How Much a Foil Wrap Blocks a Car Key Fob's Signal

Rand Simulation — Applications Engineering AI
Automotive · Ansys HFSS · 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.

Two panels of the same modeled car key fob: bare, with its solved far-field radiation lobe blazing around it, and sealed in aluminum foil, where the lobe is gone — The same modeled key fob, drawn at the same scale. Left: the bare fob, with the *solved* far-field pattern of its little antenna radiating out — the short-range chirp a relay-theft rig grabs and pipes to your car. Right: the same fob sealed in aluminum foil. At the fob's radio band the emission falls to the solver's numerical floor (around 200 dB down), so at this scale there is no lobe left to draw. Solved in Ansys HFSS.

The short version: a security article tells you to drop your car key in a foil pouch, a cookie tin, or a “Faraday” wallet so thieves cannot relay its signal. We modeled the fob as the small antenna it really is and solved its shielding effectiveness — the decibel ratio of how much signal escapes with the wrap versus without it — in eight full-wave Ansys HFSS runs at the 315 MHz (North America) and 433 MHz (Europe) fob bands. Sealed, every wrap is a strong shield: aluminum foil drops the signal below what the solver can even measure — to its numerical floor, around 200 dB — and a proper pouch holds about 103 dB. The surprise is how gracefully it fails when the wrap is imperfect: a 100 mm fold left open in the foil still blocks about 82 dB, and where the gap sits matters more than how big it is — a cookie tin's ill-fitting lid seam leaks the most of anything we tried, down to 53 dB. But none of that is the whole story of relay theft, and the honest answer needs the other radio in the system.

A Modern Car Key Is Two Radios, and Foil Only Answers One

A keyless-entry attack is a relay: two thieves, two boxes, one link between them. One box stands by your car and impersonates the fob; the other stands by your house and impersonates the car, waking your key and passing its reply back down the line. That handshake runs on two very different radios. The car wakes the fob with a 125 kHz low-frequency magnetic pulse, and the fob answers with a 315 or 433 MHz radio chirp. A foil wrap, a tin, or a pouch is specified against that second link — the UHF reply — and that is the link a full-wave solver like HFSS is built to price. So lead with the nuance: this study shows what a wrap does to the fob's radio reply, and at that band a sealed wrap is genuinely strong. It does far less to the 125 kHz wake-up pulse, which is a different kind of physics entirely — and, as the scope box below works out, thin non-magnetic foil is a poor shield there. Foil quiets the echo; it barely touches the shout that starts the whole exchange.

What Shielding Effectiveness Means Here

Shielding effectiveness (SE) is the honest number the enclosure-shielding literature uses: solve the field radiated by the bare fob, solve it again with the wrap in place, and take the ratio in decibels. Higher is better — 20 dB means the escaping field is one-tenth as strong, 40 dB one-hundredth, and so on. It is a ratio, so the exact antenna does not matter as long as it is the same in every run; ours is a small strip dipole about 17 mm tip to tip, which at 433 MHz is only about a fortieth of a wavelength — electrically tiny, exactly like a real fob antenna. Each wrap is a thin conductive shell around it: aluminum for the foil, a mild-steel box for the tin, a good conductive fabric for the pouch. At these frequencies the metal itself is never the leak — aluminum's skin depth at 433 MHz is under four microns against a wall thousands of times thicker — so a perfectly sealed shell blocks essentially everything, and the whole story lives in the openings: the fold in a foil wrap, the seam under a tin lid, the overlap of a pouch flap.

Bar chart of shielding effectiveness in decibels for sealed foil, a steel cookie tin, and a Faraday pouch, at 315 and 433 MHz — How much each wrap blocks at the two fob bands, straight from the converged solves. Sealed foil is floor-limited — its leak falls below what the far-field export can resolve, so read it as “blocks everything.” The Faraday pouch, with only a small overlap gap, holds about 103 dB. The metal cookie tin is the weakest of the three, not because steel is a poor conductor but because its lid never makes an RF-tight joint — the seam we modeled leaks it down to the low 50s.

A Gap in the Foil: Does It Really Ruin the Protection?

This is the question the request really turned on, and the folklore answer is “yes, the smallest gap wrecks it.” The solves say something more interesting. We took the sealed foil wrap and cut a single opening along the top fold — 3 mm wide, and swept its length from 5 mm out to a hand-wide 100 mm — then solved SE at each size. The protection does fall as the opening grows, monotonically: from about 121 dB at a 5 mm nick down to about 82 dB at the 100 mm fold. But look at the numbers — even the worst case, a fold as long as the fob itself, still leaves more than 80 dB at the radio band. That is not a shield that has been “ruined.” It has been dented.

Why so much better than the back-of-the-envelope rule predicts? The textbook worst-case estimate for a slot, SE ≈ 20 log₁₀(λ/2L), treats the opening as an efficient antenna and puts the 100 mm fold at only about 11 dB. Our full-wave solve sits 70 dB above that line. The reason is physical and worth keeping: that formula is a pessimistic bound for a slot near its half-wave resonance, and our openings are nowhere near it — a half-wave slot at 433 MHz would be 346 mm long. A narrow, sub-resonant fold, with the fob's electrically-tiny antenna sitting right under it where the coupling field is weakest, is a genuinely poor leak. The gap costs real decibels, and the trend is exactly what aperture theory predicts, but at the fob band the failure is graceful, not the sudden collapse the folklore promises.

