What Is a Timegrapher and What Do Its Readings Mean?

A timegrapher is an instrument that listens to a mechanical watch's ticking and turns it into three numbers: rate (how many seconds per day the watch gains or loses), amplitude (how energetically the balance wheel swings), and beat error (how evenly the watch ticks). Watchmakers use one to regulate and diagnose movements; collectors use one to check whether a watch is running within spec. A reading takes seconds and reveals more about a movement's health than days of wearing the watch.

What a timegrapher does

Every mechanical watch has an escapement — the assembly that releases the mainspring's energy in tiny, regular increments. Each release makes a sound: the familiar tick. How often it ticks is set by the movement's beat rate, quoted in beats per hour (BPH). The most common modern rate is 28,800 BPH, which works out to 8 ticks per second, one every 125 milliseconds. Other standard rates range from 18,000 BPH in vintage movements to 36,000 BPH in high-beat calibers.

A timegrapher picks up those ticks — traditionally with a piezoelectric contact sensor clamped against the watch case — and times each one against a precise quartz reference. From the spacing and internal structure of the ticks it computes the three readings below, and plots each tick as a dot on a scrolling trace. A clean movement draws two tight parallel lines; problems show up as slope, scatter, or splayed lines.

The key thing to understand is that a timegrapher gives you an instantaneous diagnosis, not a verdict on real-world timekeeping. It tells you how the movement is running right now, in one position, at one state of wind. How the watch actually performs on your wrist over a week is a related but different question — see how accurate a mechanical watch should be for what counts as normal there.

The three readings: rate, amplitude, and beat error

Rate

Rate is the deviation from perfect timekeeping, expressed in seconds per day. A reading of +8 means that if the watch kept behaving exactly as it is behaving this instant, it would gain 8 seconds over 24 hours. Rate changes with position — dial-up and crown-down readings often differ by several seconds — which is why watchmakers measure in multiple positions and why chronometer standards specify positional testing. COSC, the Swiss chronometer certification body, requires an average daily rate between −4 and +6 seconds; an ordinary modern automatic is typically regulated somewhere within about ±15 seconds per day, with specs varying widely by brand and caliber.

Amplitude

Amplitude is how far the balance wheel rotates on each swing, measured in degrees. It is effectively a fuel gauge for the movement's energy: a healthy, fully wound movement typically swings somewhere around 250–315° measured dial-up (published guidance varies by source and movement), and amplitude naturally drops in vertical positions and as the mainspring winds down. Low amplitude — consistently under about 200° on a full wind — usually means degraded lubricants or worn parts, and is the classic sign a service is due. Very high amplitude (roughly 330° and up) risks "knocking," where the balance swings so far it strikes the impulse pin from the wrong side.

One caveat: a timegrapher does not see the balance wheel. It infers amplitude from the timing of sounds within each tick, using the movement's lift angle — a per-caliber constant, commonly around 52° but ranging from roughly 44° to 58°. If the lift angle setting is wrong, the amplitude reading is wrong, even though rate and beat error remain valid.

Beat error

Beat error is the asymmetry between the tick and the tock, in milliseconds. The balance should swing equally far each side of its resting point, so the two halves of each oscillation take equal time. When the impulse jewel is off-center, one interval runs long and the other short — at 28,800 BPH, perhaps 126 ms then 124 ms instead of 125 and 125. Under 0.5 ms is good; under about 1 ms is acceptable on most modern movements. For what causes it and when it matters, see the full guide to watch beat error.

Hardware timegraphers vs phone apps

Dedicated machines span a wide price range. Entry-level units — the ubiquitous Weishi machines and similar — typically cost around $100–250 and are perfectly serviceable for hobbyists. Professional bench instruments from makers like Witschi run $500 to $3,000 and up, with faster convergence, positional test automation, and certified precision. All of them share the same advantage: a contact sensor pressed against the case, which delivers a strong, clean signal that is essentially immune to room noise.

Phone apps take the harder path: the built-in microphone, listening to the tick through open air. The signal is dramatically weaker and buried in ambient noise, which is a genuine signal-processing problem rather than a packaging difference. ChronoLog's audio timegrapher solves it with epoch folding, a technique borrowed from radio astronomy, and measures rate and beat error through the phone's microphone with typical accuracy of ±2–5 seconds per day, no extra hardware — there is a full engineering write-up on how we built a timegrapher from a phone microphone.

The practical takeaway: for regulating a movement on the bench, hardware wins. For answering "is my watch healthy, and roughly how fast or slow is it running?", an app on the phone already in your pocket is usually enough.

How to take a reading you can trust

Timegrapher numbers are only as good as the conditions you take them in. A few habits make the difference between a meaningful reading and a misleading one:

Wind the watch fully and let it run for a few minutes before measuring. Amplitude and rate both sag as the mainspring unwinds, so a half-wound reading will look worse than the movement deserves.

Keep the environment quiet and the watch still. This matters most for microphone-based apps, but even contact sensors dislike vibration from a wobbly desk or a running appliance.

Set the correct lift angle for your movement if you care about the amplitude number. Most machines default to 52°, which is right for many modern calibers and wrong for plenty of others — the figure is usually published in the movement's technical sheet.

Measure in more than one position. Dial-up, dial-down, and crown-down is a reasonable minimum. Give the trace 20–30 seconds to stabilize after each position change, and expect some spread between positions — a few seconds per day of positional variation is normal.

Don't stop at one snapshot. A single reading tells you how the movement is running at that instant; your watch's true daily rate is the average over how you actually wear and rest it. Tracking measured accuracy over several days complements the timegrapher's instant diagnosis.

Limits: what a timegrapher can't tell you

A timegrapher is a diagnostic tool, not an oracle. Its reading is a snapshot in a handful of static positions at one temperature and state of wind — it will not match the rate you observe on the wrist, where the watch moves through every position and your arm keeps the mainspring topped up. A watch can look mediocre on the machine and keep excellent time in real life, and occasionally the reverse.

It also identifies symptoms rather than causes. A suddenly fast rate with a clean trace suggests magnetization; scattered dots suggest dirt or damage in the escapement; low amplitude suggests old oil — but these are inferences, and confirming the cause is a job for a watchmaker with the movement open on the bench. The machine cannot measure power reserve directly, cannot check water resistance, and a conventional acoustic timegrapher cannot read a quartz watch at all, since quartz movements do not have a ticking escapement of the same kind.

Finally, remember the lift-angle caveat: of the three readings, amplitude is the only computed one, and it is only as trustworthy as the lift angle you entered. Rate and beat error come straight from tick timing and are robust even when the amplitude figure is garbage.