How to Read a Timegrapher Result: Rate, Amplitude and Beat Error Explained

How to Read a Timegrapher Result: Rate, Amplitude and Beat Error Explained

The tick of a mechanical watch is not one sound. It is a burst of three tiny metallic impacts, spaced milliseconds apart, produced as the impulse jewel, the pallet fork and the escape wheel strike one another in a fixed sequence up to ten times every second. A timegrapher is a machine built to hear those impacts, time them against a precision quartz reference and translate them into a short row of numbers and a scrolling graph. Learn to read that screen and a movement will tell you more about its internal condition in a minute than its case and dial ever could.

At Crown Vintage, every watch that passes through our hands is measured on a Witschi WAIO, the all-in-one testing station made by Witschi Electronic of Büren an der Aare in Switzerland, whose measuring instruments are standard equipment in service centres across the watch industry. This guide explains what a timing machine is actually listening to, which components inside the movement produce each reading, and how to interpret the rate, amplitude, beat error and trace that appear on screen.

Reading Timegrapher Result: Rate, Amplitude and Beat Error Explained

What a Timing Machine Actually Hears

Witschi describes the measuring principle of the WAIO in three words: acoustic measurement of beat noises. The watch sits on a sensor stage that works like a highly sensitive contact microphone, picking up vibrations through the case itself. There is no need to open the watch. Everything the machine needs to know travels through the caseback as sound.

Each beat of a Swiss lever escapement produces three distinct noises in quick succession. First comes the unlocking, the sharp click of the impulse jewel on the balance striking the horns of the pallet fork and levering a pallet stone free of the escape wheel. Second is the impulse itself, a softer sliding sound as an escape wheel tooth drives across the angled face of the pallet stone and pushes energy back to the balance. Third comes the drop, the sound of a tooth landing on the locking face of the opposite pallet stone while the fork arrives against its banking pin. The entire sequence lasts only a few milliseconds.

The timegrapher separates these pulses and times them. The interval from one beat to the next gives the rate. The interval between the first and third noise within a single beat, combined with the lift angle of the calibre, gives the amplitude. The difference in timing between the tick and the tock gives the beat error. Three numbers, all extracted from sound alone.

The Parts Behind the Numbers

A timegrapher result is a health report on three interconnected systems inside the movement: the oscillator that keeps time, the escapement that counts it, and the power train that drives everything. Understanding what each part does makes the numbers on the screen far easier to interpret.

The Balance Wheel and Hairspring

The balance wheel is the timekeeper. It is a weighted ring, usually of a copper beryllium alloy such as Glucydur in quality Swiss calibres, mounted on a slender staff that pivots in jewelled bearings. Coiled at its centre sits the hairspring, a spiral of hardened alloy finer than a human hair, fixed to the balance staff by a collet at its inner end and to the balance cock by a stud at its outer end. Together they form a harmonic oscillator. The balance swings one way, the hairspring winds and pushes it back, and the cycle repeats at a fixed frequency, 2.5 times per second in a vintage 18,000 vibrations per hour calibre and 4 times per second in a modern 28,800 vph movement.

The rate figure on the timegrapher is a report card on this assembly. The effective length of the hairspring, set by the regulator pins on most calibres, determines whether the watch runs fast or slow. A magnetised hairspring whose coils momentarily stick together behaves like a shorter spring and sends the rate soaring, which is exactly why the Witschi WAIO pairs its timing measurement with a demagnetisation function that reports the rate difference before and after.

The Escapement

The escapement is the intermediary between raw power and measured time, and it is the part of the watch a timegrapher hears most directly. Its main components are the escape wheel, a light wheel with distinctively shaped teeth, and the pallet fork, an anchor-shaped lever carrying two synthetic ruby pallet stones, the entry stone and the exit stone. At the balance end of the fork sit the horns and the guard pin, and on the balance staff itself is the roller table carrying the impulse jewel. Two banking pins limit how far the fork can swing in either direction.

