How this instrument works
One crest advances exactly one wavelength during one period, and a period lasts 1/f seconds. Divide distance by time and v = fλ falls out. That makes this an identity rather than physical law: no experiment could contradict it, because it merely restates what wavelength and frequency mean. Physics enters through a division of labour instead. A source picks f — your wrist, a loudspeaker coil, an oscillating charge — while a medium fixes v through its own stiffness and inertia. Wavelength is whatever remains once those two have spoken.
Newton attempted a wave speed from first principles in Book II of Principia in 1687, treating sound as a pressure disturbance passing between air particles, and reported roughly 979 feet per second. Measurement stubbornly returned about 15 percent more. His error survived 129 years until Laplace saw in 1816 that compressions happen too fast for heat to escape, multiplying by √γ and closing almost all of that gap. Typical figures worth carrying: 343 m/s in air at 20 °C, near 1,480 m/s in fresh water, around 5,900 m/s along steel rails, and 299,792,458 m/s for light in vacuum — now exact by definition, since 1983, because metres are defined from it.
Trouble begins where speed itself depends on frequency. Such media are called dispersive, and there v = fλ still holds for each sinusoid separately while pulses built from many of them spread and travel at some different rate. Deep-water swell obeys v = √(gλ/2π), so long waves outrun short ones and distant storms arrive sorted by period over several days. Optical fibre disperses enough to smear pulses across long spans, turning repeater spacing into an engineering budget. Two further caveats: standing waves propagate nowhere and have no speed to report, and at large amplitude — shock fronts, breaking surf — speed starts depending on amplitude too, at which point linear wave theory quietly retires.
- Enter your source rate in Frequency. Its menu spans Hz, kHz, MHz and GHz, so an FM carrier goes straight in as 100 MHz without counting zeros.
- Put crest-to-crest distance into Wavelength, choosing mm, cm, m or km. Measure across ten crests and divide by ten; a single gap is noisy.
- Read Wave speed underneath. It defaults to m/s, with km/h, mph and ft/s available on its unit menu.
- Audit against a medium rather than a source: roughly 343 m/s in room air, 1,480 in fresh water, 5,900 in steel. Answers far from those usually mean mismatched prefixes.
- Where speed varies with frequency — fibre, deep water, hollow waveguide — read this figure as phase velocity. Energy and signals travel at group velocity instead.
Worked example — a rope shaken twice a second
Tie a heavy rope to a wall, pull it taut, and flick your wrist twice each second. Crests stream away down its length; lay a tape measure alongside and they sit 3 metres apart. Enter 2 into Frequency and 3 into Wavelength, and Wave speed returns 2 × 3 = 6 m/s exactly. Nothing rounds, because neither input hides an irrational number.
Follow one crest and that arithmetic explains itself. Half a second completes its cycle, and during that half second it covers precisely one wavelength; two such intervals fill a second, so six metres go by. Your wrist had no say in 6 m/s — tension and mass per unit length did, through v = √(T/μ). Shake faster and crests merely bunch closer together while speed holds firm. Work backwards for a rope carrying 0.2 kg per metre: 6 m/s calls for 7.2 newtons of pull, near enough a 730-gram mass hung over a pulley.
Questions
Does raising the frequency make a wave travel faster?
No, and this is the mistake worth guarding against. Speed belongs to a medium, not to whoever is shaking it. Turn a signal generator up and wavelength shrinks in exact proportion, leaving v untouched. Evidence sits in any concert hall: piccolo and double bass reach row Z together, so air must carry every audible pitch at one rate. If it did not, chords would arrive smeared and orchestras would be impossible. Only dispersive media — glass, optical fibre, deep water, waveguides — break that rule, and they announce themselves by splitting white light or sorting swell by period.
Which units does this formula actually need?
Hertz and metres, giving metres per second. Most wrong answers trace to prefix mismatches rather than bad physics: nanometres paired with terahertz, gigahertz paired with millimetres, centimetres left unconverted. Both unit menus here handle that conversion, so pick honest units and let the sheet do the scaling. One trap the menus cannot catch is angular frequency ω, quoted in radians per second and larger than f by 2π. Divide it by 2π before entering, or work with v = ω/k instead.
Why doesn't light change colour when it enters water?
Because frequency is conserved across a boundary while wavelength is not. Fields on either side of an interface must stay in step cycle for cycle, so f survives the crossing intact; speed drops to c/n and wavelength shrinks by that same refractive index. Water sits near n = 1.33, so 500 nm green light runs at about 225,000 km/s inside and spans roughly 376 nm — still green, since retinal pigments respond to frequency. Quoted wavelengths for light almost always mean vacuum values, which is worth remembering before feeding one into any medium other than vacuum.
What is the difference between phase velocity and group velocity?
Phase velocity is what fλ returns: the speed of one individual crest in a single endless sinusoid. Group velocity, dω/dk, is the speed of an envelope built from many frequencies, and it carries energy and information. Where speed is frequency-independent both coincide and nobody needs the distinction. In deep water group velocity is exactly half phase velocity, so crests visibly run forward through a wave group and dissolve at its leading edge. Phase velocity may also exceed c — X-rays in glass, or a hollow waveguide — with no relativistic scandal, since an infinite sinusoid never delivers a message.
How is wave speed measured in practice?
Usually as a timing problem over a known distance, or its inverse. Seismologists run it backwards: P waves travel near 6 km/s through crust and S waves near 3.5 km/s, so a gap between their arrivals converts directly into distance from an epicentre. Medical ultrasound scanners assume 1,540 m/s for soft tissue and turn echo delay into depth, which is why fat at roughly 1,450 m/s produces small registration errors. Cable technicians do the same with a velocity factor near 0.66 of c for RG-58 coax, locating a break from a reflection delay of microseconds.
Does any of this apply to a standing wave?
A standing wave transports nothing along its length, so strictly it has no speed. Feed it in anyway and fλ returns something meaningful: the speed of two counter-propagating travelling waves whose superposition builds that pattern. Fundamental wavelength on a fixed string equals twice its length, so a 0.65 m guitar string sounding 440 Hz has λ = 1.3 m and components running at 572 m/s. Tighten a tuning peg and you raise that speed; length pins λ, so pitch follows.