Battery Contact Stamping: Terminals and Cells

Battery contact stamping turns flat metal strip into the terminals, tabs, busbars and spring contacts that physically and electrically link a cell to everything around it. Get the material or the weld wrong and the failure is not subtle. A high-resistance joint quietly heats, a cold (non-fused) weld opens under vibration, and exposed steel hiding under a thin nickel skin rusts through in months of humid storage. Those three modes - elevated contact resistance, non-welded cold joints, and base-metal corrosion - account for most field returns on stamped battery hardware. The rest of this guide covers the material choices, plating specs, and joining processes that keep them off your line.

What follows is written for the engineers and buyers who actually source these parts: the person naming the nickel grade, the one signing the plating thickness, and the one who has to explain a pack return in the third quarter. Every section carries the numbers that belong on the RFQ - resistivity, plating windows, weld parameters, current-density limits, and the acceptance tests that prove the joint - plus the failure modes that make each number matter. If you are specifying stamped battery hardware, the tables below are the checklist to run against your drawing before it goes to tooling.

The Snapshot

Battery contact hardware splits into four stamped families - button cell terminals, cylindrical cell tabs and busbars, battery nickel strip, and spring contacts - and each one fails by a different mechanism.
Pure nickel strip (Ni200 / N6, ≥99.5% Ni) at 6.99 µΩ·cm is the weldable default for interconnects; nickel-plated copper at 1.68 µΩ·cm core resistivity is the high-current upgrade.
Nickel plating thickness of 2-3 µm minimum holds for indoor consumer packs; automotive and marine service wants 5-10 µm, verified by porosity test on bent samples.
Joint resistance is the number that predicts survival: a clean nickel spot weld measures 0.1-0.5 mΩ, a cold weld 5-50 mΩ, and at 10 A a 10 mΩ joint wastes 1 W in a spot the size of a pinhead.

What Actually Gets Stamped

The "battery contact" family covers four distinct parts, and each one punishes a different mistake:

Button cell terminals - the positive cap and the negative can of CR2032, CR2025, CR2016 and similar coin cells. These are deep-drawn and then often stamped for the terminal geometry. The risk here is plating continuity: a pinhole in the nickel lets the stainless or steel substrate corrode and drive up internal resistance.
Cylindrical cell tabs and busbars - the strips welded across 18650, 21700 and 26650 cell tops to build series and parallel packs. The killer here is weld consistency and bulk current rating.
Battery nickel strips - the raw interconnect stock, 0.1-0.3 mm thick and 5-10 mm wide, supplied in rolls for high-volume welding. This is the part most often confused with nickel-plated steel by cost-driven buyers.
Spring contacts - beryllium-copper or phosphor-bronze stamped springs that hold a controlled pressure against a terminal. They fail on fatigue and on contact-force drift, not on corrosion.

The four families share a stamping floor but not a design logic. The button cell cap is a deep-draw with a terminal boss, dominated by material flow and plating continuity. The tab is a flat strip with a formed foot, dominated by cross-section and weld geometry. The nickel strip is a coil of simple slit stock, dominated by width tolerance and surface finish. The spring contact is a thin formed blade, dominated by bend radii and temper. A supplier that treats all four as "thin metal parts" will quote you four wrong answers; the correct approach is one material and process review per family, because the die design, the press speed, and the plating mask are different for each.

Picking the Base Metal

Three production materials dominate: pure nickel, nickel-plated steel, and copper (almost always nickel-plated for battery use). Resistivity and weldability decide, not price alone:

Material	Resistivity (µΩ·cm)	Relative cost	Weldability	Typical use
Pure nickel strip (Ni200 / N6, ≥99.5% Ni)	6.99	High	Excellent	18650 busbars, premium tabs
Nickel-plated steel (SPCC + 2-10 µm Ni)	~12-20 (steel-dominated)	Low	Good, plating-sensitive	Cost-sensitive packs, cans
Copper, bare (C1100 / ETP)	1.68	Medium	Poor (oxidizes)	Rarely bare in cells
Nickel-plated copper	1.68 (core)	Medium-high	Excellent	High-current busbars
Stainless 301/304 (springs)	~72	Medium	Difficult	Spring contacts, cans
Beryllium copper C17200 (springs)	~5.8	High	Good	Reliable spring contacts

Pure nickel (ASTM B162 / GB/T 2072 Ni200 or N6) is the safe default for welded interconnects: 6.99 µΩ·cm, non-magnetic in the sense that it does not attract like steel, and it welds cleanly because there is no plating layer to burn away. Nickel-plated steel is the budget option. The steel base carries roughly twice the resistivity of nickel, but the surface is conductive enough for spot welding - provided the plating is intact. Once the nickel is breached, the steel underneath rusts and the joint resistance climbs.