Shielding effectiveness versus fold-opening length for the foil wrap, the full-wave solve well above the textbook worst-case slot bound — The money chart: shielding effectiveness against the length of the opening cut in the foil, at 433 MHz. The full-wave solve (orange) falls smoothly as the fold grows but stays far above the pessimistic textbook slot bound (gray dashed) — the two agree on the trend, not the magnitude, because a narrow sub-resonant fold over a tiny antenna is a much worse radiator than the worst-case formula assumes. Sealed foil sits off the top of the chart, below the solver's floor.

Where the Gap Sits Beats How Big It Is

The tin is the tell. Its leak — a 0.5 mm seam under an ill-fitting lid, on the side wall — is far narrower than the 3 mm foil fold, yet it shields the least of everything we ran, about 53 dB at 433 MHz against the folded foil's 82. A thinner opening that leaks more sounds backwards until you notice where each one sits. The foil fold runs across the top, directly over the fob's antenna, where a small vertical antenna radiates least. The tin's seam runs along the side, broadside to the antenna, where the field it couples to is strongest. Position, not size, sets which opening actually leaks — the same lesson that makes real enclosure design about seam placement and gasket lines rather than raw wall thickness. It is also the honest caveat on the graceful-failure result above: a clean fold centered over the antenna is close to a best case, and a gap in the wrong place would cost more.

Shielding effectiveness across the 0.30 to 0.45 GHz band for the foil folds, the cookie tin, and the Faraday pouch — Shielding effectiveness across the fob band for every leaky variant. The four foil folds form the blue family, light to dark as the opening grows; the Faraday pouch (purple) sits with the tighter folds; the cookie tin (orange) trails everything. All of them are nearly flat across the band — there is no in-band resonance to fall into here, because every opening is far shorter than a half-wavelength.

So — Does Foil Stop Relay Theft?

At the fob's radio band, a wrap that actually stays closed is a strong shield, and even an imperfect one blocks a great deal — so the part of relay theft that reads the fob's reply is genuinely hard to do through foil. But two honest qualifiers decide whether that translates into a safe car. First, a loose kitchen-foil fold is not a purpose-built pouch. Our numbers assume the wrap holds a clean, closed shape around the fob; a scrap of foil that springs open in your pocket, or a tin whose lid rides up, is a different object than the sealed shell that scores ~200 dB at the solver's floor. A Faraday pouch or a well-latched tin earns its dB because it stays closed. Second, and more important, foil answers the wrong radio. The 125 kHz wake-up pulse that starts a relay attack is a low-frequency magnetic field, and — as the scope box shows — thin non-magnetic aluminum barely attenuates it. A ferromagnetic steel tin does better there, one of the few places its material actually helps. If you want a car key silenced, the reliable move is the purpose-built pouch or a snug metal tin, closed every time — not a loose square of foil.

Honest scope. This is a physics demonstration on a self-authored, brand-free key fob and generic foil / cookie-tin / Faraday-pouch shells, not a compliance test of any product. We solved the UHF response leg only (315 and 433 MHz), in eight converged Ansys HFSS driven-modal runs (adaptive tetrahedral meshes, 1,956–6,384 elements per variant, ΔS convergence below target in 6–8 passes; 38,033 tetrahedra across the set, 4 cores each). Shielding effectiveness is the far-field ratio of each wrapped fob to the bare fob — the with/without-enclosure emissions definition — on the same small strip-dipole radiator, so it is a per-configuration transfer number, not an absolute emission. The 125 kHz low-frequency wake-up leg is deliberately not solved: at 125 kHz the wavelength is about 2.4 km, a magneto-quasistatic near-field regime outside full-wave HFSS. We bound it by hand instead: aluminum's skin depth at 125 kHz is roughly 0.23 mm against a ~0.016 mm foil, about a fifteenth of a skin depth, so a thin non-magnetic foil absorbs well under a decibel and, having no permeability, reflects almost none of the magnetic field — a poor LF shield. A ~0.25 mm steel wall (μ_r in the hundreds) is several skin depths plus magnetic flux-shunting, so a ferromagnetic tin does meaningfully better at LF; that is stated, not solved. We do not model the relay hardware or any car receiver's sensitivity, so we report decibels escaping the wrap, not a go/no-go against a specific attack rig. Real wraps also leak at closures and creases that a clean modeled shell does not, which is exactly why a loose fold underperforms the sealed number.

Designing a shield, an enclosure, or an antenna that has to keep a signal in — or out? The decibels are decided by geometry: where the seams sit, how narrow the gaps are, whether the wrap actually closes. A full-wave solve prices every one of those before you cut metal or fabric. We run electromagnetic shielding and antenna studies in Ansys HFSS, geometry through validated results. Rand Simulation — innovation through insight.