The escapement performs two jobs at once. It releases the gear train one tooth at a time, converting the continuous pull of the mainspring into discrete, countable steps. And with every release it delivers a minute push through the impulse jewel to keep the balance swinging. Every click, scrape and landing in that exchange is a data point for the timing machine. When an escapement is worn, dry or contaminated, the noise signature smears, and the trace on screen shows it immediately.

The Mainspring, Barrel and Gear Train

The mainspring is the power station, a long ribbon of spring alloy coiled inside the barrel. Its torque flows through the centre wheel, third wheel and fourth wheel to the escape wheel, with each step trading force for speed. None of these parts is measured directly by a timegrapher, yet their condition is written all over the result, because amplitude is a measure of how much of the mainspring's torque survives the journey to the balance. A tired mainspring, thickened oils in the barrel or worn pivots anywhere in the train all bleed away energy, and the balance swing shrinks accordingly. This is why watchmakers read amplitude first: it is the single best indicator of when a movement needs servicing.

The Four Readings on the Screen

Rate, in Seconds Per Day

Rate is the headline figure, expressed as seconds gained or lost per day. A reading of +4 s/d means the watch will run about four seconds fast over the next twenty four hours. The machine measures it by comparing thousands of beat intervals against its internal quartz timebase and extrapolating. For context, the COSC chronometer standard requires a mean daily rate between -4 and +6 seconds per day for a mechanical movement. A healthy vintage calibre regulated to within single digits per day is performing well, and period specifications were often considerably looser than modern ones.

Rate is also the least revealing number on its own. It can be adjusted in minutes with a regulator, so a watch showing a flat zero may simply be well regulated, not well serviced. Rate tells you how the watch is tuned. The other readings tell you how it is running.

Amplitude, in Degrees

Amplitude is the arc through which the balance wheel swings, measured in degrees from its resting position to the furthest point of rotation. Witschi's own documentation puts the typical range for modern wristwatches at about 260 to 310 degrees at full wind, and notes that the value falls gradually as lubricants age. A fully wound movement in good order swings freely, often approaching a full turn of rotation in each direction. As friction rises, whether from dried oils, contamination or pivot wear, the swing decays.

The working rules are simple. Amplitude below about 200 degrees at full wind means the movement needs attention, whatever the rate says. Amplitude that is unusually high, pushing beyond roughly 330 degrees, brings its own danger, called knocking or rebanking, where the balance swings so far that the impulse jewel strikes the outside of the pallet fork and the watch gallops. Between those extremes, the right benchmark depends on the calibre, and vintage movements are judged against period expectations rather than modern chronometer figures.

Beat Error, in Milliseconds

Beat error is the asymmetry between the tick and the tock. When a movement is perfectly in beat, the impulse jewel at rest sits exactly centred between the banking pins, and the balance takes precisely as long to swing clockwise as anticlockwise. Beat error measures the deviation from that ideal in milliseconds. A reading of 0.0 ms is perfect. Anything under about 0.5 ms is generally acceptable on a vintage watch, while readings much beyond 0.8 ms deserve investigation, since a badly out of beat movement can struggle to self-start and is harder to regulate consistently.

Correcting beat error means rotating the hairspring's mounting relative to the balance. On calibres with a movable stud carrier it is a fine adjustment on the balance cock. On many older movements it requires removing the balance and turning the collet on the staff, which is one reason a low beat error on a vintage watch is a quiet signal of careful past servicing.

Beat Rate and Lift Angle

The remaining figures are parameters rather than diagnoses, but both must be right for the rest to mean anything. Beat rate is the frequency of the movement in vibrations per hour. Vintage calibres commonly run at 18,000 vph, mid-century and later designs at 21,600 vph, most modern movements at 28,800 vph, and high-beat calibres at 36,000 vph. The WAIO detects beat rate automatically across a span of 3,600 to 72,000 beats per hour, and a machine that fails to lock onto a rate is itself a warning that the ticking is too irregular to classify.

