Kilohertz to Microhertz

The Nyquist theorem requires a sample rate at least twice the highest frequency you want to capture. Human hearing tops out near 20 kHz, so you need at least 40 kHz. The extra 4.1 kHz provides headroom for the anti-aliasing filter to roll off. The specific number 44,100 was chosen because it factored neatly into the video frame rates of the PAL and NTSC systems used to store digital audio on videotape during early CD mastering.

Kilohertz (kHz) measures oscillation frequency — cycles per second. Kilobits per second (kbps) measures data throughput — bits transferred per second. A 44.1 kHz audio sample rate means 44,100 snapshots per second, but each snapshot may be 16 bits, yielding 705.6 kbps for one channel. The two units describe fundamentally different things: how fast something vibrates vs. how fast data flows.

AM radio was developed first and was allocated the medium-frequency band (535–1,705 kHz) because those wavelengths travel long distances by bouncing off the ionosphere at night. FM came later and was assigned the VHF band (87.5–108 MHz) — higher frequency means shorter range but much better audio fidelity and resistance to static. The allocation reflects both physics and regulatory history.

Yes. A typical dog whistle emits ultrasound between about 23 and 54 kHz — well above the human ceiling of ~20 kHz but within a dog's hearing range, which extends to roughly 65 kHz. Some "silent" whistles do leak a faint hiss that keen human ears pick up, but the dominant output is ultrasonic. Cats hear even higher, up to about 85 kHz.

Traditional landline phone calls sample voice at 8 kHz, which by Nyquist captures frequencies up to 4 kHz. Human speech intelligibility lives mostly between 300 Hz and 3,400 Hz, so 8 kHz is just enough. It is why phone calls sound muffled compared to in-person conversation — you lose all the higher harmonics that make a voice sound natural. HD Voice (VoLTE) bumps the rate to 16 kHz, doubling the bandwidth and noticeably improving clarity.

Solar oscillation modes with periods of hours to days, slow tidal harmonics, and long-period stellar variability all live in the microhertz band. Earth's free-core nutation — a wobble of the liquid outer core relative to the mantle — oscillates near 1 μHz. These are real physical processes, just far too slow for any wristwatch to track.

Ground-based detectors like LIGO are deafened below about 10 Hz by seismic noise. LISA will float three spacecraft in a triangle 2.5 million kilometers across, far from terrestrial vibrations, making it sensitive from ~0.1 mHz down into the microhertz regime. That band contains signals from massive black-hole mergers and thousands of compact binary stars in our own galaxy.

You need at least one full cycle to confirm a periodic signal, and preferably several. At 1 μHz (period ~11.6 days), a few months of data suffices. At 0.01 μHz (period ~3.2 years), you need a decade or more. This is why long-baseline observational campaigns — decades of pulsar timing or stellar photometry — are essential for low-frequency science.

Helioseismology studies sound waves trapped inside the Sun. The Sun rings like a bell with millions of overlapping oscillation modes. Most solar p-modes peak around 3 mHz (5-minute period), but gravity modes (g-modes) deep in the solar core are predicted at microhertz frequencies. Detecting those elusive g-modes would let scientists probe conditions at the Sun's very center.

A microhertz is a million times slower than one hertz. If middle C on a piano (262 Hz) were slowed to 1 μHz, a single wave cycle would take about 30 years. You would hear the first peak of the note in your twenties and the first trough around your fiftieth birthday. It puts cosmic patience into perspective.