Piano Design
Stephen Birkett

Piano Design
Piano Action Dynamics
High Speed Imaging
Physical Systems
Animal Growth Modelling
Environmental Systems

Linear Algebra
Numerical Methods
Computational Mathematics
Musical Instruments

Marcia Hadjimarkos


The overall objective of this research program is to investigate the acoustic and mechanical behaviour of the piano and relate it to design principles. Novel experimental investigations will be used to support the development of predictive simulation models.

Individual models will be constructed for the main functional subsystems of the piano:

  • action mechanism
  • string
  • soundboard & bridge
  • structural

The important interactions between these have been recognized intuitively by piano designers for centuries. Subsystem couplings will be explictly represented in the model: hammer-string interaction, string-bridge-string, string-bridge-soundboard, and soundboard-structure.

The computational model will be generic and based on principles which are applicable to ALL pianos, whether historical, modern, or unconventional (post-modern). This will provide a useful practical design and analysis tool for piano builders, re-builders, technicians, restorers, and organologists.

The project is aiming to develop a model with a realistic predictive capability for a particular set of detailed design specifications (structure and parameters). Consequently, calibration and model validation are an integral and important aspect. Experimental investigations will utilize a variety of technical approaches, including high speed imaging, controlled mechanical actuation and position measurement (linear encoders), accelerometers, load cells, and modal analysis based on data collected via laser doppler vibrometer. Particular emphasis will be placed on material properties. Interference between data collection and the system being investigated will be minimized.

The following aspects of this research program are currently underway:

  1. Dynamics of piano action mechanisms. Development of a multibody dynamics model for the operation of a piano action mechanism. The model will be validated using the results of experimental investigation of test action mechanisms operated by accurately repeatable force drivers. Experimental techniques include qualitative and quantitative analysis of data from high speed digital video imaging. Associated research will also investigate the dynamic control capabilities in the pianist/piano interface, i.e. is there a physical basis for the perceived subtleties of tone control used by trained pianists?
  2. Felt compression. Initially we have investigated all available models of force-compression characteristics of simple felt pads. This is relevant to the modelling of contact dynamics between components in the dynamic action mechanism model. The conclusion that no existing model can satisfactorily represent simple felt compression has been supported by comparing the results of simulations with published models to those obtained from experimental investigation of felt pads under load. A new model of felt compression is being developed. This will form the basis for further studies of the mechanical behaviour of felt in manufactured contexts, e.g. piano hammers.

|top of page|

©2004 Stephen Birkett