What Makes Cell Voltage Spread Grow as a Solar Module Ages

Cell voltage spread grows when one or more cells inside a module start behaving differently from the rest. The reason may be a crack, extra electrical resistance, leakage, or repeated reverse-bias stress. These problems do not affect every cell at the same speed, so the voltage gap between cells becomes wider over time.

For operators and asset owners, the useful point is not the age printed on the module. What matters is whether the voltage difference is getting larger under similar current, irradiance, and temperature. A 10 mΩ connection defect, for example, creates only 0.05 V of extra drop at 5 A but about 0.175 V at 17.5 A. One weak cell can therefore become a serious local problem while the total module voltage still looks almost normal.

Uneven Aging, Not Age Alone

A module can lose power without developing a large voltage spread. If most cells age at about the same rate, their voltages can still remain close to one another.

The problem starts when one cell, or a small group of cells, changes faster than the rest.

A solar cell does not behave like a battery with one fixed voltage. Its voltage moves with the operating point. Photocurrent, diode behavior, series resistance, shunt resistance, and temperature all affect that operating point.^[1]

Series Current Forces Weak Cells to Move in Voltage

Cells connected in the same series path have to carry the same current. That is where a weak cell starts to matter.

Take a series path with 20 cells. If nineteen cells can comfortably carry 10 A but one damaged cell cannot, that weak cell does not get to carry only 7 A while the others stay at 10 A.

Instead, its voltage shifts.

• mild mismatch: positive voltage, but lower than neighboring cells;
• strong mismatch: voltage approaches 0 V;
• severe mismatch: voltage becomes negative and the cell enters reverse bias.

Sandia describes the same basic effect in PV mismatch analysis: devices connected in series must carry the same current even when their electrical characteristics are different.^[2]

One detail matters for newer modules. Half-cut designs can contain both series and parallel branches. The same-current rule applies to cells or cell sections in the same series path, not automatically to every cell in the entire module.

A 0.01 V Spread Can Become 0.24 V

A simple example makes the scale easier to see. Assume five cells are working in the same series path at the same current, with similar irradiance and temperature.

Before the fault:

0.58 V − 0.57 V = 0.01 V spread

After Cell 4 weakens:

0.58 V − 0.34 V = 0.24 V spread

In this example, the spread becomes 24 times larger.

That does not make 0.24 V a universal failure limit. It also does not mean every aging module will see its spread increase 24 times. The example only shows how far one weak cell can move away from the other cells.

Cell 4 has not permanently become a “0.34 V cell” either. Change the current or temperature and its operating voltage may change as well.

Cracks Matter When They Cut Current Paths

Not every crack is equally serious. The key question is whether the crack actually interrupts the path that carries current out of the cell.

The electrical chain is straightforward:

crack → electrical isolation → less active area → lower current capability → larger voltage difference

A long visible crack may have little immediate electrical effect if current can still pass through working collection paths. A shorter crack may cause more trouble if it disconnects a large part of the cell from the fingers, busbars, or wire contacts.

That is why crack length alone is not a reliable measure of electrical severity.

IEA PVPS identifies cracking as an important module failure mode and notes that modern multi-wire interconnection can reduce the power-loss effect of some crack patterns by providing more current paths.^[3]

The same idea appears in modern MBB and 0BB interconnection structures. More collection points or wires give current more possible routes around certain damaged areas.

10 mΩ Can Become Several Watts of Heat

Aging connections often fail gradually rather than all at once. A ribbon, solder joint, or contact may still conduct electricity but with more resistance than before.

The extra voltage loss is approximately:

ΔV ≈ I × ΔR

The heat created in that resistance is:

P = I²R

The numbers below all use exactly the same added resistance: 10 mΩ, or 0.01 Ω.

This is why a weak connection can look harmless in low light and become much easier to detect when the module is carrying more current.

A current Tongwei TNC-G12 66 module family operates at roughly 17.35–17.63 A at maximum power, depending on model. At about 17.5 A, just 10 mΩ of added resistance creates roughly 0.175 V of extra drop and just over 3 W of local heat.

Once heating starts, the connection can deteriorate faster:

weak connection → more resistance → more heat → faster connection damage

Corrosion Raises Resistance; Shunts Increase Leakage

Corrosion and shunting can both make one cell fall behind its neighbors, but they do not work in the same way.

Corrosion raises resistance in parts that are supposed to carry useful current, including:

• grid fingers;
• wire or busbar contacts;
• ribbons;
• soldered joints.

More resistance means more voltage is lost when current flows.

Shunting creates an unwanted path inside the cell. Some of the generated current takes that shortcut rather than going through the normal external circuit.

A cell with strong shunting can struggle to maintain the same voltage as nearby cells. The effect can become more noticeable under lower irradiance because the useful generated current is already smaller.

Moisture can make several of these problems worse. It may contribute to corrosion, interface damage, leakage, or delamination. Module-level insulation leakage and cell-level shunt resistance are not the same fault, however, and should not be treated as one issue.

