After undergoing repeated charge-discharge cycles, the individual cells within a drone battery pack gradually diverge in terms of capacity, internal resistance, and voltage. Some cells degrade to the retirement threshold sooner than others, forcing the premature retirement of the entire pack, even though many other cells within it still possess significant remaining service life. Research published in *Nature Energy* in August 2026 by Chen Zhongwei’s team at the Dalian Institute of Chemical Physics provided quantitative data on this issue: the full-lifecycle energy resource utilization rates for battery packs in electric passenger vehicles and buses are merely 80.7% and 72.9%, respectively—meaning that when the weakest cell forces the pack’s retirement, approximately 20%–27% of the potential energy remains untapped. Drone battery packs face the same problem; moreover, due to the extreme demands drones place on discharge rates and weight, the impact of cell inconsistency is often even more pronounced. Why, then, do individual cells that start with nearly identical parameters end up facing vastly different fates during cycling?
I. Initial Differences
1.1 What Screening Can Achieve
Before drone battery cells are assembled into packs, individual cells undergo rigorous screening and matching. Standard screening parameters include:
Capacity: Variations in actual discharge capacity within the same batch are controlled to within ±1%–3%.
AC Internal Resistance (ACIR): Variations are controlled to within ±5%.
Open-Circuit Voltage (OCV): Variations are controlled to within ±5mV–10mV.
Self-discharge Rate: Abnormal cells are eliminated based on the K-value (voltage drop over time).
These screening methods effectively eliminate “obviously substandard” cells, ensuring that individual cells within the assembled pack are highly similar in terms of macroscopic parameters and preventing severe “weakest-link” effects.
1.2 What Screening Cannot Eliminate
Even if the parameters mentioned above are identical, significant microscopic differences remain between individual cells—differences that cannot be detected through standard screening. These stem from random fluctuations in manufacturing process parameters:
Coating Areal Density Deviation: Micron-level variations in electrode coating thickness lead to localized differences in active material loading.
Electrolyte Filling Error: Slight differences in the degree of electrolyte wetting affect the uniformity of ion transport channels.
Uneven Calendering Thickness: Localized fluctuations in compaction density alter electrode porosity and electron conduction paths.
Separator Thickness and Porosity Deviations: These affect the local uniformity of lithium-ion migration rates.
Tab Welding Quality: Even if a weld joint appears acceptable, there are differences in microscopic contact area and grain structure.
Researchers such as Long Zhou from the University of Shanghai for Science and Technology categorize the factors affecting battery consistency into three types: initial factors (capacity, internal resistance, etc., determining basic capability), current factors (voltage, SOC, etc., determining immediate status), and time-accumulated factors (capacity fade rate, internal resistance growth rate, coulombic efficiency, etc., determining future trajectory).
Screening during assembly can only control the first two categories; the third category—time-accumulated factors—is what truly determines the battery pack’s long-term performance, yet these are precisely the factors most difficult to measure accurately before the product leaves the factory. Take capacity fade rate as an example: suppose there are two cells, A and B, connected in series; the average reversible capacity fade rate per cycle is 0.005% for cell A and 0.008% for cell B. This difference is virtually undetectable during factory testing; however, after 500 cycles, Battery A shows a capacity loss of 2.5%, whereas Battery B reaches 4%. Without balancing protection, Battery B—due to its faster capacity degradation—hits the upper voltage limit first during charging. This effectively results in significant overcharging, which in turn accelerates further degradation, creating a vicious cycle.
II. Temperature Gradients
2.1 The Inevitability of Temperature Differences: Drone batteries undergo high-rate discharge during flight (typically 5C–15C, or even higher for racing drones), generating significant heat due to the combined effects of ohmic and polarization internal resistances. However, the heat dissipation conditions for individual cells within the battery pack vary naturally: Outer cells are in direct contact with external airflow or heat-dissipating structures, resulting in short heat dissipation paths and lower temperatures; inner cells are surrounded by other cells, requiring heat to conduct through multiple layers to reach the dissipation surface, leading to significantly higher temperatures. Drone battery packs are typically encapsulated in heat-shrink film, which further restricts the dissipation of internal heat. Studies indicate that the temperature difference between the interior and exterior of a heat-shrink-sealed battery pack can reach 8–15°C under natural cooling conditions; even with active cooling, the difference often exceeds 3–5°C.
2.2 How Temperature Accelerates Divergence in Lifespan: The impact of temperature on lithium-ion battery aging is non-linear. High-temperature effects accelerate aging: SEI film thickening (the growth rate of the SEI film roughly doubles for every 10°C rise in temperature; a thickened SEI film consumes active lithium ions, increases internal resistance, and reduces capacity); electrolyte decomposition (high temperatures accelerate oxidative decomposition, generating gases and by-products that degrade the electrode/electrolyte interface); and cathode material structural degradation (high temperatures accelerate lattice oxygen release and transition metal dissolution, leading to irreversible capacity loss). Low-temperature effects involve hidden damage: Lithium plating (when outer cell temperatures are low, the diffusion rate of lithium ions on the anode surface during charging decreases, making metallic lithium prone to plating out on the anode surface; this plated lithium not only causes irreversible capacity loss but may also form dendrites that puncture the separator, triggering internal short circuits).
2.3 The Positive Feedback Loop of Temperature, Internal Resistance, and Heat Generation: Temperature differences can also create a vicious cycle driven by changes in internal resistance: inner cells reach higher temperatures due to poor heat dissipation, and this high temperature accelerates SEI film thickening and increases internal resistance; the increased internal resistance generates more Joule heat during high-rate discharge, causing the temperature to rise further and accelerating the aging process. Ultimately, this causes the lifespan gap between the inner and outer cells to widen progressively. This issue is particularly pronounced in drone batteries; the discharge rate during a single drone flight is far higher than that of an electric vehicle, resulting in greater additional heat generation within the inner cells for the same temperature differential and a more intense positive feedback effect.