Copper is the conductivity champion at 1.68 µΩ·cm, but bare copper oxidizes and resists reliable resistance welding. For high-current busbars the smart move is nickel-plated copper: copper core for ampacity, nickel skin for weldability and corrosion resistance. That combination is what you want on packs pulling 20 A per parallel group or more.

The selection logic is a three-question gate, not a price lookup. First, what is the current per parallel group? Below roughly 5 A, pure nickel carries the job and the premium is justified by weld reliability; above 20 A, the nickel strip cross-section grows so fat that nickel-plated copper wins on both temperature rise and pack weight. Second, what does the joint have to survive? Vibration, thermal cycling, and humid storage all push toward pure nickel or plated copper, because the joint interface - not the bulk - is where plated steel fails first. Third, what is the cell chemistry's voltage floor? Low-voltage signal contacts cannot afford the resistance drift of a breached plating layer, so stainless and plated steel get pushed down the list. Run the three questions on every new pack design and the material table fills itself in.

Nickel Plating Is the Corrosion Insurance

For any steel or copper substrate, the nickel layer is the only thing standing between the cell environment and base-metal corrosion. Production specs that hold up in the field:

Thickness: 2-3 µm minimum for indoor consumer packs; 5-10 µm for automotive, marine, or high-humidity service. ASTM B689 and ISO 4526 govern electroplated nickel; EN 12540 covers the European plating spec.
Ductility: the deposit must bend with the strip without cracking. A bright, brittle deposit fractures at the bend line and exposes substrate exactly where stress concentrates.
Porosity: specify a porosity test (ferroxyl or humidity exposure) on bent samples. A pinhole at a bend is a corrosion seed.

For button cell cans and caps, a thin nickel or even gold flash over stainless is common to guarantee low, stable contact resistance at the external terminal. Gold flash of 0.05-0.5 µm is enough to protect the surface without adding meaningful cost.

Service environment	Nickel thickness	Substrate	Acceptance test
Indoor consumer pack	2-3 µm min	Steel or copper	Ferroxyl porosity, bent sample
Automotive / marine / high humidity	5-10 µm	Steel or copper	Humidity exposure, salt spray per ASTM B117
Button cell external terminal	Ni 1-3 µm or Au flash 0.05-0.5 µm	Stainless	Contact-resistance stability check
High-temperature service near cells	5-10 µm + Ni underplate logic	Copper core	Thermal-cycle adhesion check

Plating thickness is a location decision. On a tab that gets welded at the foot and handled at the head, the nickel has one job at the weld zone (burn through cleanly, fuse the copper) and another at the terminal face (stay intact for years). That is why we plate the coil selectively on reel-to-reel progressive stamping: the functional thickness lands where the spec names it, and the precious metal is not wasted where it does no work. The manifest range that holds in production is 2-8 µm for nickel in selective reel-to-reel application, with zinc at 5-12 µm reserved for structural hardware that never carries a weld. Whatever the thickness, the porosity check belongs on a bent sample, because a deposit that survives flat and cracks at the bend line is the exact failure mode that shows up in the field a year later.

Contact Resistance Is the Spec That Matters

Contact resistance is the single number that predicts whether a joint survives. For a clean nickel-to-nickel spot weld, joint resistance runs 0.1-0.5 mΩ. A cold or contaminated weld measures 5-50 mΩ. The arithmetic is unforgiving: at a 10 A load, a 10 mΩ joint dissipates P = I²R = 1 W in a tiny spot, raising local temperature and accelerating every other failure mode. Button cell internal resistance is 10-200 mΩ depending on chemistry and size, so the stamped terminal weld must add less than 5 mΩ or it starts to dominate the cell's own losses.