IEA PVPS treats corrosion, encapsulation damage, cell cracking, interconnection faults, PID, and related electrical failures as separate mechanisms that can interact during module life.^[4]

Shade Can Drive One Cell Below 0 V

Shade becomes an electrical problem because it cuts the photocurrent available from the affected cell.

If the rest of the series path is still strongly illuminated, the circuit may continue demanding more current than the shaded cell can comfortably supply. Its voltage falls.

less light → less current capability → lower voltage → possible reverse bias

Once the cell voltage drops below zero, the cell can begin absorbing electrical power instead of adding useful positive voltage.

The power dissipated is approximately:

P ≈ I × |V_reverse|

These are calculation examples, not typical reverse-voltage values. Actual reverse behavior varies with cell design, defect type, shade pattern, and cell condition.

The useful point is the scale. Tens of watts can be dissipated in a cell-sized area even when the module as a whole is still producing useful power.

IEA PVPS documents reverse-bias and hot-cell behavior in shaded and cracked cells and notes important differences between modern cell technologies.^[5]

Hot Connections and Hot Cells Are Different Faults

Two hot areas can look similar on a thermal image and still come from different electrical problems.

A hot region tells you where electrical energy is being lost. It does not identify the cause by itself.

That is why a thermal image is more useful when the same pattern appears repeatedly under suitable irradiance and operating conditions. IEC TS 62446-3 sets requirements for outdoor PV thermography, including test conditions, equipment, inspection procedure, and reporting.^[6]

Bypass Diodes Protect Groups, Not Individual Cells

Bypass protection works at the substring level, not one diode per cell.

A current Tongwei TNC-G12 66 configuration, for example, uses 132 cell sections and 3 bypass diodes.

The internal split layout is more complex than simply dividing 132 by 3, but the practical point is clear:

many cell sections share only a few bypass devices.

One weak cell can therefore develop a noticeable local voltage mismatch before the whole protected section reaches the condition that turns a bypass diode on.

When a bypass path does become active, total module voltage can change much more sharply than it would from a slow loss of a few tenths of a volt in one cell.

Bypass diodes can also fail, and defective bypass diodes are a recognized module failure mode.^[7]

Temperature Can Look Like Aging

A lower voltage reading does not always mean the cell has degraded. Sometimes it is simply hotter.

For crystalline-silicon PV, voltage normally falls as cell temperature rises.

Standard Test Conditions use:

• 1,000 W/m² irradiance;
• 25°C cell temperature.

Real modules often run much hotter outdoors.

As a product example, Tongwei's current TNC module data lists a module open-circuit-voltage temperature coefficient of about -0.24%/°C.

A 20°C difference is already large enough to create a meaningful voltage change. If temperature is ignored, that difference can easily be mistaken for aging.

This example uses module Voc and the stated coefficient. Individual operating cell voltage does not necessarily change by exactly the same percentage.

Thermal Cycling Turns Small Defects Into Larger Ones

The module is not mechanically static during its life. Every day it heats up and cools down.

Silicon, copper, solder, glass, polymers, and aluminum expand by different amounts. That repeated movement puts stress on weak areas that already exist.

• existing cracks;
• weak solder joints;
• ribbon damage;
• delamination;
• and poor electrical contacts.

This helps explain why some voltage spreads stay almost unchanged for years and then start growing faster. A cracked region may remain electrically connected until repeated movement finally isolates it. A weak joint may stay usable until rising resistance and heat speed up the damage.

IEC 61215 includes thermal and other environmental stress tests for module design qualification, but the IEC states that passing qualification does not predict an exact service lifetime.^[8]

Localized Delamination Matters More Than Uniform Aging

Encapsulant aging becomes a voltage-spread issue when the damage is uneven.

If light transmission falls by about the same amount across the whole module, total output may decline while cell matching remains fairly close.

If one area yellows, becomes cloudy, or delaminates more severely, the cells in that area can receive less light and generate less current than the cells around them.

Delamination can also change moisture paths and reduce mechanical support around cells and conductors. Tongwei's discussion of module delamination covers moisture ingress, UV exposure, thermal cycling, and encapsulant aging as interacting causes.

For voltage spread, the useful question is not whether the encapsulant is old. It is whether the damage is uneven enough to make one group of cells behave differently from another.

PID Can Increase Spread, but Spread Does Not Prove PID

Potential-induced degradation is a specific electrical degradation mechanism linked to high system voltage together with cell technology, module materials, moisture, temperature, and system design.

Depending on the PID mechanism, it can increase leakage or change other electrical properties of the affected cells.

If some cells are affected more strongly than others, voltage spread can increase. If the effect is relatively even, module performance may fall without producing a very large cell-to-cell spread.

A voltage-spread measurement alone cannot identify PID.

