| Component | PGA | Velocity |
|---|
A teaching model of buildings swaying under a steady periodic push. It is not a design check and not a recorded earthquake.
Each floor is a 1000-tonne mass. Stories are 3.6 m, so height is 3.6 m times the number of floors. Stiffness is set so the fundamental period is T1 = 0.1 N seconds, and f1 = 10/N hertz. Modal damping is 2%, so resonance can magnify the motion by about 1/(2ζ) ≈ 25 compared with a static push.
The cyan, amber, and rose towers are fixed at 10, 20, and 40 stories. The violet block on the left is housing: the floor slider runs from 2 to 24 stories and the height follows.
The quake slider is peak ground acceleration. It starts at 0.045 g. Displacement, inter-story drift, and base shear all grow in proportion to that acceleration. The ground motion is a steady sine wave at the frequency on the bar, held at that peak — not an accelerogram.
The distance beside the acceleration is a rough magnitude reading: where a magnitude 6 quake would produce that peak, using the median Joyner and Boore (1981) relation. At 0.045 g that distance is about 49 km. At 0.20 g it is about 12 km. Above about 0.39 g the note says the shaking is stronger than magnitude 6 even at the fault.
Fixed base is the building bolted to the ground. Base isolators puts a soft bearing under every building, tuned to a 3-second period with 15% damping. The stories above stay at 2%. The bearing slides, so inter-story drift and base shear drop, unless the drive sits on that new long period. A 40-story tower already has a 4-second period, so the same bearing shifts it less than the shorter buildings. Wind still uses the fixed base.
Vortex shedding and turbulent shear use wind speed at 10 m. The slider starts at 25 m/s, which is 90 km/h. Floor force grows with the square of the speed and is stronger higher up the building. Doubling the speed from 25 m/s to 50 m/s multiplies the sway by four.
Simulate holds three steady-sine stand-ins for the Mw 7.7 Sagaing-fault earthquake of 28 March 2025 (about 12:50 local time, depth about 10 km). They are not the recorded accelerogram.
Mandalay was at the north end of the rupture. The epicenter was about 16 km north of Sagaing, and USGS intensity in the city was IX. No on-scale station recorded the city itself. The preset uses 0.75 g, the ShakeMap step for violent shaking, at 1.0 Hz so the 10-story fundamental sits in the mid-rise band. Housing is set to 8 stories. In the city, the 8-story Golden City Hotel and the 11-story Sky Villa collapsed.
Nay Pyi Taw (နေပြည်တော်) holds the only near-fault on-scale record, station NPW, 2.6 km west of the fault and 246 km from the epicenter. Fault-parallel acceleration peaked at 6.10 m/s² (0.622 g), the east–west component at 5.63 m/s², and the vertical at 10.53 m/s². The fault-normal velocity pulse was about 0.3 Hz (period 3.3 s), and the acceleration spectrum also peaked near 0.15 s (about 6.5 Hz). The preset uses 0.622 g at 0.30 Hz. That period sits between the 20-story building (2 s) and the 40-story building (4 s). The station also shifted 1.6 m along the fault; this model does not include that permanent offset.
The same rupture was felt in the Bangkok basin, about 1000 km south. Records there were about 0.02 g, but the soft soil amplified periods of roughly 1 to 6 seconds, and a 33-story building under construction collapsed. That preset uses 0.02 g at 0.25 Hz, a 4-second period on the 40-story fundamental. The 40-story tower stands in for a tall building in that long-period band.
Noto Peninsula, Japan, 1 January 2024, 16:10 local time, Mw 7.5. K-NET station Togi (ISK006) recorded a horizontal peak of 2.78 g, and that site amplifies motion near 0.2 seconds. The preset uses 2.78 g at 5 Hz and sets the housing to 2 stories, on that short-period peak. In Wajima and Anamizu the shaking that damaged houses was amplified at about 1 to 4 seconds; this sine is the Togi record, not those towns.
Southern Mindanao, Philippines, 8 June 2026, 07:37 local time, Mw 7.8, about 25 km southwest of Kablalan, 57 km deep. USGS ShakeMap puts the peak acceleration near 0.97 g, with the mapped spectrum strongest around 0.3 seconds. The preset uses 0.967 g at 3.33 Hz and sets the housing to 3 stories.
Flores, Indonesia, 15 August 2026, 05:58 local time, Mw 7.8, about 64 km north-northwest of Ende, 10 km deep. USGS ShakeMap puts the peak acceleration on land at 0.612 g, again with the mapped spectrum strongest near 0.3 seconds. The preset uses 0.612 g at 3.33 Hz and sets the housing to 3 stories.
Two shocks near San José del Palmar, Chocó, the same morning. The first, 07:34 COT (12:34 UTC), was Mw 7.4 at intermediate depth (USGS 108 km, Colombian Geological Survey 103 km). The second, 08:18 COT, is the M 4.8 the Survey lists at the same town; USGS calls that shock mb 5.0. Both presets are steady sines, not the records.
Filandia station FLND recorded the largest horizontal peak in the Survey’s station table for the mainshock: 422.7 cm/s² east–west (0.431 g), 268.3 cm/s² north–south, and 105.5 cm/s² vertical. Manizales station MAN1C recorded 0.289 g east–west. USGS ShakeMap maxima are 0.448 g and 54 cm/s, intensity VIII. The mapped spectrum is strongest near 1 second, so the Filandia preset uses 0.431 g at 1.0 Hz, on the 10-story fundamental. A preliminary report also noted Jamundí at 0.427 g east–west and asked for more review of that record, so the sine does not use it. Station velocities were not in the acceleration table; the 54 cm/s figure is the ShakeMap maximum.
San José del Palmar station PAL for the 08:18 shock: 23.6 cm/s² east–west (0.024 g), 21.3 cm/s² north–south, and 13.9 cm/s² vertical. No velocity and no dominant period were published for that shock. The preset uses 0.024 g at 1.0 Hz so the two shocks can be compared on the same building period.
Kathmandu, 25 April 2015, 11:56 local time, Gorkha Mw 7.8. At Kantipath the horizontal peak was 0.23 g and the vertical peak 0.19 g. The valley shook hardest at periods of about 3 to 5 seconds, so the preset uses 0.23 g at 0.25 Hz, the 40-story fundamental. The largest velocity in the valley was 99 cm/s north–south at TVU in Kirtipur, a different station from Kantipath.
Türkiye, 6 February 2023, Pazarcık Mw 7.8. Two stations from the processed AFAD records are listed with north–south, east–west, and vertical peaks. Pazarcık station 4615 is about 10 km from the rupture: horizontal peaks near 0.60 g, vertical 0.68 g, and the north–south velocity 152 cm/s. Hatay station 3129 in Defne: north–south 1.39 g and 171 cm/s, east–west 1.22 g and 76 cm/s, vertical 0.84 g and 44 cm/s. The building sway uses the larger horizontal acceleration. Vertical shaking is shown in the table and is not applied to the frame.
The frequency bar is logarithmic and is shared by every building. Keys 1–3 tune the three towers’ fundamental frequencies, 4–6 their higher modes, and 7 the housing fundamental. When the drive matches f1, f2, or f3, that building resonates and the roof displacement, drift, and base shear grow together. Hold a building on resonance and move only the shaking slider to see strength change without leaving the peak.
Skyline sway is magnified so the lean is visible. The charts stay at true scale. Column color is the inter-story drift ratio, from 0 to 4%.