Measure joint resistance with a four-wire (Kelvin) milliohm meter on production samples, not a two-wire multimeter. Two-wire readings include lead and probe resistance and will hide a bad weld.

Joint condition	Typical resistance	Verdict at 10 A
Clean nickel-to-nickel spot weld	0.1-0.5 mΩ	Accept; loss under 5 mW
Good nickel-plated copper weld	0.2-0.8 mΩ	Accept; verify by peel test
Cold or contaminated weld	5-50 mΩ	Reject; 0.5-5 W local heating
Loose mechanical contact, no weld	50 mΩ and climbing	Reject; thermal runaway risk

Resistance readings have to be paired with a mechanical test to mean anything. A joint can measure fine cold and still be a cold weld that will open under vibration, so the control plan runs both: four-wire milliohm on a per-shift sample and destructive peel testing on the same frequency. The two numbers together - resistance under 1 mΩ and a peel that tears the parent metal instead of the weld - are the only reliable proof that the joint is fused and not just stuck.

Spot Welding vs Laser Welding

Both processes weld thin battery strip, and both have hard limits you need to design around.

Resistance Spot Welding

This is the workhorse for 18650 and 21700 pack assembly. Typical parameters for 0.15 mm nickel strip on a nickel-plated steel or pure-nickel surface:

Weld current: 3000-6000 A (capacitive-discharge or inverter DC supply)
Weld time: 5-20 ms (about 1-4 cycles at 50 Hz)
Electrode force: 40-80 N
Electrode tip: truncated cone, 3-5 mm face, RWMA Class 2 (zirconium copper)
Target nugget diameter: ≥ 4√t, where t is total joint thickness - for a 0.3 mm total stack, that is ≥ 2.2 mm

The failure to watch is the cold weld: enough current to stick but not enough to fuse, or oxide between surfaces that blocks nugget formation. It passes a pull test weakly and opens later. Destructive peel testing on a per-shift basis is the only reliable guard.

Laser Welding

Fiber lasers at 300-1000 W, 1064 nm, weld with a small heat-affected zone and no electrode force, which protects heat-sensitive cells. Practical numbers:

Penetration: 0.1-0.8 mm per pass on thin strip
Weld speed: 10-60 mm/s
Mode: keyhole, tight HAZ

Reflective metals are the catch. Nickel absorbs reasonably at 1064 nm, but copper absorbs only about 5%, so copper and nickel-plated copper busbars often need a green (515 nm) or blue (450 nm) laser where absorption climbs to 40-60%. If you spec laser welding on copper, confirm the wavelength with your supplier before tooling.

Decision	Resistance spot welding	Laser welding
Best for	Nickel and nickel-plated steel strip, cell-top taps	Copper and nickel-plated copper busbars, thin HAZ
Heat input	Localized at electrode pair	Tight keyhole, minimal cell heating
Electrode wear	Real; RWMA Class 2 tips dressed per shift	None - no contact force
Reflectivity problem	None (resistive heating)	Copper needs green or blue wavelength
Process control	Current, time, force logged per weld	Power, speed, focus logged per weld
Primary failure	Cold weld from oxide or low current	Incomplete penetration on reflective surface

The process choice is a joint-location decision, not a technology preference. Resistance welding wins on nickel interconnects where hundreds of welds per pack make cycle time and electrode cost the controlling factors. Laser wins where the joint sits close to a heat-sensitive cell, where the material is copper, or where the weld must be cosmetically clean for an exposed busbar. Many packs run both: spot welds on the nickel taps, laser on the copper busbar ends. Specify the joint, the material, and the current, and let the process follow.

Sizing for Current

A 0.15 mm × 8 mm pure nickel strip presents a 1.2 mm² cross-section. Carrying 10 A puts current density near 8.3 A/mm², and that strip stays under a 30 °C rise over short interconnect runs. Pure nickel thin-strip ampacity lands around 10-15 A for that geometry; nickel-plated copper of the same size carries roughly three times the current thanks to the copper core. The limiting factor on most packs is not the strip bulk but the weld joint resistance - a 5 mΩ joint on a 10 A path wastes half a watt the strip never would.

