On this page
Every seismometer, from Zhang Heng's bronze vessel to the sensor in your phone, works on one idea from Class 11 physics: a heavy object resists changes in its motion. When the ground shakes, the frame moves and the mass lags behind. Zhang Heng turned that lag into a signal nearly 1,900 years ago. This post covers what the historical record says, what we do not know, and a worked example of the physics in the language of pseudo-forces.
Who was Zhang Heng?
The MacTutor biography gives his dates as 78 to 139 CE, born in Nanyang. He served at the Han court and held the post of chief astrologer (court astronomer) more than once. He built an armillary sphere for mapping the stars, described about 2,500 visible stars, and proposed π = √10, about 3.162. He was also a celebrated poet.
What does the Book of Later Han say about the seismoscope?
The main source is his biography in the Book of Later Han (Hou Hanshu), the dynastic history of his era. You can read the original Chinese text of chapter 59 online. In summary, it says:
In the first year of the Yangjia era (132 CE) he built the houfeng didong yi, roughly "instrument for watching the winds and the movements of the Earth".
It was cast from refined bronze, eight chi across, with a domed lid, shaped like a wine vessel.
Inside was a central pillar (du zhu) with eight channels and a trigger mechanism. Outside were eight dragons, each holding a bronze ball, with eight toads below, mouths open.
When the earth moved, the vessel shook, a dragon released its ball and a toad caught it with a loud noise. Only one dragon fired; the other seven stayed still, so the direction showed where the quake was.
Once a dragon fired but no one in the capital felt anything. Scholars said the device had failed. A few days later a messenger arrived: there had been an earthquake in Longxi, to the west. After that, officials were ordered to record the direction each earthquake came from.
MacTutor dates this event to February 138.
How could a heavy pillar sense a distant earthquake?
Because the inside is lost, we can only say what physics allows. One possibility that fits the phrase "central pillar" is a heavy pendulum or upright column. When the ground and vessel jerk sideways, the heavy mass tends to stay where it was (Newton's first law). Relative to the vessel, it seems to swing the opposite way and can push a lever in one of eight channels. Engineering papers such as Yan and Hsiao's reconstruction design in Mechanism and Machine Theory (2007) explore how such linkages could work.
Worked example: how far does a hanging mass tilt when the ground accelerates?
Model the sensor as a 2 kg mass hanging on a string inside the vessel. Suppose the ground, and the vessel with it, accelerates east at a = 0.5 m/s². Take g = 9.8 m/s².
Pick the vessel's frame. It is accelerating, so it is non-inertial. Add a pseudo-force of magnitude ma, opposite to the acceleration: 2 × 0.5 = 1.0 N, pointing west.
List the forces on the mass. Weight mg = 2 × 9.8 = 19.6 N down, pseudo-force 1.0 N west, tension T along the string.
Find the tilt. In equilibrium in this frame, the string lines up with the combined "effective gravity". So tan φ = ma / mg = a / g = 0.5 / 9.8 = 0.051, giving φ ≈ 2.9° from the vertical, with the bottom of the string swinging west.
Notice what cancels. The mass drops out: tan φ = a/g. A heavier pendulum does not tilt more, but it pushes harder on a lever, which helps it trip a trigger.
The mass swings west when the ground accelerates east. A direction-finder has to account for this. The common mistakes are putting the pseudo-force in the same direction as the acceleration, or adding it in the ground frame, where it does not belong. Our post on pseudo-forces in non-inertial frames has more practice.
Was it a seismograph?
Not in the modern sense. MacTutor calls it a seismograph, but a seismograph records ground motion over time. Zhang Heng's device gave a one-off signal and a rough direction, so "seismoscope" is the more accurate word. The Longxi story also has a scientific lesson: a single confirmed detection is encouraging, but you would want many more trials before trusting an instrument.
What can a student learn from Zhang Heng?
Zhang Heng built a sensor for a problem his government cared about: knowing quickly where a disaster had struck. He turned an invisible event into a visible signal, a dropped ball. When you study a measuring instrument, whether a galvanometer, a spring balance or a pendulum, ask the same two questions: what physical effect does it use, and what turns that effect into a reading?
If seeing how a pendulum responds to a shaking support is easier than reading about it, the Fermi Simulation Lab has interactive simulations for ideas easier to see than read.
Try it now: redo the worked example for a = 1.5 m/s² and find the tilt angle.
