Spring Stamping: Clip, Contact and Leaf Spring Design

A stamped spring fails quietly. Not with a bang, but with a number drifting the wrong way on a test bench at hour 5,000 of a 10,000-hour qualification run. Stress relaxation lets a battery clip lose its grip a few grams at a time; fatigue cracks a leaf spring after a few million cycles instead of the ten million the print asked for; a thickness tolerance that wandered makes the engagement force fall outside the spec band. Each failure looks small in isolation, and each one sends a small stamped part back to the design table, because in spring design the material, the geometry and the process are inseparable.

This guide covers the decisions that actually decide whether a stamped spring works: which spring types are stamped rather than wound, which alloys carry which load, how relaxation and fatigue are checked against published data, how geometry and springback are controlled in a progressive die, what plating does to contact performance, and how the part is tested and qualified before it goes into production. The reference process throughout is high-speed progressive stamping of strip from 0.10 to 0.50 mm, which is where most stamped springs live.

The Snapshot

• Stamped springs are planar parts made in one press stroke: battery clips, contact fingers, leaf springs, constant-force strips and snap detents, typically from 0.10-0.50 mm strip.
• 301 stainless half-hard is the default spring strip for non-current-carrying clips; C17200 beryllium copper is the contact spring standard because it combines high strength with 22-25% IACS conductivity.
• Relaxation and fatigue are the two failure modes that matter; they are checked with published material curves and a Goodman-type mean-stress correction, not with guesswork.
• Spring stiffness scales with the cube of strip thickness, so blank thickness control of ±0.01 mm is what keeps the force band tight in production.
• High-speed progressive stamping holds ±0.005 mm positioning accuracy at up to 300 SPM, which is how spring force and geometry stay repeatable across millions of parts.

Why a Stamped Spring Fails Quietly

Three failure modes show up again and again in stamped springs for automotive connectors, medical devices and industrial locks: stress relaxation that lets a clip lose its grip, fatigue fracture that cracks a leaf spring after a few million cycles, and geometry drift that pushes engagement force outside the print band. All three share a characteristic that makes them dangerous: the part still looks fine. There is no visible deformation, no crack on the surface, just a function that stops being reliable.

The failure analysis almost always comes back to a decision made before the part was stamped: alloy selection, operating stress, or the tolerance the die was built to hold. That is why the first article of spring design is not the drawing, it is the operating envelope - temperature, cycle count, load, environment - because every one of those inputs maps to a different alloy and a different geometry.

Stamping a Spring vs. Winding One

Stamping builds flat, planar springs in a single press stroke: battery clips, contact fingers, leaf springs, constant-force strips and snap detents. A progressive die turns a coil of 0.10-0.50 mm strip into millions of identical parts, holding blank thickness to ±0.01 mm and feature position to ±0.03 mm in production. Wire forming cannot match that planar complexity - a wireform is a line in space, a stamping is a two-dimensional shape with engineered stiffness - and nothing else matches the piece price at high volume.

The trade-off is that stamping only produces flat or bent-from-flat geometry. If the spring must wrap around a cylinder or carry load in three dimensions, a wound coil or wireform is the right route. If the spring is a clip, finger, beam or strip that works in a plane, stamping wins on cost, repeatability and integration: the spring can share the strip with a terminal or contact, be plated selectively, and be delivered on reels for automated assembly. For connector spring contacts specifically, the stamped route is the industry default, as covered in our stamped spring contacts design guide.

Volume decides the process as much as geometry does. Progressive die tooling is a capital investment, and the breakeven against wire forming or laser cutting typically lands in the hundreds of thousands of parts. Below that, a simple forming die or prototype stamping makes sense; above it, the progressive die amortizes fast. The progressive die high-speed stamping route is where spring cost per piece reaches its floor, because the strip, the die and the press are all running at their designed speed.

Spring Strip Alloys: The Selection Map

Four alloy families cover most stamped spring work, and each one buys a different combination of strength, conductivity, relaxation resistance and corrosion performance. The selection is a trade, not a ranking: no single alloy is best at everything.

301 Stainless, Half-Hard - The Workhorse

301 in the half-hard temper (1/2 hard per ASTM A666) is the default spring strip for non-current-carrying clips. The numbers are worth memorizing: tensile strength 1275 MPa (185 ksi), yield 860 MPa (125 ksi), elongation about 15%, and an elastic modulus E of 193 GPa (28×10⁶ psi). 301 has a high work-hardening rate, so it bends stiff and springs back hard - exactly what you want in a snap clip. Corrosion resistance covers most underhood and indoor applications without plating, which removes a whole process step from the cost.

