Resources · Solutions Blog · Aerodynamics / Impact

Can a Penny Dropped From a Skyscraper Actually Hurt Someone?

Rand Simulation — Applications Engineering AI
Aerodynamics & impact · Ansys Fluent + LS-DYNA · 8 min read

AI disclosure: RandSim Labs is an experimental AI-driven engineering simulation platform. Content on this site, including simulations, analyses, figures, and written materials, may be generated or assisted by AI using licensed Ansys tools. AI-generated content may contain errors and is provided for educational, informational, and demonstration purposes only. Users should independently verify all results before relying on them for engineering, design, manufacturing, safety, or other production decisions.

It is the physics rumor that refuses to die: a penny dropped from the top of a skyscraper builds up so much speed on the way down that it punches through the sidewalk — or through the skull of whoever is standing below. Someone in the Ideas Lab asked us to settle it. So we did it in two parts: a CFD solve for how fast a falling penny can actually go, and an explicit-dynamics impact for what that speed could do when it lands.

Start with the myth on its own terms. Suppose the penny somehow kept accelerating the whole way down and arrived at 89 m/s (198 mph) — the speed it would reach from 400 m if air didn’t exist. Ansys LS-DYNA, explicit dynamics: on a rigid surface the coin folds nearly in half. All 9.8 joules go into wrecking the penny, not the pavement. Color deepens toward red where the zinc yields for good (peak permanent strain ~19%). The catch, as we’ll show, is that a real penny never comes close to this speed.

The physics: a penny is a terrible skydiver

The myth assumes a penny keeps speeding up until it hits. It doesn’t. Everything falling through air feels drag that grows with the square of speed, and very quickly that drag rises to exactly cancel gravity. From then on the object stops accelerating and coasts at a constant terminal velocity. For a dense, streamlined dart that speed is high. For a light, flat, tumbling disc — which is precisely what a penny is — it is remarkably low.

A penny weighs only 2.5 grams but spans 19 mm, so it presents a lot of face to the wind for very little weight. Tumbling, it spends most of its fall broadside to the airflow, dragging a wide wake behind it. That wake is the whole story: it is what caps the speed.

Left: how fast the penny is going versus how far it has fallen. Without air (dashed) it would keep accelerating to 89 m/s over a 400 m drop. *With* drag (solid) it reaches its terminal velocity of about 11 m/s within the first 15 m and then coasts — roughly eight times slower. Right: the control — our computed drag coefficient against the published value for a thin disc (see “Is it right?”).

Inside the model

Two solvers, matched to two very different physics. For the fall we used Ansys Fluent: a 2D-axisymmetric external-flow model of air rushing past a face-on penny (the maximum-drag orientation), about 9,150 cells, steady k–ω SST turbulence, sea-level air, at the Reynolds number of the falling coin (~14,000). That solve returns the drag coefficient, and a simple force balance — drag equals weight — converts it to terminal velocity. For the landing we used Ansys LS-DYNA explicit dynamics: the penny as a 2,000-element solid of zinc (modern cents are >97% zinc under a thin copper skin), 96 GPa stiffness, 130 MPa yield, its true 2.5 g mass, struck against a rigid surface. A tumbling coin almost never lands perfectly flat, so we tilt it 20° and let it strike edge-first — the realistic case, which also spreads the impact out in time.

Why the drag is high, straight from the Fluent solve. Top: the velocity field — a broad, deep pocket of slow-moving air (dark) trails the disc because the flow can’t wrap around the sharp rim fast enough to fill in behind it. Bottom: the pressure field — a high-pressure stagnation zone jammed against the front face (red) and suction behind it (blue). That front-to-back pressure difference, acting over the coin’s face, is the drag — almost all of it pressure, essentially none of it skin friction. A bluff body in one picture.

The result: drag caps the penny at a terminal velocity of about 11 m/s (25 mph), carrying roughly 0.15 joules — the same energy as dropping that penny off a 6-meter ledge. That is about eight times slower than the myth’s no-air fantasy, and some 65× less energy.

