Shot Peening

How shot peening works

Take a piece of steel and hit one spot with a tiny round ball, hard enough to leave a dent. Right under that dent the metal has flowed — it has yielded plastically and stretched sideways. But the material a millimetre below didn't yield; it just deformed elastically and now wants to spring back to its original size. It can't, because the stretched skin is in the way. So the bulk clamps down on the dimpled skin and holds it in compression. Repeat that across the whole surface, dimple overlapping dimple, and you build a continuous, uniform layer of compressive residual stress a few tenths of a millimetre deep.

That is the entire trick of shot peening. Nothing is added to the part — no coating, no plating, no extra material. You simply rearrange the internal stress state of a thin surface layer so that, before any service load is applied, that layer is already squeezed. The mechanism that drives fatigue — a tensile crack opening cycle after cycle — now has to fight its way out of compression before it can even begin.

Why compression matters so much: fatigue cracks initiate at the surface (that's where stress is highest in bending and torsion, and where defects live) and they propagate only when the crack tip is pulled open by tension. A crack tip sitting in residual compression stays shut. So the figure of merit is simple — the more compressive the surface, and the deeper that compression extends past the zone where cracks nucleate, the longer the part lives.

The media itself is the other half of the story. Peening shot must be round and of controlled size and hardness — cast steel shot, conditioned cut wire, ceramic beads, or glass beads. Roundness is non-negotiable: a sharp or broken particle gouges the surface and leaves a stress concentration, which is precisely the kind of crack starter you came to eliminate. Broken shot is continuously screened out of the recycling loop.

The governing physics

The residual stresses through a peened section must balance — they cannot create a net force or moment out of nothing. So the through-thickness profile always has the same shape: high compression at and just below the surface, crossing zero, then a small balancing tension in the bulk.

Through-thickness equilibrium (no external load):

  ∫ σ_res(z) dz = 0        (forces balance)
  ∫ σ_res(z)·z dz = 0      (moments balance, for a free part)

Typical steel profile:
  surface           σ ≈ -0.3 to -0.5 × UTS   (compression)
  peak (just below) σ ≈ -0.5 to -0.6 × UTS   (max compression)
  depth to σ = 0    z₀ ≈ 0.1 to 0.5 mm
  bulk              σ ≈ +0.05 to +0.1 × UTS  (balancing tension)

The effect on fatigue is captured by treating residual stress as a shift in mean stress. Fatigue strength depends on both the cyclic amplitude σ_a and the mean stress σ_m; the Goodman relation expresses how a more compressive (more negative) mean stress lets a part survive a larger alternating stress:

Goodman line:    σ_a / σ_e  +  σ_m / σ_uts  =  1

  σ_a   = alternating (cyclic) stress amplitude
  σ_m   = mean stress  (service mean + residual stress)
  σ_e   = fully-reversed endurance limit
  σ_uts = ultimate tensile strength

Peening makes σ_m more negative → allowable σ_a rises.

The intensity of the process is the kinetic energy each particle delivers, which sets the depth and magnitude of the compressed layer. For a single impact the energy scales as ½mv², so for a particle of diameter d and density ρ striking at velocity v:

Single-impact kinetic energy:

  KE = ½ m v² = ½ · (π/6 · ρ d³) · v²

  → larger shot (d) drives a DEEPER compressive layer
  → faster shot (v) drives HIGHER peak compression
  → both raise Almen intensity

Because the depth scales with shot diameter and the peak with velocity, engineers tune the two media size and air pressure (or wheel speed) independently to hit a target profile, then verify the result with an Almen strip rather than trusting the inputs.

Almen intensity: measuring an invisible stress

You cannot see residual stress, so the industry measures it indirectly with the Almen strip — a small spring-steel coupon (SAE 1070) clamped flat and peened on one face only. The compressive layer on the peened side makes the strip arch into a convex curve once unclamped; the height of that arc, read on an Almen gage, is the proxy for how much compression went in.

Peen identical strips for increasing exposure times, plot arc height versus time, and you get a saturation curve that climbs steeply then flattens. Almen intensity is formally defined as the arc height at the point where doubling the exposure time produces no more than a 10% increase in arc height (per SAE J443). That saturation point is repeatable and machine-independent, which is why it — not air pressure or time — is the spec.

