The string's voltage, however, remains roughly intact, which is why the array can still reach a high voltage while delivering almost no current, a state that looks healthy on a voltmeter but produces very little power. In a parallel bank, the shaded panel's current collapses on its own, but the bank's voltage stays fixed by the healthy panels, so the loss appears only as reduced total current. Voltmeters and wattmeters tell two different stories here, and many a confused builder has measured a healthy series voltage and assumed all was well while the array produced a fraction of its potential.

Why the Choice Multiplies Across a Larger Array

The single-panel example scales in a predictable but punishing way. In a larger series string of eight or ten panels, a single shaded module throttles an even larger group, so the percentage loss can actually grow as the string lengthens. By contrast, a parallel bank scales more gracefully because every panel keeps contributing independently. This is one reason large ground-mount arrays are so often split into many shorter strings rather than one long string, and it is the conceptual bridge between our small 4-panel example and a full-scale residential or commercial design.

Bypass Diodes and Their Role in Series vs Parallel Solar Panels Shading

No discussion of shading in series vs parallel solar panels is complete without understanding the bypass diode, the small component that is quietly responsible for most of the shading tolerance of modern arrays. Almost every commercially made panel sold today contains several bypass diodes mounted in its junction box, and they exist specifically to stop the weakest-link effect from destroying an entire string. They do not eliminate series shading losses, but they change their character dramatically, and anyone designing around shade needs to know exactly what they can and cannot do.

A bypass diode is wired in parallel with a group of cells, typically a substring of 20 cells in a 60-cell panel, so that a 60-cell panel is divided into three independently bypassable sections. In normal operation the diode is reverse-biased and does nothing. But if one section of the panel is shaded and can no longer source the string's current, the voltage across that section reverses, and the diode becomes forward-biased, effectively creating a short detour around the dark cells. The shaded section is taken out of the circuit, and the string's current is no longer throttled by it. This is why modern series strings lose roughly one-third of a panel's output when a single substring is shaded, rather than the entire string's output.

The crucial nuance is that bypassing a substring is not free. When a diode conducts around a shaded section, that section contributes no voltage, so the panel loses that substring's voltage share, typically about one-third of the panel's voltage, plus the diode itself drops a small amount, around 0.5V. On an unshaded panel, the current still flows through the panel's remaining two active substrings and their cells still generate, but the shaded third is simply skipped. The net result is that a shaded panel in a series string produces perhaps two-thirds of its voltage at full current, instead of full voltage at a fraction of the current. The loss is far smaller than the catastrophic no-diode case, but it is still a real loss.

What Bypass Diodes Fix and What They Do Not

Bypass diodes fix the worst failure mode, the total collapse of a series string when one cell goes dark, and they prevent the hot-spot damage that can physically destroy a panel. What they do not do is recover the energy from the shaded area, and they do not make a shaded series string perform as well as a parallel bank. Even with diodes working perfectly, the shaded substring contributes nothing, and the string as a whole loses that fraction of output. The parallel bank, by contrast, does not need to bypass anything; the shaded panel simply contributes less current while the others keep working.

How MPPT Charge Controllers Interact With Diodes

There is another subtlety that matters in the real world. A maximum power point tracking (MPPT) controller scans the voltage-current curve of the array to find the best operating point. When a diode bypasses a shaded substring, the array's power-versus-voltage curve develops multiple peaks, one near the voltage of the full string and another at a lower voltage corresponding to a bypassed panel. A good MPPT tracker can find the lower peak and recover substantial power, which is why bypass diodes and MPPT work so well together. A basic PWM controller, which simply clamps the array to battery voltage, cannot exploit this and will fare worse in shade.

How Many Diodes Do Panels Actually Have

The number of bypass diodes inside a panel varies by design, and that number sets the granularity of shading protection. A 60-cell panel is conventionally divided into three substrings of 20 cells, each protected by its own diode, while a 72-cell panel is often split the same way into three groups of 24. Larger commercial modules may use more substrings, and a few budget panels historically shipped with fewer, which leaves more cells exposed to a single bypass event. More substrings means finer-grained protection, so a shadow that would have darkened a 24-cell group on a 72-cell panel only darkens a 20-cell group on a 60-cell panel, preserving more voltage. When you are shopping for panels destined for a shaded roof, the internal diode layout is a specification worth checking, because it directly affects how gracefully the panel behaves in a series string.