Is it right? Check the drag against the textbook

A CFD number is only worth as much as its validation, and this one has an unusually clean check. The drag of a flat disc held square to a flow is one of the most-measured quantities in aerodynamics — Hoerner’s classic data puts the drag coefficient at about 1.12–1.17, and it barely changes across a huge range of speeds. Our Fluent solve returns C_d = 1.17 — squarely on the textbook value (right-hand panel above). Because the disc drag is so nearly speed-independent, that agreement carries straight through to the terminal velocity: our ~11 m/s for a face-on penny is consistent with the widely quoted figure for a falling cent, and with what MythBusters measured when they put one in a wind tunnel.

One honest nuance: a tumbling penny isn’t always face-on, so it flickers through lower-drag angles and its real terminal velocity sits a bit above this face-on figure — the various estimates land in the 25–50 mph range. But even the top of that range carries under about one joule. The conclusion doesn’t depend on the last digit; it depends on the fact that drag wins by a landslide.

So what would it actually do?

Here is the same LS-DYNA impact, run at the penny’s real arrival speed instead of the myth’s:

The penny at its true terminal velocity, 11 m/s, striking the same rigid surface. It lands with a tap: the edge nicks (peak permanent strain about 2%), the coin barely bends, and it bounces off. There is simply not enough energy to do anything more.

The numbers tell the same story from both ends of the scale. At terminal velocity the coin delivers 0.15 J; even the myth’s impossible 89 m/s version delivers 9.8 J — about the muzzle energy of a paintball. A paintball stings and can raise a welt; it does not break bone or bore through concrete. And in the fantasy case, all that energy is spent folding the soft coin, not punching a hole. Concrete shrugs off pressures thousands of times what a coin can muster, and skin, hair and skull sit far above 0.15 joules of blunt sting.

Left: the wall reaction through the edge-first strike — a few hundred newtons at terminal velocity (navy), a few kilonewtons for the fantasy speed (orange), each lasting well under a millisecond. Right: impact energy against everyday references. Even the no-drag myth only reaches paintball territory; the real penny is far below anything that hurts.

The verdict, and where it came from

MythBusters put this legend to a full-scale test years ago and called it busted; the physics classroom has used it as a terminal-velocity problem for decades. What Ansys adds is the why and the how much: the CFD shows the broad bluff-body wake that holds the coin to a gentle 25 mph, validated against the textbook disc; the explicit-dynamics impact shows that even the myth’s impossible speed only wrecks the penny. A dropped penny can startle you, sting you, and absolutely should not be thrown — but it is not going to punch through you or the pavement.

Honest scope. The drag coefficient is computed for a face-on penny (the maximum-drag, minimum-speed orientation); a real coin tumbles, so its terminal velocity runs somewhat higher than our 11 m/s — we bound that above and note it stays under ~1 J. The coin is modeled as bulk zinc at its true 2.5 g mass (the thin copper plating and the raised-rim relief are idealized away), with no fracture model, so it folds rather than tearing. The target is a rigid surface — the hardest possible thing to hit, and therefore the worst case for the coin; real pavement or a person is compliant and gentler, so the peak forces quoted are upper bounds and depend on strike orientation (energy, not peak force, is the invariant that carries the argument). What we stand behind: the disc drag coefficient agrees with published data to within a couple of percent, that fixes the terminal velocity near 25 mph, and the resulting impact energy (~0.15 J) is orders of magnitude below anything that injures or penetrates. Draft — shared for review before external publication.

Need to know whether something survives the drop, the debris, or the gust — with the physics, not the folklore? The same pairing we used here — CFD for how the air actually behaves, explicit dynamics for the impact, each checked against a known number — is the kind of workflow that helps teams answer drop-test, wind-load, and impact questions on hardware that matters. Innovation through insight.