• Published September 1st, 2016

Jeremy E. Schaffer, Chase Lockwood, Wayne Buchan
Fort Wayne Metals Research Products Corp, Fort Wayne, IN, USA

Introduction

Fort Wayne Metals is engaged in alloy design, process development, and thermomechanical conditioning and test development of low through high temperature nitinol and NiTi ternary alloys for actuator applications. Custom product forms range from ultrafine filament (e.g. 50 µm) through larger wire (e.g. 2-5 mm), cables, strip and other custom product forms. The present work on low temperature actuation using superelastic grade NiTi is adapted from a talk given by the authors at SMST 2015 [1].

Implementation of both small and large scale Nitinol actuators is gaining interest due to the high value offered compared to alternative technologies [2, 3]. Nitinol provides high force-to-weight actuation compared to alternative technologies: the same amount of work can be done by a solid state SMA element with ten times less weight than an electrical or hydraulic system. Actuator design using shape memory alloy (SMA) wire requires careful understanding of alloy selection and processing, loading rates and axes, environmental variation and overall system stability. It must be recognized that the factors driving design complexity can be harnessed to provide powerful application-specific tuning. Strain and/or load evolution during SMA element actuation is critical to both system stability and durability. It is known today that actuation is possible for millions of cycles at modest conditions of stress, temperature, and stroke length. In spite of this, durability testing and validation of these materials in application-specific designs is costly because of slow cyclic test frequencies (often less than 0.1 s-1). This work is an early step exploring stabilization behavior of a Ni-rich, high stress (700 MPa), low temperature (290 to 430K), binary Ti49.2Ni50.8 actuator wire at 0.1 and 0.3 Hz test frequencies.

Materials and Methods

Ti_49.2Ni_50.8 (hereafter NiTi#1) nitinol wire was tested in a uniaxial test rig at 0.1 s^-1 and 0.3 s^-1 actuation rates in order to understand strain evolution during near-isobaric conditions. A thin wire diameter of 0.25 mm was selected in order to enable enhanced heat transfer and thus response time during high thermal gradient testing. Figure 1 shows the room temperature constitutive behavior of the binary NiTi#1 wire pulled to 7% strain, reversed to zero stress, and pulled to failure at a strain rate of 10^-3s^-1. The order of curves a-d also corresponds to lower internal stress in the initial annealed state and diminishing ultimate wire strength. Grain sizes of the samples given by a, b and d are nominally 50, 100, and 1000 nm respectively. Aging of the wire giving curve Fig. 1c was conducted at about 500 K for 600 ks, compared to the un-aged sample producing Fig. 1d. Active Af temperatures are given in the fig. 1 legend.

Figure 1: Uniaxial engineering stress-strain behavior for examined conditions of Ø 0.25 mm, NiTi#1 wire tested at 298K at a strain rate of 10^-3s^-1. All wires were annealed under 70 MPa uniaxial tension, reel-to-reel, in a dry argon atmosphere for less than 60 seconds. Fig. Label (Anneal Temperature K , label, grain size nm, Af K): a (773, GS1, 50, 283), b (873, GS2, 100, 274), c (973, GS3+age, 1000, 290), d (973, GS3, 1000, 270).

Results

Figure 2 shows the actuation strain behavior as a function of accumulated cycles. In each curve, the top of the cyclic data range gives the fully-extended (cold) position, while the lower boundary gives the contracted position in strain. Less than 0.2% strain evolution was observed in the low temperature boundary for condition GS1 over 100 cycles compared to 3% strain for condition GS2 and about 16% for GS3. The stroke length after 100 cycles remained more constant for the GS3 samples, where the aged sample maintained double the steady-state stroke (2.6%) compared to the un-aged condition (1.3%). A significant correlation between actuation magnitude and grain size was also observed with significantly improved stroke in ultrafine grained metal. This result requires further investigation and may be related to variant selection, strain evolution, and wire texture.

Figure 2: Load-biased actuation strain versus cycle count data for Ø 0.25 mm NiTi#1 wire tested at temperature limits of 290 K (top of plots) to 433 K (bottom of plots) at thermally-induced rates of 0.1 s^-1 and 0.3 s^-1 with a constant applied stress of 700 ± 25 MPa. 50 nm GS1 wire (a); 100 nm GS2 wire (b); Comparison of GS1, GS2, GS3 and reduced evolution associated with low temperature aging in GS3 + age

Conclusion

Similar strain evolution and slightly contracted (thermally limited on cooling) stroke occurs in Nitinol wire during actuation at a constant uniaxial stress of 700 MPa and thermal fluctuation test rates ranging from 0.1 to 0.3 s^-1. Much work remains to explore the usefulness of a higher rate test and/or actuator training mode.

References:

1. J. E. Schaffer, et al., "High Rate Testing of NiTi Actuator Wire." The International Conference on Shape Memory and Superelastic Technologies (SMST),(May 18-22, 2015). ASM, 2015.
2. J. Mabe, et al., "Flight test of a shape memory alloy actuated adaptive trailing edge flap Part I," in The International Conference on Shape Memory and Superelastic Technologies, Asilomar, CA (2014). (2013).
3. O. Benafan, et al., "Shape memory alloy actuator design: CASMART collaborative best practices and case studies," Int J Mech Mater Des, vol. 10 (2014), pp. 1-42. 

Click here to see previous highlights.

Disclaimer: Our monthly highlights are sneak peeks of what our R & D department is working on. This does not mean we have what is referenced above ready for manufacturing.