Diodes Change the Math, Not the Conclusion

With bypass diodes in the picture, our 4x220W series string with one panel shaded to 3A no longer collapses to 360W. Instead, the shaded panel loses one of its three substrings, so the string operates at roughly 7.3A but at a reduced voltage, losing about a third of one panel's 30V, or about 10V, dropping the string from 120V to about 110V. Output lands near 110V times 7.3A, roughly 800W, a loss of only about 10%. That is far better than the no-diode catastrophe, but it is still a larger loss than the parallel bank's 750W... and here is the surprise: with diodes and MPPT, a series string with a lightly shaded panel can actually approach the parallel bank, which is why the answer to the series-versus-parallel question depends on how the shade falls, not just on whether shade exists.

Series vs Parallel Solar Panels: Performance Under Partial Shading

View more>>The Right Wire Gauge for Series-Parallel Solar Arrays: Avoid Voltage Drop and Fire Risks

Partial shading is the most common shading condition a rooftop ever sees, and it is also the condition where the difference between series and parallel wiring is most nuanced. "Partial" can mean one panel partially covered, one panel out of several fully covered, or a soft diffuse reduction in light across the whole array from thin cloud. Each of these behaves differently in a series string than in a parallel bank, and each deserves its own treatment if you want to predict real performance rather than guess.

The first and most important distinction is between uniform and non-uniform shade. Uniform shade, the kind you get from a high, thin cloud layer, reduces every panel's illumination by roughly the same fraction. Because every panel is affected equally, the weakest-link problem never really materializes in a series string, and both topologies lose output in rough proportion to the light reduction. Non-uniform shade, the kind you get from a tree branch, a chimney, or a vent pipe, hits one panel or one section of a panel much harder than the rest, and that is where series and parallel part ways.

A second distinction is between shade that spans an entire substring and shade that only partially covers one. Bypass diodes only help when a whole substring is pulled down far enough to reverse-bias its diode; a thin stripe of shadow that crosses a single cell without fully darkening it may not trigger the diode at all, leaving the string throttled at the reduced current. This is why even a small, sharp shadow in the wrong place can punch far above its weight in a series string, and why the parallel bank's relative immunity to such details is so valuable in cluttered rooftop environments.

Uniform Versus Non-Uniform Shade

Under uniform shade, series vs parallel solar panels perform almost identically. If a cloud reduces light to 60% across the whole array, both configurations drop to roughly 60% of output, because in the series string every panel is throttled together and in the parallel bank every panel is reduced together. The series string retains its efficiency advantage in wiring, since its higher voltage keeps current and resistive losses low. This is why the series-versus-parallel decision should not be driven by the mere presence of clouds, which are uniform, but by the presence of localized shadows, which are not.

Localized Shadow Across One Panel

When a hard shadow falls on a single panel, the two topologies diverge sharply, as our 4x220W example showed. The parallel bank sacrifices only the shaded panel's own output, while the series string, even with bypass diodes, sacrifices the shaded substring and possibly the shaded panel's entire voltage contribution for that portion. The exact numbers depend on the MPPT controller's ability to find the lower power peak, but the structural advantage of parallel is consistent: in parallel, damage stays local; in series, damage propagates at least one substring deep.

Microinverters and Power Optimizers: The Third Option

Before settling on series or parallel wiring, it is worth acknowledging that the industry has developed a third approach that sidesteps the dilemma almost entirely. A microinverter attaches to each individual panel and performs maximum power point tracking at the module level, so every panel operates independently and a shaded panel simply produces less while its neighbors produce full power. A DC optimizer does something similar, keeping each panel at its own best operating point before feeding a central inverter. Both technologies replicate the shade tolerance of a parallel bank while preserving the high-voltage, low-current wiring advantages of a series string, effectively giving you the best of both worlds at a higher upfront cost. For a heavily shaded roof where neither pure series nor pure parallel is ideal, microinverters or optimizers are often the cleanest engineering answer, even though they cost more than a conventional string setup.

Partial Coverage Within a Single Panel

The most insidious case is a shadow that covers only part of a single panel without fully darkening any one substring. In a parallel bank, this simply reduces that one panel's current by roughly the covered fraction, and the array barely notices. In a series string, the same shadow forces the whole string down to the reduced current unless a substring happens to reverse-bias and trip its diode, and even then the string loses that substring's voltage. Repeated across a day as the shadow moves, these small differences compound into a meaningful annual energy gap, and they are a primary reason parallel or hybrid configurations are preferred on heavily obstructed roofs.

Designing Your Array Around Shading in Series vs Parallel Solar Panels

Once you accept that shading is a design input rather than a nuisance to be ignored, the wiring decision becomes a genuine design task with a clear method. The goal is to match your array topology to the specific way shade falls on your site, and in most cases that means abandoning the idea of a pure series string or a pure parallel bank in favor of a deliberate hybrid. Understanding how series vs parallel solar panels respond to shade lets you arrange modules so that any single shadow touches as little of the array as possible.