IEC 61215-1-1 includes a specific PID detection test for crystalline-silicon modules, which is why PID should be treated as a defined failure mechanism rather than a general label for module aging.^[9]

Modern Modules Do Not All Fail the Same Way

Module construction changes how cracks, resistance, shading, and bypass protection show up electrically.

• full-cell versus half-cut layouts;
• MBB versus 0BB or dense-wire interconnection;
• different substring layouts;
• glass-backsheet versus glass-glass construction;
• TOPCon versus HJT cell technology.

More current-collection paths can reduce the effect of certain crack patterns. Different cell technologies can also respond differently to temperature and reverse bias. IEA PVPS notes that newer interconnection and cell technologies change several degradation and failure mechanisms.^[10]

The differences between TOPCon and HJT cell structures are one example of why voltage and thermal behavior should not be assumed to be identical across technologies.

A current Tongwei module example operates around 40–41 V at maximum power and roughly 17.35–17.63 A, depending on model. At those current levels, small resistance changes can matter much more than they would in a lower-current module.

A 0.20 V Cell Loss Can Hide Inside a 40 V Module

Module voltage can hide a local problem because it is the sum of many cell or cell-section voltages.

Suppose a module is operating at about 40 V and one cell loses 0.20 V while the bypass state remains unchanged.

The module-level change is approximately:

0.20 ÷ 40 × 100 = 0.5%

A 0.5% change in total module voltage can be easy to miss in routine monitoring.

At the cell level, however, 0.20 V is a large change because one silicon cell normally operates at only a fraction of a volt.

This is why one cell can become a strong electrical outlier while the inverter still shows a module that appears broadly normal.

The reverse also applies. A large module-voltage drop may come from bypass-diode operation, a substring fault, shading, or wiring rather than slow aging across every cell.

Compare Voltage Only Under Similar Conditions

A voltage reading is only useful if the test conditions are known.

IEC 60891 defines procedures for translating measured PV I-V characteristics between different temperature and irradiance conditions.^[11]

So if a cell reads 0.55 V one year and 0.49 V the next, that alone does not prove aging. The later test may simply have been taken at a different current or temperature.

Use Several Tests Before Calling a Cell Bad

A single low-voltage reading is weak evidence. A fault becomes much more convincing when several tests point to the same place.

IEC TS 62446-3 defines requirements for outdoor PV thermography.^[12] IEC TS 60904-13 covers electroluminescence image acquisition, processing, and interpretation.^[13]

For example, growing voltage difference + repeated hot-cell pattern + EL inactive area + worsening I-V curve is much stronger evidence than one unusual voltage reading on its own.

These Five Signs Matter More Than One Voltage Threshold

There is no single cell-voltage-spread number that works as a failure limit for every module.

• Persistent: the difference remains after shade, dirt, or temporary temperature effects are removed;
• Repeatable: the same cell remains the outlier under similar test conditions;
• Growing: the difference increases over time;
• Hot: the same location shows abnormal heat during operation;
• Confirmed: I-V, EL, infrared, or physical inspection supports the same fault.

If these signs are missing, the voltage difference may simply be temporary operating variation rather than permanent cell damage.

Do Not Use Voltage Spread to Predict Remaining Life

Voltage spread can show that cells are no longer behaving alike. It cannot tell you, by itself:

• remaining module life;
• the year of failure;
• fire probability;
• the exact fault mechanism;
• or whether immediate replacement is necessary.

Those decisions also depend on temperature, insulation condition, physical damage, I-V performance, fault progression, module design, and manufacturer limits.

IEA PVPS failure guidance also separates fault identification, severity, risk, inspection evidence, and corrective action instead of treating one number as a complete diagnosis.^[14]

Owners Should Check External Causes First

Before treating the problem as internal cell aging, rule out the conditions that can create the same symptoms:

• new tree or building shade;
• bird contamination;
• leaves or debris;
• localized snow;
• broken glass;
• visible discoloration;
• delamination;
• and changes in inverter or monitoring data.

Do not puncture encapsulation or expose internal cell conductors just to measure voltage.

PV strings can operate at hazardous DC voltages, and damaged encapsulation can create new moisture and insulation problems.

Installation and maintenance should follow the relevant manufacturer instructions. Tongwei's PV module installation documentation contains product-specific handling and installation requirements.

Cell-level electrical tests, energized string measurements, junction-box work, and invasive diagnostics should be left to qualified PV personnel.

Finally

Growing cell voltage spread matters most when the same cell stays abnormal, the difference keeps increasing, and other evidence supports the fault. The scale can be easy to underestimate: a 10 mΩ connection defect creates only 0.05 V of extra loss at 5 A but about 0.175 V and 3.06 W of heat at 17.5 A. One cell losing 0.20 V may change a 40 V module by only about 0.5%, so module-level monitoring can miss a serious local problem. Compare readings under similar current and temperature, then use thermal imaging, I-V testing, EL imaging, or trend data to confirm what is actually happening.