Silo quaking is a very complex industrial phenomenon and is often occurred in silo structures during discharge. There are three variants of the silo quaking phenomenon such as silo honking, silo shock and silo pulsation. Accelerometers are often used in practice to measure the responses of the structure. The accelerations generated by the silo structure during discharge are of non-linear and non-stationary nature. Therefore traditional algorithm, such as Fast Fourier Transform, which was designed to analyse linear and stationary signals and often used to analyse the collected acceleration data, is not suitable for signals generated by the silo structure during discharge. In the present research, Hilbert-Huang Transform and Hilbert Marginal Spectrum created by Norden Huang, a mathematician at NASA to analyse non-linear and non-stationary signals, were used to analyse the accelerograms collected from the experimental silo. The results revealed the existence of a silo quake spectrum which lays a solid foundation for further research into the silo quaking phenomenon and especially structural design methods to mitigate the effects of silo quaking on the silo structure. Crown Copyright (C) 2018 The Society of Powder Technology Japan. Published by Elsevier B.V. All rights reserved.