Lift angle is the number of degrees the balance rotates while it is actually coupled to the pallet fork, and it is the one value the operator must set by hand, adjustable between 10 and 90 degrees on the WAIO. It matters because amplitude is calculated from it. Many ETA and Sellita calibres sit at 50 or 51 degrees, most machines default to 52, and many vintage Seiko movements need 54.5. Setting the wrong lift angle can distort the displayed amplitude by 10 to 20 percent, which is worth remembering whenever an amplitude figure looks surprisingly flattering.

Reading the Trace

Below the numbers scrolls the diagram, and it often says more than the figures above it. Each dot represents one beat, plotted by how early or late it arrived against the reference. A healthy movement in beat draws a single clean line. The slope of the line is the rate: perfectly horizontal means the watch is keeping reference time, a slope upward means it is gaining, a slope downward means it is losing.

Two parallel lines mean beat error, the tick and the tock arriving with a consistent offset, and the wider the gap between the lines, the larger the error. A regular wave rippling through the trace points to a fault that repeats with the rotation of a wheel in the train, such as a bent pivot or a damaged tooth, with the period of the wave hinting at which wheel is responsible. Scattered dots, the so-called snowstorm, mean the machine is hearing noise it cannot resolve into a pattern: debris in the escapement, a cracked jewel, or a movement so worn that the beat itself is unstable. A watch that produces a snowstorm at full wind is not a candidate for regulation. It is a candidate for the bench.

Why Watches Are Tested in Positions

A single reading dial up is only part of the story, because gravity changes how a movement runs. With the watch flat, the balance staff rests on its cap jewels and friction is at its lowest. Turn the watch vertical and the staff bears on the sides of its pivots, friction rises, and amplitude typically drops by 20 to 50 degrees, shifting the rate with it. Poise errors in the balance, meaning tiny imbalances in its mass distribution, also express themselves only in the vertical positions.

This is why the WAIO measures rate accuracy in four test positions automatically: dial up, dial down, and two vertical orientations with 6 o'clock and 12 o'clock uppermost. The spread between positions, called the delta, is a sensitive measure of the condition of the balance pivots and the quality of past servicing. A well serviced vintage movement might hold its positions within a handful of seconds per day. A large positional spread, even with a good flat rate, points to pivot wear or poise problems that no amount of regulating will cure.

The Witschi WAIO

The WAIO, short for Witschi All In One, condenses several bench instruments into a single automated station. Alongside its acoustic chronometry in four positions, it performs leak testing up to 5 bar using a built-in compressor, analysing the microscopic deformation of the case under vacuum and pressure, and it carries out demagnetisation with a measured before and after rate difference. Results appear on a 7 inch touchscreen in a format legible to owner and watchmaker alike. For a vintage dealer, that combination matters: it means every watch can be assessed for timekeeping, magnetism and case tightness in one automated pass, without opening the case, before it is ever offered for sale.

Final Thoughts

A timegrapher result rewards being read as a whole rather than as a row of isolated numbers. Amplitude describes the movement's health, because it measures how freely power flows from the mainspring to the balance. Rate describes its tuning, the easiest thing to change and the least meaningful on its own. Beat error describes its symmetry, and the trace ties everything together into a picture that flags faults the numbers can miss. Read together, they turn a few seconds of ticking into an honest account of what is happening beneath the dial, which is precisely why every watch we sell crosses the Witschi WAIO before it crosses the counter.

References

1. Witschi Electronic AG, WAIO product documentation and technical specifications, witschi.com/en/products/waio

2. Caliber Corner, Amplitude in Watches, calibercorner.com/amplitude

3. Caliber Corner, Lift Angle, calibercorner.com/lift-angle

4. Beyond the Dial, Collector Guide: Interpreting Timegrapher Results, beyondthedial.com

5. Firgelli, Lever Escapement Mechanism: How It Works, firgelliauto.com/blogs/mechanisms/lever-escapement

6. Contrôle Officiel Suisse des Chronomètres (COSC), chronometer certification criteria, cosc.swiss

Back to blog

Leave a comment

Please note, comments need to be approved before they are published.