At low frequencies, the transmission loss depends on the modal behavior of the partition wall (here a 18-mm plaster board partition). The elastic parameters of the partition wall (including the mass density) and the boundary conditions at the junctions with the floor and roof, are needed to describe the real behavior of the partition (which appears in grey on the graph between 63 and 1k Hz). The higher the amplitude of the partition vibration, the lower the transmission loss.

A rough estimation of the transmission loss at low frequencies can however be obtained by only considering the mass of the partition. This mass law modeling, only valid at low frequencies, below the critical frequency, is depicted in black on the graph.

The dip at 2k Hz corresponds to the critical frequency. At this frequency (and higher ones), the wavelengths of the waves propagating inside the partition are small compared to the dimensions of this partition so that the boundary conditions have a negligible impact. The elastic parameters (particularly the loss factor) control the vibrating behavior of the partition.
The critical frequency results from the coincidence frequencies of multiple angles of sound incidence.

Above the critical frequency, the transmission loss increases, as the vibrations of the partition have smaller amplitudes.

About the plot :
The cartoon style is a tribute to the work by late Randy Glasbergen. The original drawing has been done from scratch, inspired by multiple drawings by R. Glasbergen (so I assume this is not a copyright infringement). The final result has been vectorized for the purpose of this page.

The real modal behavior curve is a FEM simulation (Finite Element Modeling). The black curve (from the mass law modeling at low frequencies up to 20k Hz) is a FTMM simulation (Finite Transfer Matrix Method, also known as Windowed Transfer Matrix Method).

Unless otherwise indicated, all the content of this page is copyleft under
the creative commons license Attribution 3.0 Unported (CC BY 3.0).
Luc Jaouen (@ljaouen), ISSN 2606-4138.
Site hosted by Matelys.