The most powerful tool you have is the series-parallel hybrid, usually written as something like 2S2P, meaning two panels in series repeated in two parallel strings. A 2S2P arrangement of our four 220W panels produces about 60V at 14.6A, a voltage high enough to keep wiring losses modest and to feed an MPPT charge controller comfortably, while still limiting the blast radius of any single shaded panel to one string of two panels rather than the whole four-panel chain. This configuration captures much of the voltage advantage of series wiring and much of the shade tolerance of parallel wiring at the same time.

Site analysis is the other half of the design. Before you decide on a topology, you should map the shade that actually crosses your roof, ideally by observing the site across a full day in multiple seasons, or by using a solar shading analysis tool. Identify the sources: trees, chimneys, neighboring buildings, vent pipes, satellite dishes, and even your own adjacent panel rows. The question is not just "is there shade?" but "where and when does shade fall, and can I route the wiring so that a single shadow lands on one string instead of one cell of a long series chain?"

When to Choose Series

Series wiring remains the right choice when your site has little localized shade and you want the simplest, cheapest installation. Higher string voltage means lower current, which means thinner and cheaper wire, lower resistive losses, and longer home runs without voltage drop. If your array sits on an open ground mount or a south-facing roof with no nearby obstructions, a full series string feeding a high-voltage MPPT controller is clean, efficient, and inexpensive. The 4x220W panels in series produce about 120V, which is well within the input range of most MPPT controllers and keeps the array wiring simple.

When to Choose Parallel or a Hybrid

Parallel, or more realistically a hybrid like 2S2P, becomes the better choice the moment a hard localized shadow can reach your modules. If a chimney shades one panel at 9 a.m. and a tree shades another at 5 p.m., you want those two shaded panels in different strings so that neither shadow takes down the whole array at once. Grouping the panels that share a common obstruction into the same string is a clever trick: if two panels are always shaded together by the same tree, putting them in series means one bypass event handles both, while the clean panels run in a separate parallel string untouched.

Accounting for Temperature and Voltage Limits

Any shading-aware design must also respect the electrical limits of the equipment it feeds, because the wiring decision affects voltage and current in ways that shading can exaggerate. A long series string of cold panels on a bright winter morning can produce a surprisingly high open-circuit voltage, and if that voltage exceeds the input rating of the charge controller or inverter, the equipment can be damaged. A large parallel bank, by contrast, runs a high current that demands thick conductors and can exceed the current rating of a controller or combiner. These limits pull the design back toward a balanced hybrid: enough series to keep voltage high but not so much that it approaches the controller's ceiling, and enough parallel branches to keep current manageable. Shading is not the only constraint on series vs parallel solar panels, but it is the one that most often tips a design that would otherwise sit safely within limits into one that must be reconsidered.

Placing Shade-Prone Panels Deliberately

Layout matters as much as topology. Wherever possible, put the panels most likely to be shaded at the bottom of a string or in their own string, and keep the sunniest panels together so they never share a series path with a chronically shaded neighbor. In a 2S2P setup, this might mean pairing the two morning-shaded panels in one series pair and the two afternoon-shaded panels in another. The result is that at any given moment, only one parallel branch is impaired, and the other branch keeps producing at full power, which is precisely the behavior you want from series vs parallel solar panels designed around real-world shade.

Series vs Parallel Solar Panels: Real Shading Scenarios Compared

Abstract numbers are useful, but the decision finally gets made on a real roof with real obstructions. Three shading scenarios dominate residential installations: the tree, the chimney, and the passing cloud. Each has a distinct shadow shape, duration, and intensity, and each rewards a different wiring choice. Walking through them with our 4x220W array makes the tradeoffs concrete and gives you a template for evaluating your own site.

The tree is the classic hard-shadow source. Its shadow is dense and sharply defined, often covering a large fraction of one or two panels for an hour or two as the sun moves behind the canopy. Because the shadow is hard and localized, it triggers bypass diodes cleanly and throttles the covered panels to near zero. In a series string, a tree shadow over one panel is exactly the worst case we described, and even with diodes it costs the array a substring of voltage; in a parallel or hybrid bank, it costs only the covered panel's own output. For a heavily treed site, the tree scenario alone usually tips the balance toward hybrid wiring.

The chimney is a different beast: a small, extremely hard shadow that sits near the top of a panel and moves only slowly across it. Because it is small, it often fails to fully darken a substring, meaning a bypass diode may not even activate, and the panel simply contributes reduced current. In a series string, a stubborn little chimney shadow can throttle the entire string for a long stretch, which is frustrating precisely because the shadow looks too small to matter. In a parallel bank, the same shadow barely registers. This is the scenario that most often surprises first-time builders and convinces them to avoid a pure series string.