17-4PH - When Load Beats Everything

17-4PH is a precipitation-hardening martensitic stainless steel, and in the H900 condition it carries serious load: tensile 1310 MPa (190 ksi), yield 1170 MPa (170 ksi), hardness 40-44 HRC, modulus around 197 GPa. That is the highest yield of the alloys in this guide, which is why 17-4PH clips are used in corrosive or elevated-temperature environments where a 301 clip would relax into uselessness. The cost is formability: 17-4PH is stamped in the annealed condition and precipitation-hardened after forming, which adds a heat treat step and a dimensional check after treatment.

Beryllium Copper C17200 - The Contact Spring King

C17200 in spring temper TD04 is the reason gold-plated connector contacts work. Tensile 1140 MPa (165 ksi), yield 965 MPa (140 ksi), modulus 128 GPa (18.6×10⁶ psi), and electrical conductivity 22-25% IACS. That single combination - high strength plus real conductivity plus low relaxation - does not exist in any stainless. Its bending endurance limit lands around 230–100 MPa and it clears 10⁷ cycles in standard fatigue tests. When a contact must both carry signal and hold mating force through temperature cycles and millions of matings, C17200 is the benchmark, and it is the reference material in our connector terminal stamping guide.

Phosphor Bronze C51900 - The Current-Carrying Compromise

Phosphor bronze C51900 (8% tin) is the lower-cost current-carrying spring alloy. In hard temper it delivers tensile around 700 MPa, yield around 570 MPa, a modulus of roughly 110 GPa and conductivity near 13% IACS. It does not match BeCu on strength or conductivity, but it costs a fraction as much, relaxes acceptably at moderate temperatures, and solders and welds well, which is why it dominates terminals and clips where the spring function is modest. It is the standard choice for many stamped terminal springs, detailed in our phosphor bronze stamping guide.

Spring Steel - High Load, No Conductivity

High-carbon spring steels such as 1075 and 1095 deliver the highest strength per dollar of any spring strip, regularly above 1300 MPa in hardened and tempered strip form. They are the right choice for heavy leaf springs, latches and mechanical springs that never carry current and operate in dry, protected environments. The catches are corrosion (they rust without plating or oiling) and relaxation at elevated temperature, which limits them to moderate thermal duty. Where a steel spring must resist corrosion, 301 stainless is the usual upgrade, as covered in our stainless steel stamping guide.

The full material landscape for stamped parts, including formability and plating notes for each family, is in our metal stamping materials guide.

Stress Relaxation - The Silent Drift

Relaxation is loss of clamp force under constant deflection at temperature. Do not confuse it with creep; creep is constant load with growing strain, relaxation is constant strain with falling stress. A battery contact held at 0.4 mm deflection at 85°C for ten years is a relaxation problem, full stop. Published relaxation curves are blunt about the alloy gap: at 150°C, 301 half-hard sheds 25-30% of its initial stress over the first 1000 hours, while C17200 loses only a few percent in the same window. If the spring holds a critical clamp force for the life of the product, the alloy is chosen on relaxation data, not on strength tables.

Designing against relaxation means three things. First, keep the operating stress well below yield: relaxation rate climbs steeply as initial stress approaches the yield point, so a spring run at 80% of yield relaxes dramatically faster than one run at 50%. Second, know the temperature: every 25°C increase roughly doubles the relaxation rate for stainless and copper alloys, so the qualification test must run at the worst-case service temperature, not room temperature. Third, verify with a long-duration test: the standard approach is to hold a sample at the service temperature and deflection for 1000 hours or more, measure the retained force, and compare it against the design limit. A supplier that cannot show relaxation test data for the alloy and temperature you need is guessing, and a spring program should not run on guesses.

Fatigue Life and the Goodman Check

Spring failure in service is usually fatigue, not overload. A leaf spring in a latch cycles between zero and full deflection a few hundred thousand times a year, and a constant-force spring sees that load on every extension. You size against the endurance limit with a Goodman or Gerber correction for mean stress. For BeCu C17200 the bending endurance limit sits near 250 MPa; hold a factor of safety on top of that and you design for 10⁷ or more cycles with margin. For 301 stainless the endurance limit in bending is lower as a fraction of tensile, so the operating stress must be derated accordingly.

The Goodman line is the practical tool: it connects the endurance limit at zero mean stress to the yield strength at zero amplitude, and any operating point below the line is safe for infinite life. The two inputs that matter are the true stress amplitude at the critical section and the mean stress from the preload. Stamped springs carry residual stress from forming - the bend sets the material beyond yield - and that residual stress acts as a beneficial mean stress on the inside of the bend, which is why a well-designed stamped clip can outlast a nominally identical machined part. Getting the residual stress direction right requires knowing which side of the bend is in tension in service; get it backwards and the part fails early. Fatigue testing on a sample population, not just calculation, is the only way to confirm the margin, and it belongs in the PPAP-style qualification file for any cycled spring.

