Ball Collision Hydrodynamics and Coefficient of Restitution

PRO CUE: The ball is not a passive projectile — it is a pressurized, deformable energy capsule that steals 40–50% of every joule you deliver into it. Elite power comes from understanding exactly where that energy goes during the 5-millisecond collision and how to harvest the restitution you are legally given.

Video: The Physics of Bouncing a Ball | Science Project (Science Buddies).

1. Mechanical Foundations: The Collision as a Hydrodynamic Event

Every shot in tennis terminates in a violent, sub-10-millisecond collision between a pressurized rubber shell and a tensioned string grid (or the court itself). At impact speeds above 30 m/s, the ball does not behave as a rigid sphere — it flattens into a pancaked disc, its internal gas column compresses like a piston, and its rubber membrane stretches and rebounds like a fluid membrane under pressure. This is why collision scientists describe the tennis ball using a hydrodynamic model: a pressure vessel whose stored energy is governed by gas thermodynamics, membrane elasticity, and viscous dissipation simultaneously.

During the collision, the ball's kinetic energy partitions into four transient reservoirs:

Compressible gas energy ($E_{gas}$): The internal pressurized air (≈2 atmospheres of overpressure in a new ball) compresses as the shell flattens, storing energy thermodynamically and returning most of it during decompression.
Membrane strain energy ($E_{membrane}$): The rubber shell stores elastic strain as it bends, stretches, and shears around the impact zone. Rubber is viscoelastic, so a significant fraction of this strain is never returned.
Stringbed strain energy ($E_{strings}$): When the collision partner is a racket, the string grid stretches and stores energy with extraordinarily low hysteresis — returning roughly 95% of what it absorbs.
Dissipated heat and acoustic energy ($E_{loss}$): The remainder leaves the system as heat (measurable as a 1–2 °C temperature spike in the ball), sound, and internal friction.

The governing principle of this entire article is the coefficient of restitution (COR) — the single scalar that compresses this entire hydrodynamic energy ledger into one number, and the number that ultimately fixes the ceiling on your ball speed.

MECHANICAL DEFECT WARNING: Players who "muscle" flat, compressed hits at maximum grip tension do not add energy to the collision — they only lengthen the deceleration phase of the racket and stiffen the interface where the ball must deform. Poorly timed collisions waste restitution; they do not create power.

2. The Coefficient of Restitution: Definition, Equations & ITF Calibration

The coefficient of restitution $e$ is defined as the ratio of separation velocity to approach velocity along the line of impact:

$$e = \frac{v_{\text{after}} - v_{\text{racket,after}}}{v_{\text{racket,before}} - v_{\text{before}}}$$

For a simple vertical drop test onto a rigid surface, this collapses to the square-root ratio of rebound height to drop height:

$$e = \sqrt{\frac{h_{\text{rebound}}}{h_{\text{drop}}}}$$

Because kinetic energy scales with velocity squared, the fraction of energy retained through the collision is $e^2$, and the fraction destroyed is $1 - e^2$. This quadratic relationship is the most important number in racket sports physics: a ball with $e = 0.75$ keeps only $0.75^2 = 56\%$ of its vertical impact energy and loses 44% to internal hydrodynamic losses.

The International Tennis Federation (ITF) calibrates every legal ball with a standardized drop test:

ITF Ball Specification Parameter	Required Value	Restitution Implication
Drop height (onto concrete plate)	254 cm (100 in)	Standardized incident velocity ≈ 7.0 m/s
Legal rebound window	135 – 147 cm (53 – 58 in)	Implies $e = 0.729 – 0.761$
Mass	56.0 – 59.4 g	Sets momentum ($m \cdot v$) delivered per impact
Diameter (Type 2)	6.54 – 6.86 cm	Fixes shell surface area and gas volume
Forward deformation (8.2 kg load)	5.6 – 7.4 mm	Indexes shell stiffness and internal pressure

The rebound window is deliberately narrow. Tournament balls must sit inside $e \approx 0.73 – 0.76$, which guarantees that roughly 42–47% of the collision energy is always destroyed inside the ball itself — before the strings or the court have taken their share. No racket technology, string setup, or swing mechanic can recover this loss; it is thermodynamically sealed inside the rubber.

