Open Access
Article Number 06037
Number of page(s) 7
Section Applied Research and Applications: Testing and Modelling
Published online 01 September 2009
ESOMAT 2009, 06037 (2009)
DOI: 10.1051/esomat/200906037

Response of NiTi SMA wire electrically heated.

C. Zanotti1, P. Giuliani1, A. Tuissi1, 2, S. Arnaboldi1, 2 and R. Casati1, 2

1  Institute for Energetics and Interphases, Via R. Cozzi, 53 Milano (Italy)
2  Institute for Energetics and Interphases, Corso Promessi Sposi, 29 Lecco (Italy)

Published online: 1 September 2009

In the present work an experimental-numerical approach is used to study the thermo-mechanical behaviour of NiTi wire for defining what parameters are the most important in actuator designing. Tests were carried out heating, by an electrical current, a wire having a diameter of 150 μm, under constant stresses of 200 MPa. Data concerning strain, applied current and voltage are acquired during the tests by PC while wire temperature is recorded by thermographic system.
The numerical code integrates the ordinary differential equation describing the volumetric wire heating and cooling transients assuming that the electrical power is the only external heating source. Temperature distribution in the wire is considered uniform and its change in time depends on wire specific heat, latent heat of the Martensitic-Austenitic transformation, free convection heat exchange coefficient and total emissivity of the surface wire.
Comparisons between experimental and numerical results, obtained under different operating conditions, indicate that the wire heating rate mainly depends on the applied electrical power, latent heat of the Martensitic-Austenitic transformation, free convection heat exchange coefficient.
In the case of the cooling transient, the capability to decrease the wire temperature to reach the ambient value is strongly effected, once the free convection heat exchange coefficient is defined, by the shape of the curve representing the Austenitic-Martensitic transformation. Simulations of the wire behaviour point out that the time necessary to obtain the maximum strain can be reduce to few ms if current of 10 A is used. Moreover, the strain dependence on time can be tailored by choosing the suitable heating transient.

© Owned by the authors, published by EDP Sciences 2009