Spring Geometry, Springback and Tolerances

A leaf spring is the simplest stamped spring: a cantilever or a doubly-clamped strip whose deflection follows δ = PL³/(3EI) for a cantilever. The lever you actually have is thickness cubed - double the strip thickness and you get eight times the stiffness. That is why clip engagement force is so sensitive to blank thickness tolerance; hold ±0.01 mm and your force band stays tight, let it wander and the same die produces parts that range from too stiff to too soft. Spring rate, engagement force and deflection all trace back to thickness, which is why the blank is bought to a thickness tolerance, not just a nominal.

Springback is the second geometry lever. Every bend in a stamped spring springs back when the punch lifts: 301 and BeCu spring back more than steel, and the amount varies with the coil lot's yield strength. Die design compensates by overbending - the die angle is set past the target so the part lands on spec after springback - and the compensation is developed against measured springback on the actual material. In production, the control plan watches bend angles because coil-to-coil yield variation shifts springback. A coining or bottoming operation at the bend can stabilize the angle by flattening the material at the apex, at the cost of a small work-hardened zone.

Constant-force springs are the other signature stamped spring: a pre-stressed rolled strip that, once the free end is pulled past its natural radius, delivers near-constant load across most of its extension. The classic uses are retracting a printer carriage, counterbalancing a monitor arm and feeding a label web. The strip is 301 or BeCu, 0.05-0.25 mm thick, wound around an arbor sized to hold the natural curvature without yielding the strip. The spring rate is effectively zero across the working range, which is exactly the point: the load stays flat, so the mechanism does not need a compensating cam or linkage. The manufacturing challenge is consistency of the pre-stress, because the constant-force behavior comes from the strip being uniformly bent past yield, and any variation shows up as a wandering load.

Plating and Surface Finish

For contact springs the plating is half the design. Bare BeCu oxidizes and its contact resistance climbs, so you plate it. The standard stack is a 1.25-2.5 μm nickel strike underplate followed by 0.76-2.5 μm (30-100 μin) of hard gold on the contact area, which holds contact resistance at 1-2 mΩ through the life of the part. Tin at 2-5 μm is the cost play for non-gold applications and solders cleanly, but tin creeps under load and its contact resistance is higher, so it belongs on terminals, not on high-cycle mating contacts. Silver is the middle option where conductivity matters and the environment stays clean.

Selective plating matters as much as the coating itself. Reel-to-reel selective plating deposits gold only on the contact zone, which keeps the precious metal bill proportional to the function instead of the part area. The plating must survive the stamping and forming steps or be applied after forming, depending on the geometry; pre-plated strip is cheaper but the plating cracks at tight bends, while post-plate protects the formed geometry at higher cost. For contact springs, post-plating or selective strip plating with the contact zone positioned correctly is the reliable route, and the plating thickness is verified by X-ray fluorescence on a sampling plan tied to the control plan. FOD and cleanliness also matter: a stamped spring destined for a sealed connector must be free of loose burrs, lubricant residue and metal fines, because a single particle can bridge contacts in a high-impedance circuit.

Manufacturing, Tolerances and Quality Control

Stamped springs are made on high-speed progressive dies, and the process window is set by three numbers: the press speed, the die accuracy and the material consistency. A high-speed line running up to 300 SPM with ±0.005 mm positioning accuracy produces springs whose geometry is set by the die, not by operator adjustment. The material enters as coil strip within a tight thickness tolerance, the die holds feature position, and the press feeds precisely; if all three are right, millions of parts are identical within the SPC band.

Quality control on springs is functional, not just dimensional. First article inspection measures geometry on CMM or optical equipment and tests force at the critical deflections. Production sampling repeats the force test at the control plan frequency, because force is the characteristic that catches thickness drift and material temper shifts before dimensions do. Life testing on a sample population - cycling to the required cycle count, holding at temperature for relaxation - qualifies the design, and the results become part of the submission file for automotive or medical programs. A spring supplier should be able to show, for any part number, the force-at-deflection distribution, the relaxation and fatigue evidence, and the plating thickness records.

In-house tooling is what keeps springs repeatable across the life of the program. The tool room builds the die, develops the springback compensation, and maintains the cutting edges that control burr; when the die is sharpened or a station is repaired, the change is documented and the first article is re-verified. The quality system around all of this - measurement lab, SPC, calibration, traceability - is described on our quality page, and it is the same discipline that applies to any precision stamped part, spring or not.

Get the Alloy and the Geometry Right the First Time

If you are sourcing stamped springs and the failure modes above sound familiar, send us the drawing and the operating envelope - temperature, cycle count, load, environment and any plating spec. We will run the material and geometry trade-off against real published data, not a guess, and come back with a stamped prototype in the right alloy, with the relaxation and fatigue check built into the design review. That is what separates a spring that drifts out of spec at hour 5,000 from one that holds its force for the life of the product.

Request a quote for your stamped spring program and include the operating envelope; we will confirm the alloy, the geometry and the qualification test plan before you commit to tooling.