Reading an intensity callout:   0.012A

  0.012  = arc height in inches (12 thousandths)
  A      = which Almen strip thickness
           N strip ≈ 0.031 in (light intensities)
           A strip ≈ 0.051 in (most common)
           C strip ≈ 0.094 in (heavy intensities; 1 C ≈ 3.5 A)

Saturation definition (SAE J443):
  T  = exposure time at the curve knee
  2T gives ≤ 10% more arc height  →  T is "saturation"

Three strip thicknesses cover the working range. Light work (thin springs, electronics, aluminum) uses N strips; most aerospace and automotive work uses A strips; very heavy peening (large gears, thick forgings) uses C strips. A handy conversion: a reading of about 1 unit on a C strip corresponds to roughly 3.5 units on an A strip.

Coverage and the saturation rule

Intensity tells you how hard you hit; coverage tells you whether you hit everywhere. Coverage is the fraction of the surface dimpled at least once. Because shot lands at random, the last few percent of bare surface take a wildly disproportionate amount of time to cover — the math follows the Avrami (Poisson-process) equation:

Coverage growth (Avrami / Poisson):

  C(t) = 1 - exp(-k t)

  t      = exposure time
  k      = impact rate × dimple area
  C → 1  only asymptotically (never literally 100%)

Convention:  98% measured coverage  =  "100% (full) coverage"
Spec often requires 200%  =  twice the time to reach full

Because you can never truly reach 100%, the industry defines 98% visually-measured coverage as "full coverage" (100%). Specs then call for multiples — 150%, 200%, even 400% — meaning the part is peened for two, three, or four times the time that first reached full coverage, guaranteeing no soft, untreated window survives where a crack could nucleate. Coverage is checked at 10× magnification or, more reliably, with a fluorescent tracer dye that wears off only where shot has struck (the Peenscan / Valley method).

Real-world process specs

Application	Media	Typical intensity	Why
Automotive coil / valve springs	Cut wire / cast steel	0.008 to 0.014A	Springs cycle billions of times; peening is mandatory for life
Leaf springs (truck/trailer)	Cast steel shot	0.012 to 0.020A	High bending load; often stress-peened under preload
Carburized gears (gear roots)	Steel shot S230 to S330	0.012 to 0.020A	Adds to case compression; doubles to triples root bending life
Aircraft landing gear (300M steel)	Steel shot, high intensity	0.012 to 0.024C	Ultra-high-strength steel is brittle; compression fights crack growth
Turbine / compressor blades	Glass / ceramic beads	0.004 to 0.010N	Light intensity, low roughness, no iron contamination on Ni alloys
Fan-blade leading edges (LSP)	Laser pulse (no media)	1 to 2 mm deep	Foreign-object-damage tolerance needs deep compression
Wing skins (peen forming)	Large steel shot	Tuned to bend, not just stress	Compression bends the panel to an aerodynamic contour

Shot peening vs other surface treatments

	Shot peening	Shot blasting / grit blasting	Laser shock peening	Case hardening	Roller burnishing
Primary goal	Compressive residual stress	Clean / roughen surface	Deep compressive stress	Hard, wear-resistant case	Smooth + compress surface
Media / energy	Round controlled shot	Angular grit (cutting)	Laser pulse + ablative layer	Carbon / nitrogen diffusion	Hardened roller pressure
Compression depth	0.1 to 0.5 mm	Negligible / random	1 to 2 mm	0.5 to 2 mm (case)	0.1 to 0.5 mm
Surface finish effect	Roughens slightly (dimples)	Roughens heavily	Minimal roughening	Unchanged by process	Smooths (mirror)
Geometry coverage	Complex shapes, fillets, roots	Line-of-sight	Line-of-sight, slow, costly	Whole part (furnace)	Rotationally symmetric only
Relative cost	Low	Lowest	Very high	Moderate	Low to moderate
Typical use	Gears, springs, shafts, blades	Paint prep, descaling	Aero fan blades, welds	Gears, cams, bearing races	Crankshaft fillets, hydraulic rods

