Skip to content
Murat Güven

Phone vibrometer

Lay your phone on a desk, a floor or a footbridge and read its natural frequency.

  1. Lay the phone flat on the structure, screen up, where it can’t fall.
  2. Tap Start and take your hands off it.
  3. Let the structure move: walk or jump once on a footbridge, or measure while people cross.

No measurement yet. Start measuring, or upload a CSV.

– Hz

Spectrum, amplitude against frequency

The spectrum of your measurement appears here.

Acceleration, mean and drift removed

The recorded acceleration appears here.

Method

The browser reads the phone’s accelerometer through motion events. The record is resampled to a uniform rate, the mean and drift are removed, and a Hann-windowed fast Fourier transform gives the spectrum of the axis that moved most.

The highest peak above 0.5 Hz is the main frequency, refined between frequency bins. Smartphone accelerometers have been checked against reference sensors on shaking tables and a bridge (Feng et al. 2015).

ReferencesCode

Assumptions and limits

Browsers deliver about 60 to 100 samples a second, so only frequencies below half that rate show, and they round the readings (Chrome to 0.1 m/s²), so only clear vibrations register. Phone sensors are uncalibrated and noisy: frequencies are more reliable than amplitudes. A reading is indicative, not a structural assessment.

Nothing leaves your device: the recording stays in this page until you export it.

Cite

@software{guven_vibrometer,
  author = {Güven, Murat},
  title  = {Phone vibrometer},
  year   = {2026},
  url    = {https://likemaestro.github.io/tools/vibrometer}
}

References

  1. Cooley, J. W., & Tukey, J. W. (1965). An algorithm for the machine calculation of complex Fourier series. Mathematics of Computation, 19(90), 297–301. doi:10.1090/S0025-5718-1965-0178586-1
  2. Feng, M., Fukuda, Y., Mizuta, M., & Ozer, E. (2015). Citizen sensors for SHM: Use of accelerometer data from smartphones. Sensors, 15(2), 2980–2998. doi:10.3390/s150202980