For 18650 packs, remember the cell rating. A typical 18650 delivers 2-5 A continuous, with high-drain cells rated 20-30 A. Your busbar must handle the sum of the parallel group, not the single-cell number. Size for the parallel-group current plus a 50% thermal margin and you will not see summer field returns.

Strip geometry	Cross-section	Practical current (short run)	Material
0.1 mm × 5 mm	0.5 mm²	4-6 A	Pure nickel
0.15 mm × 8 mm	1.2 mm²	10-15 A	Pure nickel
0.2 mm × 10 mm	2.0 mm²	18-25 A	Pure nickel
0.2 mm × 10 mm	2.0 mm² (copper core)	40-60 A	Nickel-plated copper
0.3 mm × 15 mm	4.5 mm² (copper core)	90-120 A	Nickel-plated copper

Two sizing rules keep the pack honest. First, size the strip for the weld-zone constraint, not the strip bulk: the joint is the smallest cross-section in the path, and its resistance sets the local temperature. Second, size for the worst parallel group with margin, because pack current is never evenly split - cell internal resistance varies by a few milliohms and the healthiest cell in the group carries the most current. A busbar sized to the average is a busbar that runs hot in one corner of every pack.

Spring Contacts and Stable Force

Spring contacts live or die on contact force and fatigue, not corrosion. Beryllium copper C17200 in spring temper (TM00/TM04) gives 22-25% IACS conductivity with 1100-1400 MPa tensile, and a fatigue life beyond 10&sup6; cycles. Phosphor bronze (C5191/C5210) is the cheaper alternative with lower conductivity. Stainless 301 is used when corrosion resistance outranks conductivity.

Design the contact for 0.3-2 N of force per point, with 0.5-2 mm of working deflection. A nickel underplate of 1-3 µm keeps the surface weldable and corrosion-free; a thin gold flash drops and stabilizes contact resistance for low-signal or low-voltage contacts. Avoid designing a spring that sits at the very top of its deflection range - that is where force drift and relaxation live.

Spring material	Conductivity (% IACS)	Tensile (MPa)	Fatigue life	When to use
Beryllium copper C17200	22-25	1100-1400 (aged)	>10&sup6; cycles	Reliable contacts, high force in small envelope
Phosphor bronze C5210	10-13	700-750 (hard)	10&sup5;-10&sup6; cycles	Cost-sensitive spring contacts
Phosphor bronze C5191	12-15	550-650 (hard)	10&sup5;-10&sup6; cycles	Deeper-formed springs, softer force
Stainless 301	~2-3	1200-1500 (full hard)	10&sup5;-10&sup6; cycles	Corrosion-critical, non-current-carrying springs

Force stability is a stamping outcome, not a material promise. The bend radii, the grain direction, and the stress-relief treatment after forming decide whether the contact holds its 0.3-2 N at the end of life. A bend radius below 1.5 times the material thickness work-hardens the inside of the bend and becomes a crack origin; sheared edges with heavy burrs act as stress raisers on the same feature. Specify the temper, the grain direction relative to the bend, and the stress-relief cycle on the drawing, and verify with a force-at-deflection measurement at first article and on per-lot samples. A spring that drifts 20% in force over 100,000 cycles has failed even though it never broke.

DFM Notes That Save the Program

Bend radius: keep it at 0.5-1× material thickness for nickel; spring steels tolerate tighter radii but work-harden fast.
Burrs: hold burr height under 10% of thickness. A burr across a weld gap or a spring contact face creates a spot that will not seat.
Strip range: production battery contacts run 0.05-0.5 mm thick, held to ±0.02-0.05 mm. Progressive dies with 15-40 stations handle tabs with formed feet, slots, and spring features in one pass.
Plating before or after forming: plate the coil, then form. Forming after plating risks cracking the deposit at bends - verify with a bent-sample porosity check.

