How Electronic Pest Repellers Work
Start with the hearing gap
Human hearing tops out somewhere between 15 and 20 kHz in young adults, and declines with age — most adults over 30 hear very little above 17 kHz. Rodents hear far higher: mice are sensitive well past 70 kHz and communicate ultrasonically as a matter of course. Many insects detect vibration and high-frequency sound through specialised organs rather than ears.
That gap is the whole basis of the category. A sound loud enough to make a space genuinely unpleasant for a mouse can be entirely inaudible to the people living there.
Engineering note
The gap is real, and it's the reason these devices are physically plausible. It doesn't by itself prove any given product is effective — that depends on output, frequency behaviour and, above all, whether the sound reaches the animal at all.
Technology 1 — Ultrasonic sound
A piezoelectric transducer converts an electrical signal into mechanical vibration and radiates it as sound above the human hearing threshold, typically in the 20–65 kHz band.
For a rodent, sustained exposure at effective intensity is aversive in a specific way: it interferes with the acoustic environment they rely on for navigation and communication. The animal doesn't experience pain — it experiences a place that feels wrong, and its instinct is to route around it. That's what deterrence is. Nothing dies.
Where it stops. Higher frequency means shorter wavelength, and shorter wavelength means the sound behaves more like light and less like bass. It travels in straight lines, reflects off hard surfaces, is absorbed by soft ones, and does not meaningfully pass through solid partitions.
A concrete or plasterboard wall is a boundary, not an obstacle. So is a closed door.
Treat coverage the way you treat a lamp. Whatever the device can "see", it protects. What it can't see, it doesn't.
Technology 2 — Electromagnetic pulsing
Electromagnetic devices use the building's existing wiring as a distribution network, pulsing a field along the cable runs inside walls and under floors.
The intended advantage is exactly where ultrasound is weakest: the cavities, voids and under-floor spaces that rodents use as motorways and that no airborne sound can reach.
Be sceptical of the claims, and here's the honest framing. Independent evidence for electromagnetic repellency is considerably thinner than for ultrasound, and the strength of the field diminishes rapidly with distance from the cable. It's best understood as a supplementary layer that may add pressure in wiring runs — not as a whole-house solution, and not as a substitute for ultrasound in the room itself.
Any product marketed on electromagnetic performance alone, with claims of whole-building coverage, deserves your scepticism.
Technology 3 — Seismic vibration
Airborne sound is close to useless against burrowing animals. Moles, voles and gophers live in a medium that transmits vibration extremely well and airborne sound extremely badly.
Vibration-based outdoor units invert the approach: a motor inside a stake driven into the ground generates low-frequency mechanical pulses that propagate through the soil itself. To a burrowing animal, the ground reads as unstable and occupied.
Soil type changes everything. Dense, moist, clay-rich soil transmits vibration well and gives a large effective radius. Dry, sandy or very loose soil damps it quickly and the radius shrinks substantially. Any manufacturer quoting a single coverage figure without mentioning soil is quoting a best case.
The strongest outdoor systems combine both layers: vibration into the ground for burrowers, ultrasonic sweep above ground for surface visitors.
The mechanism that separates real devices from cheap ones: frequency sweeping
This is the single most useful thing to understand before buying anything.
Habituation is the nervous system's basic efficiency mechanism: a stimulus that repeats identically and predictably, without consequence, gets filtered out. You stop hearing your own refrigerator within a day of moving in. The same process happens to an animal exposed to a fixed tone — and it happens within days, not months.
This is why so much of the category has a bad reputation. A device emitting one unchanging frequency is, biologically speaking, designing its own obsolescence. It works for a week and then becomes furniture.
Sweeping is the answer. Instead of holding one pitch, the device moves continuously and unpredictably across a frequency band — for example 25 to 40 kHz — changing many times per minute. There's no stable pattern to filter out, so the nervous system can't file it as background.
When you read a spec sheet, this is the line that matters most. "Ultrasonic, 25 kHz" describes a device that will stop working. "Sweeping, 25–40 kHz, variable interval" describes one designed for month twelve, not week one.
If a product page doesn't state its frequency range and whether it sweeps, that omission is itself information.
What none of these technologies do
Being clear about this is the point of the article.
- They don't kill. Every one of these is a deterrent. Nothing dies, and there's nothing to clear up.
- They don't pass through walls. Ultrasound doesn't. Electromagnetic pulsing travels in wiring, not through masonry as a field you can rely on. Vibration travels in soil, not into buildings.
- They don't remove the reason animals came. An open food source or a 6 mm gap under a door will outcompete any device. Deterrence raises the cost of staying; exclusion removes the invitation.
- They don't work instantly. Breaking an established territorial habit takes about four weeks, with a counter-intuitive increase in visible activity in the first fortnight as animals are disturbed.
How to read a spec sheet in ninety seconds
Frequency range and sweep behaviour. A range plus the word "sweep" or "variable". A single number is a red flag.
Coverage, stated per room. Any figure above roughly 200 m² for a single indoor plug-in unit is describing an unobstructed floor area, not a real home.
Output channels. More emitters means better angular coverage in an irregular room. One emitter covers one cone.
Build and rating. Outdoor units need a real IP rating and a stated operating temperature range. "Weather resistant" is not a specification.
Warranty length. In a category where the honest evidence is mixed, a manufacturer's willingness to carry risk over several years is the most reliable quality signal available to you — precisely because a low-quality product would make that guarantee expensive to honour.
What they refuse to claim. A manufacturer who tells you what their product can't do has told you something verifiable. One who claims 100% elimination has told you only that they're willing to say it.