What Frequency Cleans Speakers? (165Hz Explained)

Every phone speaker cleaning tutorial mentions 165Hz. Some say 160Hz. A few say 200Hz. Apple Watch plays around 165Hz for Water Lock. What's actually happening at that frequency, and why not higher or lower?

This is the physics of cleaning tones, written for the reader who wants to understand rather than just follow instructions.

Speakers are pumps

A phone speaker is a small voice-coil transducer. Voltage in, motion out. The voice coil sits in a magnetic gap; when current flows through it, the coil is pushed or pulled by the magnetic field, which moves the attached diaphragm. The diaphragm pushes air, and air pressure waves hit your ear as sound.

The diaphragm moves back and forth at whatever frequency you drive the coil. 440Hz in = diaphragm oscillates 440 times per second. 60Hz in = diaphragm oscillates 60 times per second. The maximum distance the diaphragm travels on each cycle is called excursion.

Here's the key fact: excursion is much larger at low frequencies than at high frequencies for the same input power.

A speaker driven at 60Hz moves far more physical distance than the same speaker at 1000Hz. That's because low-frequency sound requires pushing more air per cycle, which requires longer diaphragm travel.

When you're trying to clean a speaker, large excursion is what you want. The diaphragm moving through a large distance creates enough air pressure to push debris out of the grille and shake particulate off the cone itself.

Why not just go as low as possible?

If more excursion means more cleaning, why not just play 20Hz and maximize it?

Two reasons:

1. Resonant frequency matters. Every speaker has a natural resonance — the frequency at which its physical construction produces maximum efficient excursion. For small phone speakers, that resonance is typically between 300Hz and 400Hz. Below the resonance, excursion gets limited by the suspension stiffness. The speaker wants to move but the suspension fights it, producing heat in the voice coil rather than displacement.
2. Voice coil heating. At very low frequencies, the coil is pushed to its mechanical limits for extended periods. The electrical energy that doesn't become motion becomes heat. A 30-second run at 20Hz generates far more voice coil heat than the same 30 seconds at 165Hz.

So the cleaning frequency target isn't "as low as possible." It's "as low as the speaker can comfortably handle without excessive heating."

Why 165Hz specifically

165Hz sits in the range where small phone speakers can produce near-maximum sustainable excursion without overheating. Here's the breakdown by band:

• Below 100Hz: too much suspension resistance, too much heat, not enough excursion benefit.
• 100-150Hz: usable, but less excursion than ideal for the heating cost.
• 150-175Hz: sweet spot. Large excursion, manageable heat, efficient coupling to the speaker cavity air volume.
• 175-250Hz: still effective but excursion drops noticeably as you approach the driver's resonance.
• 250-400Hz: driver is at or near resonance. Maximum efficiency for sound production, but smaller excursion distance than 165Hz. Works for mild dust but weaker for water ejection.
• Above 500Hz: excursion too small for meaningful debris ejection.

165Hz specifically also happens to be outside the typical speech frequency range (most voice energy sits between 85Hz and 255Hz, but peak voiced energy is usually around 100-130Hz), which means a 165Hz tone feels distinctly like "cleaning" rather than a human voice. That's useful for users who want to know the tone is doing something.

Water ejection is different from dust cleaning

The two problems respond to slightly different frequencies.

Water ejection wants maximum excursion at low frequency because it's literally pumping air and water out of a cavity. 150Hz-165Hz is ideal. Apple Watch's Water Lock tone is in this range.

Dust cleaning benefits from vibration across a wider range. The coupling between diaphragm motion and dust dislodgement works at frequencies from 150Hz up to about 1000Hz, with effectiveness decreasing as frequency rises. 165Hz is a compromise that handles both dust and moisture.

Some cleaning apps play a swept tone — a frequency range rather than a single frequency — to catch debris that might be easier to dislodge at specific spots within the range. This is marginally more effective than a pure tone for dust, no more effective for water.

Why your speaker vibrates in your hand during the tone

The 165Hz tone at maximum volume is loud enough (and at a low enough frequency) that the mechanical energy transfers noticeably to the phone chassis. Your hand feels it because the phone body itself is oscillating in sympathy with the speaker.

This is not the speaker breaking. It's the expected physical coupling between a small speaker and a small chassis at a frequency near the chassis resonance. The vibration decreases at higher frequencies because less mechanical energy is in the tone.

What about ultrasonic cleaning?

Some advertisements claim "ultrasonic speaker cleaning," implying frequencies above 20kHz. Phone speakers physically cannot play ultrasonic frequencies at meaningful volume — the driver has a high-frequency rolloff that attenuates anything above about 15kHz, and most phone speakers fall off by 8kHz.

Dedicated ultrasonic cleaners (like the ones jewelry stores use) work on a completely different principle — they use a liquid bath and cavitation bubbles. You can't do that to a phone. Any app claiming ultrasonic cleaning through the phone's own speaker is misrepresenting what's happening.

Why apps can play slightly different frequencies

You'll see cleaning apps that use 160Hz, 165Hz, 170Hz, 200Hz, or a sweep. Small differences in the chosen frequency reflect:

• The specific phone models the app was tuned for (slightly different speaker resonances)
• Whether the app prioritizes water ejection (lower) or dust cleaning (higher)
• Whether the app sweeps a range or sticks to a single tone
• Proprietary choices by the app developer

All of these are within the effective range. The differences are small enough that the effectiveness is essentially the same across any reputable cleaning app.

What definitely doesn't work

A few frequency-related claims you'll see that are wrong:

• "432Hz healing frequency cleans speakers." 432Hz is in a healthy audible range but produces small diaphragm excursion — no better than a random music track for actual cleaning.
• "Binaural beats clean speakers." Binaural beats are interference patterns between two slightly-different tones played in each ear. Irrelevant to mechanical speaker cleaning.
• "Brown noise cleans speakers." Brown noise has energy mostly in low frequencies. It can work somewhat, but a pure 165Hz tone at max volume is far more effective than broadband noise at the same RMS level.

Pseudoscience catches users who aren't sure what they're supposed to be doing. A calibrated single tone between 150 and 175Hz is the correct answer.

The duration question

How long should you run a cleaning tone? The general answer: 15-30 seconds per pulse, repeated up to three times with 30-second gaps.

Why the gap between pulses: voice coil heat dissipation. During a pulse, the coil absorbs electrical energy. The 30-second gap lets it cool back to ambient before the next pulse. Continuous runs of several minutes build up real heat.

Why not longer single pulses: diminishing returns. By 30 seconds of a 165Hz pulse, free debris has already been ejected. Additional time mostly generates heat without removing more debris.

The short version

165Hz is the standard cleaning frequency because it's the sweet spot for small phone speakers: large excursion, manageable heat, effective for both dust and water ejection. Frequencies within about ±15Hz of that work similarly well. Frequencies far from that range either under-excite the diaphragm (too high) or overheat the voice coil (too low). Single-tone pulses for 15-30 seconds are more effective than continuous multi-minute runs.

Understanding why 165Hz works makes the cleaning routine feel less like magic and more like mechanical engineering — which is what it is.