Add three more notes from the press floor. First, pilot-hole geometry: the carrier strip that feeds a 30-station progressive die is part of the part, and the pilot holes that keep the tab in position at 300 SPM have to be designed in from the start, not added after tooling. Second, weld-location tolerance: the formed foot that lands on the cell terminal has to hold its position relative to the weld datum, because a 0.1 mm shift moves the weld off the center of the tab and turns a good joint into an edge weld. Third, edge condition on the weld face: a sheared edge with a torn zone welds differently than a coined or shaved edge, so if the weld quality matters, name the edge finish on the drawing. These three notes are where stamping and pack assembly stop being separate disciplines.

Feature	Held tolerance	Why it matters
Strip thickness	±0.02-0.05 mm	Sets cross-section and weld energy demand
Burr height	<10% of thickness	Weld gap seating, spring face contact
Weld-foot position	±0.05-0.1 mm to datum	Weld lands centered, not on the edge
Bend radius	0.5-1× thickness (Ni)	Plating continuity, fatigue life
Plating thickness at weld zone	2-8 µm per spec, XRF-verified	Clean burn-through, corrosion life

Failure Modes That Drive Field Returns

Four failure signatures account for nearly every returned battery contact assembly we have examined. Recognizing them tells you which feature to audit first.

Cold Weld Opens Under Vibration

A pack passes final test, ships, and fails intermittently after three months of shipping vibration. The tab is still attached - weakly. The spot weld fused a few grains and then stopped: low current, worn electrodes, or oxide on the strip surface blocked nugget growth. The audit trail is the per-shift peel-test record and the weld-current log; a supplier that cannot produce either is quoting you luck.

Plating Breach at the Bend Line

A nickel-plated steel tab corrodes at the formed foot after a humid season. The deposit cracked during forming, the crack exposed the steel base, and the rust ran along the grain. This is the failure that incoming inspection never sees, because the crack is invisible until it corrodes. The audit trail is the bent-sample porosity test and the XRF measurement at the bend, not on the flat coupon.

Current Density Meltdown on an Undersized Busbar

A high-drain pack runs hot at the busbar, the nickel-plated copper strip discolors, and the weld pads lift. The strip was sized to the single-cell rating instead of the parallel-group sum, and the missing 50% thermal margin turned a 60 A design into a 40 A part carrying 60 A. The audit trail is the sizing table against the pack current map and the temperature-rise measurement at full load.

Spring Force Drift Kills the Contact

A spring contact measures correct force at first article and drops 30% after a year in service. The bend radius was tighter than the drawing, or the stress-relief cycle was dropped, so the blade relaxed toward its yield point. The audit trail is the force-at-deflection measurement at first article and per-lot, plus the die revision history that shows the bend radius never changed.

All four failures share one property: the part passed a dimensional inspection and still failed functionally. That is why the battery contact RFQ should name functional targets - joint resistance, weld nugget, plating porosity, spring force at end of life - and let the supplier's control plan prove them, instead of trusting a drawing that measures what is easy to measure.

Sourcing Checklist and Next Step

Put these items on the RFQ before you send it, and you will get quotes that are comparable instead of four different products.

Cell type and chemistry, peak and continuous current per parallel group, and the ambient environment (indoor, automotive, marine).
Material grade and temper on the drawing - pure nickel Ni200/N6, nickel-plated steel with the plating spec, or nickel-plated copper with the core grade.
Plating thickness and location, named per surface, with the test method (XRF on the part, ferroxyl porosity on bent samples).
Weld process and parameters or the joint acceptance criteria - joint resistance target, nugget size, peel-test requirement.
Spring parts: force-at-deflection target, working range, cycle-life requirement, and the stress-relief spec.
Tolerance table: strip thickness, burr, weld-foot position, and the features that carry current or weld.
Volume and tooling ownership terms, because progressive-die tooling is where battery contact economics are decided.
Ask for the first-article package: dimensional data, material and plating certificates, weld samples, and the four-wire resistance readings.

If you are sourcing battery terminals, 18650 busbars, nickel strips, or spring contacts and want the material and joining spec locked to your pack's real duty cycle, send us the cell type, peak current, and target environment - we will return a stamped-part proposal with the plating and weld process called out, not a generic quote. The DFM review is free, and the first-article package with resistance and peel data is standard on every battery contact program. Request a quote and include the pack current map; that is the one document that makes the material and weld answers obvious.