Worked example: a peened valve spring

Consider an automotive engine valve spring made from oil-tempered chrome-silicon wire (σ_uts ≈ 1900 MPa). At 6,500 RPM the spring cycles about 54 times per second, so over a 200,000 km / 4,000-hour life it sees roughly 8 × 10⁸ cycles — deep in the high-cycle fatigue regime where any surface flaw is fatal. Run the numbers on what peening buys:

Spring wire UTS:           σ_uts = 1900 MPa
Peening intensity:         0.010A, ~150% coverage
Peak compressive stress:   σ_res ≈ -0.55 × 1900 ≈ -1045 MPa
Compressive layer depth:   ~0.20 mm

Unpeened torsional endurance limit:   ~620 MPa
Peened torsional endurance limit:     ~800 MPa   (+29%)

Fatigue life at the same service stress:
  unpeened   ≈ 2 × 10⁸ cycles  (fails before target)
  peened     ≈ 1 × 10⁹ cycles  (survives — ~5× longer)

That roughly 5× life multiplier is why peening is not optional on a production valve spring — an unpeened spring simply will not survive the engine's design life. Stress peening (peening while the spring is compressed under load) pushes the benefit further by biasing the residual stress in the direction that opposes the service load, and is standard for the highest-duty springs.

Where shot peening is used

• Springs. Coil, leaf, valve, and clutch springs are almost universally peened. The cyclic torsional or bending stress is high and unrelenting; peening is the cheapest large lever on fatigue life. Stress peening (under preload) is common on the hardest-working springs.
• Gears. Carburized gear teeth are peened in the tooth root fillet, where bending fatigue cracks start. Peening stacks compressive stress on top of the carburizing case and reliably doubles or triples root bending fatigue life — critical in aerospace and motorsport transmissions.
• Aero engines and airframes. Compressor and turbine blades, disks, shafts, and landing gear are peened. High-strength steels like 300M (σ_uts ≈ 1900 MPa) used in landing gear are notch-sensitive and peening is mandatory. Laser shock peening protects fan-blade leading edges against bird-strike and foreign-object damage.
• Crankshafts and connecting rods. Fillet rolling or peening at the journal fillets is a standard durability measure in production and racing engines.
• Welded structures. The weld toe is a fatigue hotspot — high tensile residual stress plus a sharp geometric notch. Peening (shot or needle/hammer peening) the toe converts that tension to compression and is a recognized fatigue-improvement method in offshore and bridge structures.
• Peen forming. The same compressive layer that protects a thick part will bend a thin one. Aircraft wing skins are deliberately peened on one face to curve them into their aerodynamic contour — no dies, no press, just controlled residual stress.

Common misconceptions and pitfalls

• "More peening is always better." Wrong, and dangerously so. Past saturation, extra impacts fold and tear the surface, raising roughness and seeding microcracks (laps and folds). Over-peening can give worse fatigue life than no peening at all. Both intensity and coverage are capped, not just floored.
• Confusing peening with blasting. Sandblasting and grit blasting use sharp angular media to cut and clean — exactly the opposite of what you want. Sharp impacts create stress raisers. Peening media must be round; broken shot is screened out continuously.
• Ignoring distortion of thin parts. The compressive layer that simply strengthens a thick part will warp a thin one (the basis of peen forming). Thin sections, especially when peened on one side, can bow out of tolerance. Either peen both sides, support the part, or design for the deflection.
• Thermal relaxation of the residual stress. Compressive residual stress is not permanent if the part runs hot. Above roughly 230 °C for many steels — and lower for aluminum — the locked-in stress relaxes over time, eroding the fatigue benefit. Hot-section components need this checked against service temperature.
• Treating intensity inputs as the spec. Air pressure, wheel speed, and time drift as nozzles wear and shot degrades. The controlled variable is Almen intensity verified on strips, plus measured coverage — never the machine setpoints alone.
• Iron contamination on non-ferrous parts. Steel shot embeds iron particles that corrode aluminum, titanium, and nickel alloys. Those materials are peened with glass or ceramic beads, or with stainless/conditioned-cut-wire media, to avoid galvanic and corrosion problems.