Mid-stage lithium battery terminology: winding and stacking, term by term

Winding and stacking form the electrode assembly at the middle of the lithium-cell line, and the equipment vocabulary around them trips up newcomers. Here are the high-frequency terms, what they measure, and why they matter.

Geometry: overhang, slitting accuracy, tab pitch

Overhang (electrode alignment) is the positional offset between cathode and anode, usually stated as how far the cathode edge extends past the anode. It is measured with optical microscopy or laser ranging, feeding servo correction in real time. Mainstream 2026 winders hold ±0.3 mm; high-energy-density cells demand under ±0.2 mm. Excessive offset invites lithium plating, capacity fade, and internal shorts — hence per-roll inspection.

Slitting accuracy covers edge straightness and width tolerance after electrode slitting. Foils develop wavy edges and burrs; high-speed cameras monitor the cut profile, tied to blade-wear maintenance cycles. Stacking demands more than winding here — uneven edges damage separators during stacking — with common tolerances of ±0.1 mm and width variation coefficients under 0.5% for EV cells. Judge blade material (carbide, ceramic) and edge life, not just day-one precision.

Tab pitch is the spacing between tabs on one electrode, critical in multi-tab winding: pitch drift misaligns tabs, hurting welding and resistance consistency. Equipment combines vision positioning with asynchronous servo compensation; 2026 high-speed winders hold ±0.5 mm in closed loop.

Dynamics: tension and speed

Tension control spans unwinding through winding/stacking. Too much tears foil or creases it; too little loosens the web and degrades alignment. Typical ranges: copper (anode) 0.5–2 N/cm width, aluminum (cathode) 1–3 N/cm. Control has evolved from open-loop constant torque to closed-loop PID with feed-forward and dancer rolls; 2026 machines run dual loops (inner torque, outer tension-sensor feedback). Keep peak-to-peak fluctuation under ±5% of setpoint or expect wrinkling.

Winding and stacking speed set throughput: line speeds of 0.5–3 m/s for winding, station cycle times of 0.6–1.2 s per sheet for stacking. Speed is limited by pick-and-place stability and alignment retention; high-speed 2026 machines use linear motors and lightweight suction plates, with air-film lubrication effects appearing above 2 m/s. Evaluate acceleration-inertia compensation and speed ripple, not just peak speed — ripple quietly redistributes electrode stress.

Cut-and-stack integration merges die-cutting with stacking in continuous or intermittent modes: continuous cut-stack is faster but demands extreme geometric consistency (and tolerance of cutter vibration); intermittent is more flexible with a different tension path.

Interface quality: burrs, dust, wettability

Burrs are residual metal protrusions from slitting; above 5 μm they can pierce the separator. Forms include rollover, serration, and curl, detected by optical profilometry or SEM; with in-line detection now common, machine-learning classifiers help tune cutter gaps. Key controls: blade-dulling cycles (blades typically replaced every 100k meters) and edge side-pressure.

Dust management targets metal particles and active-material debris from cutting. High dust drives self-discharge; laser particle counters check compliance, with cleanrooms at ISO Class 5 or better. In stacking, dust trapped between electrode and separator forms micro-shorts — extraction hoods and static eliminators help, verified by cleanliness sampling.

Wettability — electrode surface state (burrs, coating density) governs electrolyte penetration. Contact angle reflects surface energy (smaller wets easier); calendered surface roughness correlates negatively with wetting, and stacked cells are more sensitive due to uneven interlayer gaps. Labs use sessile-drop or capillary methods; lines usually track absorption time.

From line to cell: geometry-performance links

Thickness uniformity (coating distribution, tracked via Cpk) — non-uniformity causes local over/under-charge and accelerated aging; 2026 practice watches local compaction-density variation rather than mean thickness alone (cathode compaction deviating over 2% can cost ~15% cycle life). Jellyroll diameter / stack height must fit the can, allowing separator shrinkage and electrode swelling; in-line laser gauging holds ±0.2 mm, with ovality flagging uneven winding stress. Insertion depth (electrode into the core) sets inner-tab weld position; placement accuracy in stacking (suction-plate xy offset) is typically ≤±0.15 mm — both drive internal resistance and short risk, and some 2026 suppliers add temperature-adaptive compensation. OEE wraps it together — distinguish theoretical from effective speed, and judge from a month of downtime distribution (alignment excursions, tension alarms, burr defects), not one trial run.

Questions & answers

How is winder alignment judged? Laser or vision edge scanning; typically within ±0.3 mm, under ±0.2 mm for demanding cells — sensor zero calibration is the key discipline.

Where is the stacking speed bottleneck? Pick-and-place cycle and alignment retention; mainstream 2026 cycles reach 0.8 s, with suction-plate vibration and air-pressure ripple the barriers below that.

Why does slitting accuracy matter? Poor edges mean burrs that pierce separators; width drift blocks stack-into-can assembly. EV cells commonly specify ≤±0.1 mm width tolerance.

How should tension be set? Copper 0.5–2 N/cm width, aluminum 1–3 N/cm, adjusted dynamically for thickness and roll diameter, with fluctuation under ±5% of setpoint.

What burr height passes? Generally ≤5 μm; check by optical profilometer with periodic sampling, and maintain blades on ~100k-meter cycles.

What placement accuracy do stacked cells need? xy offset ≤±0.15 mm; 2026 equipment adds vision plus temperature compensation for stability.

How is OEE assessed properly? Separate theoretical from effective speed and log stoppage causes over a month or more — single-run peaks mislead.