3. Internal Hydrodynamics: Gas Compression, Rubber Hysteresis & the Energy Budget

Where exactly does the missing 44% go? High-speed internal studies of ball compression (up to 30% flattening at professional impact speeds) resolve the loss into three components:

Gas thermodynamic loss (≈5–10% of incident energy). The compressed air column heats as it compresses and cannot fully re-expand isentropically during the few milliseconds of contact. Part of the gas work is conducted into the rubber as heat rather than returned as pressure.
Viscoelastic hysteresis in the rubber shell (≈25–35%). Rubber's stress–strain loop is not closed: the stress curve during loading sits above the curve during unloading, and the area between them is heat. This is the single largest sink in the collision. It is why a ball struck repeatedly at high speed warms measurably and why "dead" balls lose rebound height long before they lose pressure.
Membrane wave and vibration loss (≈5%). The impact launches elastic waves that rattle around the shell and decay before they can contribute to rebound velocity.

The Hydrodynamic Energy Ledger (Vertical Drop, $e = 0.75$)

Energy Pathway	Fraction of Incident Energy	Recoverable?
Gas compression & re-expansion	~15% stored	Mostly yes (quasi-elastic)
Rubber membrane strain	~30% stored	~50% of it returns; rest is hysteresis heat
String/court elastic storage	Separately counted	Strings ≈ 95% return; court ≈ 60–70%
Hysteresis + acoustic + wave loss	~40–47% destroyed	No — permanent loss
Total rebound energy ($e^2$)	~53–58%	Defines the legal COR window

A critical hydrodynamic insight follows: because gas pressure contributes the most elastic (least lossy) storage pathway, a fully pressurized ball at 23 °C rebounds measurably higher than the same ball at 10 °C, and a pressureless ball — which stores its energy almost entirely in thick, hysteretic rubber — sits at the low end of the COR window by design.

4. The Coupled Restitution System: Ball × Stringbed

When the collision partner is a racket rather than the court, the two restitution systems couple in series. The stringbed acts as a low-hysteresis spring that shares the deformation load with the ball: for a given closing velocity, a softer stringbed lets the strings deform more and the ball deform less. Since strings lose only ~5% of stored energy while rubber loses ~45%, transferring deformation from ball to strings is the single most legal power gain available in equipment setup.

The combined rebound follows the two-body restitution equation. Treating the racket as a mass $M$ with an effective mass at the contact point $M_{\text{eff}}$ (typically 0.12–0.17 kg for a midplane groundstroke hit, because the racket recoils and rotates about the wrist), the outgoing ball speed is:

$$v_{\text{out}} = \frac{(1+e)\,M_{\text{eff}}\,v_{\text{racket}} + (m - e\,M_{\text{eff}})\,v_{\text{in}}}{m + M_{\text{eff}}}$$

Worked Example: Elite Rally Groundstroke

Parameter	Value	Comment
Racket speed at contact ($v_{\text{racket}}$)	30 m/s (108 km/h)	Peak kinetic chain output (Art. 008, 019)
Incoming ball speed ($v_{\text{in}}$)	−27 m/s	Opposite sign: closing collision
Effective racket mass ($M_{\text{eff}}$)	0.15 kg	Midplane impact, relaxed wrist
Ball mass ($m$)	0.058 kg	ITF legal range
Combined COR ($e$)	0.75	Ball (0.73–0.76) conditioned by stringbed
Outgoing ball speed	≈ 45 m/s (162 km/h)	Before spin and air-drag penalties

The formula exposes three levers, in descending order of player control:

$v_{\text{racket}}$ dominates. Racket-head speed is multiplied by $(1+e) \approx 1.75$ — the physical reason the entire kinetic chain of Pillar I exists.
$M_{\text{eff}}$ is tunable through contact geometry and grip. A hit at the node near the throat, a death-grip, or a floppy wrist each drain effective mass and leak the collision (Art. 007, 012, 021).
$e$ is nearly fixed by ITF ball law — but its stability is not. String tension, gauge, and temperature shift it within a usable band, which is the subject of Sections 7 and 8.

MECHANICAL DEFECT WARNING: Off-center hits (outside the 15 cm "power valley" around the stringbed node) do not merely lose power — they add racket torque, drain $M_{\text{eff}}$, and shock-load the arm (Art. 020, 021). The collision is only hydrodynamically efficient inside the sweet zone.