III. Welding Process
3.1 Consistency Requirements for Weld Joint Resistance: In drone battery packs, the quality of the weld between the cell tab and the nickel strip or busbar directly impacts electrical consistency and thermal behavior. A robust welding process ensures high consistency in contact resistance across all weld joints (typically requiring <0.5 mΩ, with a variation of <0.1 mΩ between joints). However, in actual production, weld quality is influenced by various factors: minor fluctuations in welding parameters (current, pressure, time), batch-to-batch variations in the thickness of the oxide layer on the tabs, and uneven pressure distribution caused by weld head wear. These factors can lead to contact resistance at specific weld joints being significantly higher than at others—a discrepancy often overlooked during factory inspections.
3.2 The First Hazard of High-Resistance Weld Joints: Localized Overheating. Drone battery packs frequently operate with discharge currents reaching tens of amperes. Taking a 4S battery pack with a continuous discharge current of 50A as an example: for a normal weld joint (contact resistance 0.3 mΩ), heat generation power = 50² × 0.0003 = 0.75 W; for an abnormal weld joint (contact resistance 1.5 mΩ), heat generation power = 50² × 0.0015 = 3.75 W. The heat generated by the abnormal joint is five times that of the normal joint. This excess heat is conducted directly to the connected cell tab and terminal, causing the local temperature of that specific cell to rise above that of the others, accelerating SEI layer growth and electrolyte decomposition, and thereby hastening its aging process relative to the other cells.
3.3 The Second Hazard of High-Resistance Weld Joints: Voltage Sampling Distortion. This is the most insidious and easily overlooked hazard associated with welding issues. In a battery system, the BMS monitors the voltage of each individual cell via voltage sampling lines (sense lines). If high contact resistance exists at any of the solder joints between the voltage sensing lines and the battery electrodes, an additional voltage drop is introduced into the sensing circuit. This creates a discrepancy between the voltage “seen” by the BMS and the actual voltage of the cell. During charging, current flows from the charger into the battery pack; the high-resistance solder joint causes a voltage drop in the sensing circuit, making the voltage measured by the BMS appear higher than the cell’s true voltage. As the voltage approaches the charging cutoff limit, the BMS perceives the cell as having reached the cutoff voltage and terminates charging. Conversely, during discharge, the current flows in the opposite direction, causing the BMS to measure a voltage lower than the cell’s actual voltage. The BMS perceives the cell as having reached the discharge cutoff voltage and stops discharging—even though the cell has not actually reached that limit. If the BMS features active balancing, it may even force other cells to transfer energy to this cell, which appears to have a low voltage. The net result is that this cell operates in a state of shallow charge and discharge, experiencing “milder” cycling conditions than the other cells. Because this cell undergoes shallow cycling, its usable capacity remains lower, thereby dragging down the total capacity of the entire battery pack. Over long-term operation, this cell degrades more slowly than the others, creating a disparity in lifespan compared to the rest of the battery pack.
IV. Differences in Mechanical Stress
4.1 Uneven Distribution of Expansion Force: During the charge and discharge cycles of lithium-ion batteries, electrode materials undergo repeated lithium intercalation and de-intercalation, accompanied by volumetric expansion and contraction. Taking a graphite anode as an example, the volume expands by approximately 10%–13% upon full lithiation; silicon-based anodes exhibit much more drastic volume changes, exceeding 300%. In drone battery packs, individual cells are subject to varying constraints: internal cells are tightly surrounded by adjacent cells and structural components, resulting in greater constraints on expansion, whereas external cells are constrained on only one side, with the other side potentially resting against an end plate.
4.2 How Mechanical Stress Affects Lifespan: Uneven mechanical stress impacts cell lifespan through the following mechanism—changes in internal resistance: gas generation is inevitable during cell cycling. Since drones predominantly use pouch cells, the ability to constrain gas expansion is relatively poor. In regions of the cell subject to high constraint, gas generated due to stress migrates toward lower-pressure peripheral areas; conversely, in low-constraint regions, gas may become trapped between electrode sheets, impeding lithium-ion transport and manifesting as increased internal resistance, electrode deformation, and delamination. Uneven expansion forces can cause electrode bending, reduce adhesion between active materials and current collectors—potentially leading to localized delamination—and increase resistance along electron transport paths. Regarding separator compression and porosity changes, separators in high-constraint zones undergo compression, resulting in reduced porosity, hindered ion transport, and increased localized polarization. Interface contact degradation also occurs, as repeated expansion and contraction cycles cause the contact between internal cell layers to deteriorate progressively. These forms of mechanical damage are incremental; while almost imperceptible during early cycling, the damage accumulates over time, ultimately causing more heavily constrained internal cells to exhibit accelerated capacity fade and rising internal resistance.
The divergence in cycle life among individual cells within a drone battery pack is not caused by a single factor; rather, it is the result of multiple factors—ranging from microscopic process fluctuations during manufacturing to thermodynamic imbalances, electrical connection variances, and uneven mechanical stresses post-assembly—coupling, accumulating, and amplifying non-linearly over long-term cycling. Understanding this complex scenario implies that extending the cycle life of drone battery packs cannot be achieved merely by optimizing a single stage; instead, it requires coordinated control across the entire value chain—including cell manufacturing, cell sorting and matching, thermal management design, and welding processes—to fully unlock the lifespan potential of every individual cell within the pack.