Skip to content
WattCurious

How Partial Shade Affects Solar Panels (Bypass Diodes Explained)

Updated 8 October 20267 min readSolar Energy

Cells in a module are wired in series, which means the same current passes through every one of them. A cell in shadow cannot produce that current, and because it cannot, it limits what all the others may deliver. A bypass diode exists to contain the damage — not to prevent it — and understanding where those diodes sit is what turns shading from a mystery into something you can predict and design around.

Key takeaways

  • A series string carries one current, so the least illuminated cell sets the limit for every cell in that string.
  • A bypass diode does not restore the shaded cell's output — it removes its whole sub-string from the circuit so the rest of the module can keep working.
  • Shading makes the module's power curve grow several local peaks, and an inverter tracking maximum power can settle on the wrong one.
  • Sub-strings run the length of a module, so a shadow crossing them is far more costly than one running along a single sub-string.
  • A shaded cell in a string that is still delivering current is reverse-biased, dissipating power instead of producing it, which is where hotspots come from.
On this page
  1. One current, sixty cells
  2. What a bypass diode actually does
  3. The power curve grows extra peaks
  4. Which way the shadow runs
  5. Reverse bias, and where heat comes from

One current, sixty cells

A module is not sixty small generators feeding a common bus. It is a chain. The cells are wired in series so their voltages add, and a series connection has one inescapable property: the same current flows through every element.

A series string of cells drawn as a chain, all carrying the same current. Most cells are in full sun and could each deliver the full current. One is shaded and can only deliver a fraction of it. Arrows show that the current through the whole string is held down to what the shaded cell can pass, so the fully lit cells are producing far below their capability. A note compares this to a chain of pipes where the narrowest section sets the flow.A series string of cells drawn as a chain, all carrying the same current. Most cells are in full sun and could each deliver the full current. One is shaded and can only deliver a fraction of it. Arrows show that the current through the whole string is held down to what the shaded cell can pass, so the fully lit cells are producing far below their capability. A note compares this to a chain of pipes where the narrowest section sets the flow.
One current, set by the least illuminated cell. The other cells are not broken and not shaded — they are simply not allowed to deliver what they could.

So a cell in shadow does not reduce output by its own share. It reduces the current available to every cell wired with it. Shade a twentieth of a module's area and you can lose vastly more than a twentieth of its output, which is the single most surprising thing about photovoltaics and the reason shading is treated so seriously at design time. Cloud is the opposite case: it dims every cell evenly, so the series current falls in step with the light and the loss stays proportional — the behaviour behind how panels perform on cloudy days.

This is the same mechanism that makes a single opaque patch — a leaf, a bird dropping — so much worse than an even film of dust, the contrast drawn out in cleaning panels without damaging them.

What a bypass diode actually does

The fix is not to rescue the shaded cell. It is to take it out of the circuit along with its neighbours.

A module drawn as three sub-strings of cells, each with a bypass diode connected across it. In normal operation all three diodes are reverse-biased and do nothing, and current flows through all the cells. When one cell in the middle sub-string is shaded, the voltage across that sub-string reverses, its diode becomes forward-biased and conducts, and the current bypasses the entire middle sub-string. The two unshaded sub-strings continue to deliver normally, and the module's voltage drops by a third.A module drawn as three sub-strings of cells, each with a bypass diode connected across it. In normal operation all three diodes are reverse-biased and do nothing, and current flows through all the cells. When one cell in the middle sub-string is shaded, the voltage across that sub-string reverses, its diode becomes forward-biased and conducts, and the current bypasses the entire middle sub-string. The two unshaded sub-strings continue to deliver normally, and the module's voltage drops by a third.
The diode conducts and the current goes around. The shaded sub-string stops contributing entirely, so the module loses roughly a third of its voltage and keeps its current.

A module is divided into sub-strings, commonly three, each with a bypass diode wired across it. In full sun the diodes are reverse-biased and invisible. When a cell is shaded enough that its sub-string can no longer pass the string current, the voltage across that sub-string reverses, its diode becomes forward-biased and conducts, and the current takes the diode path instead.