The cloud is the soft-shadow scenario, and it is the one where series wiring actually holds its own. A thin or broken cloud layer dims the whole array uniformly, so every panel loses current together and the series string's weakest-link problem never appears. Since the cloud does not localize, parallel wiring confers no special advantage, and the series string's lower current and cheaper wiring win on simplicity. The lesson is that not all shade is the same, and the series-versus-parallel answer genuinely changes with the character of the obstruction.

Scenario 1: The Tree Line

Picture a row of tall pines on the east side of a south-facing roof. From 8 a.m. to 10 a.m., the shadow of the trees sweeps from left to right across the four 220W panels. In a series string, at any moment one panel is heavily shaded, triggering a bypass and cutting the string's output by roughly a third of a panel's voltage; averaged over those two hours, the array loses a meaningful chunk of morning production. In a 2S2P hybrid with the left two panels in one string and the right two in another, the shadow spends its time on one string and then the other, so the unshaded string keeps producing near full power the whole time. The hybrid recovers most of the morning energy that the series string gives away.

Scenario 2: The Chimney or Vent Pipe

Now add a chimney near the peak of the roof that casts a shadow the size of a dinner plate. Because it is small and hard, it may sit inside a single substring without ever tripping the bypass diode, so in a series string it throttles the entire string to a reduced current for as long as it lingers, sometimes an hour or more. In a parallel bank, the affected panel loses a few percent of its own output and the array hardly notices. This is the scenario where series wiring looks worst relative to parallel, because the loss is disproportionate to the tiny size of the shadow, and it is the strongest practical argument for parallel or hybrid wiring on cluttered roofs.

Scenario 4: Seasonal Tree Growth

A fourth, often overlooked scenario is the tree that is fine in winter and a problem in summer. A deciduous tree that is bare during the low-sun winter months can leaf out into a full canopy by late spring, suddenly casting a hard, moving shadow across panels that enjoyed six months of unobstructed sun. If the array was wired as a single series string and performed beautifully all winter, the same string can begin collapsing every afternoon once the leaves appear, and the owner may not even connect the dots until the summer electric bill arrives. This is a strong argument for designing the array around the worst-case seasonal shading rather than the snapshot of the moment, and it is another reason a hybrid topology, which degrades gracefully as the canopy fills in, is the more robust long-term choice for series vs parallel solar panels.

Scenario 3: Passing Clouds

Finally, watch a partly cloudy afternoon. Clouds roll over the whole array at once, dimming all four panels uniformly. During these periods, series and parallel configurations track each other almost exactly, both dropping to some fraction of nameplate and recovering together. The series string's higher voltage keeps its resistive losses lower, so if anything it slightly outperforms a pure parallel bank under uniform cloud. The takeaway is to reserve your worry for localized shadows, and to recognize that the cloud scenario is the one place where a pure series string loses its handicap.

Final Recommendations for Series vs Parallel Solar Panels in Shaded Areas

View more>>How to Increase Solar Self-Consumption Without Buying More Panels

By now the picture should be clear: the choice between wiring configurations is, in any site with real obstructions, a choice about how to manage shade. A pure series string maximizes voltage and minimizes wiring cost, but it concentrates risk, allowing a single shadow to throttle the entire chain. A pure parallel bank keeps every panel independent and shrugs off localized shade, but it runs at low voltage and high current, demanding thick, expensive cable and short runs. The sweet spot for almost every shaded residential array is the series-parallel hybrid, which pairs the voltage advantages of series wiring with the shade tolerance of parallel wiring.

For our 4x220W example, a 2S2P configuration is the default recommendation. It produces about 60V and 14.6A, which is high enough to keep wiring losses low and to make efficient use of a standard MPPT charge controller, while limiting the blast radius of any single shaded panel to one two-panel string. If your site is open and unobstructed, a full series string at 120V is perfectly fine and slightly cheaper. If your site is so cluttered that nearly every panel catches a hard shadow at some point, a full parallel bank at 30V is defensible, though you should pair it with appropriately sized conductors and a controller that can handle 29A or more.

Whatever topology you choose, make the decision deliberately and verify it against your actual site. Map the shade, group panels by their shared obstructions, confirm every panel has functioning bypass diodes, and use an MPPT controller so the array can find the best power point when part of it is shaded. Shading will cost you energy no matter what, but with the right wiring it does not have to cost you the whole array. Before you run the first wire, take stock of every obstruction that touches your roof across the year, check that your panels carry proper bypass diodes, and pair the array with an MPPT controller that can chase the best power point when part of the string is dark. If in doubt, split the difference with a 2S2P hybrid, which is forgiving enough to tolerate a chimney here and a tree there while still running at a voltage your controller likes. That is the real lesson of series vs parallel solar panels: a little thought about where shadows fall, applied before the first wire is pulled, is worth more than any premium panel you could buy.