Constarium
← Search

Data · dataset · 2026

Physics-based emulation of spring reverberation

Listed in ZivaHub and Deakin Research Online and DMU Figshare — shown once because both records carry DOI 10.17034/32641557.v1

Physical modelling for sound synthesis aims to capture the underlying physics of a system via a mathematical description.

Description

The main draw of this approach is getting authentic digital reproductions of the vibrational behaviour, as is encoded in the underlying physics. A typical focus beyond accurate modelling is arriving at a light-weight computational algorithm including real-time control over physical parameters.<br><br>This thesis focuses on physical modelling of spring reverberation.

In past physics-based research, a good reproduction is achieved with a simplified helical spring model, but this does not fully capture the spring geometry. Attempts with a more complex model, typically employing a `thin spring' formulation in which Timoshenko effects are ignored, have not yet seen a close match with measurements.<br><br>The lack of success with such models so far is largely down to simplifications at the input and output.

Read the rest (3 more)

On this basis, the work here sets out to improve on past thin spring attempts by now incorporating the magnetic beads' dynamics at the input and output. A more complex damping model is also considered.<br><br>The magnetic beads and their supporting wires are modelled as stepped beams. Experimental work using a high-speed camera informs how input and output are defined in a model with beads incorporated.

Developing an alternative thin spring formulation allows modelling the full system with the same set of equations. The methodology to arrive at a discrete-time modal algorithm starting from the reverb tank formulation uses the Chebyshev pseudospectral method and impulse invariant method to discretise space and time, respectively.<br><br>The main outcome of the improved physical model is achieving, for the first time, a close match with a measured impulse response starting from a helical spring model that fully captures the helical geometry.

The reproduction of a measurement's main features is verified visually via both spectrogram and time-domain plots; aural comparisons concur. The final algorithm is real-time capable, and a proof-of-concept audio plugin demonstrates scope for online control of physical parameters.

Links

Where it is published

Catalogue records · 1

Topics

Inferred from text
Audio 75%
Provenance · 3 source records, 8 field assertions
SourceKeyLast seenRaw
ZivaHuboai:figshare.com:article/3264155710 d agoJSON v1
Deakin Research Onlineoai:figshare.com:article/3264155710 d agoJSON v1
DMU Figshareoai:figshare.com:article/3264155710 d agoJSON v1
FieldAssertionExtractorEvidence
concepts[field].local:field:earth-environmentalmapping · zivahub uct ac zaconnector:zivahub_uct_ac_za@1.0.0
concepts[field].local:field:earth-environmentalmapping · figshare dmu ac ukconnector:figshare_dmu_ac_uk@1.0.0
concepts[field].local:field:earth-environmentalmapping · dro deakin edu auconnector:dro_deakin_edu_au@1.0.0
concepts[modality].local:modality:audioenrichment · zivahub uct ac zakeyword-concept-rules@1.0.0title+description (75%)
descriptionsource · zivahub uct ac zaconnector:zivahub_uct_ac_za@1.0.0/metadata/dc/description
license_textsource · zivahub uct ac zaconnector:zivahub_uct_ac_za@1.0.0
publication_datesource · zivahub uct ac zaconnector:zivahub_uct_ac_za@1.0.0
titlesource · zivahub uct ac zaconnector:zivahub_uct_ac_za@1.0.0/metadata/dc/title