The result is deliberately blunt: the shaded sub-string contributes nothing, and the module continues at roughly two thirds of its voltage. That is a large loss for one shaded cell — and far smaller than the alternative, which is the whole module, and behind it the whole string, dragged down to what the shaded cell can pass.

The power curve grows extra peaks

Bypass diodes solve the current problem and create a subtler one. A module with some sub-strings active and some bypassed no longer has a single, smooth power curve.

Two charts sharing a voltage axis. The upper shows current against voltage: unshaded, a single smooth curve with one knee; partially shaded, a stepped curve with distinct shoulders where bypass diodes begin conducting. The lower shows the corresponding power curve: unshaded, a single clear maximum; partially shaded, two separate local maxima of different heights separated by a dip. A tracker settled on the smaller peak is marked, alongside the larger global peak it has not found.Two charts sharing a voltage axis. The upper shows current against voltage: unshaded, a single smooth curve with one knee; partially shaded, a stepped curve with distinct shoulders where bypass diodes begin conducting. The lower shows the corresponding power curve: unshaded, a single clear maximum; partially shaded, two separate local maxima of different heights separated by a dip. A tracker settled on the smaller peak is marked, alongside the larger global peak it has not found.
Illustrative shapes. Each bypassed sub-string puts a step in the curve, and every step creates another local maximum for the tracker to mistake for the best one.

An inverter finds the operating point by maximum power point tracking — nudging the voltage and watching whether power rises. That works perfectly on a curve with one peak. On a shaded curve with several, a local search can settle on a smaller peak and stay there, delivering less than the array is capable of.

Inverters handle this by sweeping the full voltage range periodically to locate the global maximum. It is why output sometimes steps up by itself after a brief dip, and why a system that appears to be underperforming under scattered shade may recover without intervention — one of the patterns worth recognising when a system produces less than expected.

Which way the shadow runs

Because sub-strings occupy specific parts of the module, shading is directional. The same shadow, on the same module, costs completely different amounts depending on how it crosses.

A module shown twice with its three sub-strings marked as bands running the length of the module. In the first, a shadow lies along one band and touches only that sub-string, so one bypass diode conducts and two thirds of the module keeps working. In the second, a shadow of the same area crosses all three bands, so a cell in each sub-string is shaded, all three diodes conduct, and the module delivers nothing at all.A module shown twice with its three sub-strings marked as bands running the length of the module. In the first, a shadow lies along one band and touches only that sub-string, so one bypass diode conducts and two thirds of the module keeps working. In the second, a shadow of the same area crosses all three bands, so a cell in each sub-string is shaded, all three diodes conduct, and the module delivers nothing at all.
Identical shadow area, completely different outcome. A shadow along one sub-string costs a third; the same shadow across all three costs everything.

Sub-strings run the length of a module, so a shadow lying along one band touches one sub-string and costs one third. The same shadow lying across the bands clips a cell in every sub-string, every diode conducts, and the module produces nothing.

This is what turns mounting orientation into an engineering decision rather than an aesthetic one. A low shadow advancing across a row of modules — a parapet, a fence, the row in front, whose shadow lengthens with steeper tilt and tighter row spacing — is best met with the sub-strings parallel to it. A shadow that sweeps through the day, from a pole or a chimney, gives a different answer. The rule is not "portrait" or "landscape"; it is "know which way the shadow travels and keep it inside as few sub-strings as you can."

Reverse bias, and where heat comes from

There is a window before the diode conducts where the shaded cell is in genuine difficulty.

