Smart materials activation analysis on example of nickel and titanium alloys

This paper is focused on research concerning activation time of elements made of Ni-Ti alloy (55/45% vol.) The activation time is a period of time required for alloy to reach it’s austenitic transformation (Af) temperature. For examined wire it reached values up to 60 °C. Heating of NiTi wire was conducted by retaining heat. In this paper the influence of wire length and electric current power on heating time is presented. This research allows to determine the correlation between the increase of temperature and time. For given electric current values. This data is useful for effective design of SMA actuators..

temperature Af the spring returns to the programmed shape, as a result of reverse transformation of martensite to austenite. The austenite finish temperature (Af) of the analysed alloy was ca. 60°C [4]. The easiest way to obtain such temperature is by using the test specimen as a resistor and thus increase the temperature of the wire by passing current. Uneven distribution of temperature in the initial phases of heating is a disadvantage of this method (Fig. 1).

Test procedure
The objective of these experiments was to investigate the effect of the duration of flow of electric current through the Ni-Ti alloy wire on its final temperature. The effect of the current on the time needed to reach the Af temperature was also investigated. The test specimens were prepared from 1 mm Ni-Ti wires in two different lengths, supplied from a regulator providing controlled current and voltage values (Table 1). For comparison 33 mm long controls were made of the same wire. Where: L -length of tested wire, I -current, U -voltage, P -power, ΔP -power measurement uncertainty.
The test stand was based on textolite plate acting as an insulator and used for mounting wires of the specified lengths. Power was supplied by Voltcraft PS 1440 benchtop power supply unit, connected to the test specimen by cables. The last element of the test stand was the Reveal PRO thermal imaging camera from Seek thermal whose role was to record the variation of temperature of the tested wire (Fig. 2). According to the test procedure, specimens of different lengths were mounted on the textolite base plate. Then the electricity was applied at the specified current and voltage levels. For each power level the heating process was repeated five times. The time needed to increase the wire temperature from 24°C to ca. 70°C was the measured parameter.

Results of experiments
The research program comprised series of tests performed on Ni-Ti wires of different lengths, namely 33 mm, 66 mm and 100 mm. All the test specimens were made of straight sections of 1 mm wire. Examples of time-temperature curves are presented in Fig. 3.
The experimental results were used to determine the time needed to increase the temperature of wire made of Ni-Ti alloy up to the activation temperature of ca. 70°C. The analysis of results started with determining of effect of the wire length on the time needed to increase the specimen temperature to Af (Fig. 3). The results allow us to conclude that the length of wire has a significant effect on the length of time needed to reach the activation temperature. The values recorded at 6,6 W power input were: ca. 45 sec. for 100 mm length, ca. 22 sec. for 66 mm length and ca. 35 sec. for 33 mm length. With the shortest specimens (33 mm) the whole test set-up acted as a radiator transferring heat to the environment. As a result, it took longer to heat up the 33 mm long specimen than the 66 mm long one. Consequently, the heating time depends on the ambient temperature and the volume of parts to which the Ni-Ti wire is connected.   4 presents the time-temperature curves obtained in heating 100 mm long specimens of Ni-Ti alloy up to the activation temperature at different power input levels. Heating the specimens with 6.5 A and 1.2 V current (9.75 W) produced time-temperature curves having a curved (bulged) shape. Higher values of current make the curve more linear, allowing determination of the trendline slope. The trendline slope depends on the level of power applied to the Ni-Ti wire. For input powers above 10 W the curve can be treated as linear. There is over 98% match between the trend line established for such power inputs and the experimentally determined curve. This approximation allows us to assume possibility of modelling the behaviour of nitinol wire specimens as a function of their length and power input level.
The test results were used to draw the curve representing the effect of power input on the time needed to heat up the specimen to the activation temperature. The obtained curve was subjected to linearization following determination of slope as a function of time. The slope depends on the values of applied current and voltage. Thus, by controlling these parameters we can control the behaviour of the specimen made of NI-Ti alloy. Over 99% fit is established between the experimental values and the values obtained with the function of slope variation vs. input power. The graph in Fig. 5 presents the variation of the trendline slope as a function of input power. The experiment included also evaluation of the effect of low input power values (not exceeding 9 W) on the degree of bulging (curvilinearity) of the time-temperature curve. Such bulging of curve for low input power levels results from ineffective heating of the specimen with excessive loss of heat in the final stage of the heating process. This was caused by too small current which did not made up for the heat losses determined by the test stand elements and by the ambient temperature of ca. 23°C. That the curved shape (bulging) of the time-temperature curve can be attributed to insufficient current can be confirmed by comparing 100 mm and 66 mm long samples (Fig. 6). The time-temperature curves (Fig.6) show that with the decreasing volume of specimen smaller currents become more effective.