What happens to a shaded cell as the shading deepens
ShadingThe cellThe sub-stringThe module
Light, diffuseProduces a little lessCurrent drops slightlySmall, proportional loss
ModerateBecomes a load, not a sourceVoltage fallsLoss larger than the shaded area
Enough to reverse the sub-stringReverse-biased, dissipatingDiode conducts, sub-string outLoses about a third
Hard shade, one sub-stringProtected by the diodeBypassed entirelyTwo thirds still working
Hard shade across all sub-stringsEvery one bypassedAll bypassedNo output

A description of behaviour, not thresholds. Where each transition occurs depends on the cell, the module's construction and how hard the shading is.

A cross-section through two cells in a series string. The lit cell drives current through the circuit. The shaded cell cannot pass that current, so the voltage of all the other cells in the string is applied across it in reverse. Arrows show power flowing into the shaded cell rather than out of it, concentrated in a small area, with heat spreading into the encapsulant around it. Alongside, the bypass diode is shown conducting and the reverse voltage collapsing once it does.A cross-section through two cells in a series string. The lit cell drives current through the circuit. The shaded cell cannot pass that current, so the voltage of all the other cells in the string is applied across it in reverse. Arrows show power flowing into the shaded cell rather than out of it, concentrated in a small area, with heat spreading into the encapsulant around it. Alongside, the bypass diode is shown conducting and the reverse voltage collapsing once it does.
Before the diode conducts, the shaded cell is a load with the rest of the string's voltage across it. The power has to go somewhere, and it goes into a very small area.

The row that matters is the third. A cell that cannot produce the current being pushed through it has the other cells' voltage applied backwards across it, and it dissipates that power rather than generating it. Concentrated in one cell, that becomes a hotspot — localised heating that stresses the encapsulant and, over time, the cell itself. It is a different thing from the uniform warming that costs a module voltage in hot weather: that affects every cell equally, while a hotspot is one cell absorbing power the others produce.

Bypass diodes exist largely to cut this short, and modules are qualified against hotspot endurance before they ship. But the underlying point stands: hard, persistent shading on one spot is worth removing, and the invisible cell damage it can leave behind is the same kind that makes hail impacts so hard to diagnose.

What actually helps, in order: remove the shadow if it can be removed; arrange the array so what remains crosses as few sub-strings as possible; and where neither is possible, move the tracking down to the module, which is the case for module-level electronics rather than an argument against string inverters in general.

Frequently asked questions

Does shading one panel really affect the whole string?

It can, and it is the reason shading gets so much attention. Modules in a string share one current just as cells within a module do, so a module producing less limits the string. Bypass diodes cap the damage at sub-string granularity, but the loss is still far larger than the shaded area alone would suggest.

How many bypass diodes does a module have?

Commonly three, each protecting one sub-string of cells, though the number varies with the module's construction. That is why shading a single cell often costs about a third of a module: the diode protecting its sub-string conducts, and the entire third stops contributing.

Why does my inverter sometimes seem to settle on lower output than it could?

Partial shading gives the power curve more than one peak, and a tracking algorithm looking locally can settle on a smaller one. Most inverters periodically sweep the whole voltage range to find the global peak, which is why output occasionally steps up on its own after a pause.

Is it better to mount modules portrait or landscape under shading?

It depends entirely on which way the shadow travels. Sub-strings run the length of the module, so the good orientation is the one where a moving shadow stays within as few sub-strings as possible. A low shadow creeping across a row favours one orientation; a pole's shadow sweeping through the day may favour the other.

Can shading damage a panel?

It can, through reverse bias. A shaded cell in a string still carrying current has the other cells' voltage pushed backwards across it, and it dissipates that power as heat. Bypass diodes exist largely to limit this, and modules are qualified against hotspot endurance — but persistent hard shading on one spot is still worth removing.

Sources

Named organisations whose published material underpins this article. Where no link is given, the source is named rather than linked.

Editorial Team

Research, drafting and review

Articles are drafted from primary engineering and physics references with AI-assisted tools, then reviewed and fact-checked line by line by a human editor before publication. We publish explanations, not recommendations: no products, no pricing, no country-specific rules, and no invented author personas.

Last reviewed 8 October 2026. How we research and review