0
Research Papers

# The Design and Experimental Validation of an Ultrafast Shape Memory Alloy ResetTable (SMART) Latch

[+] Author and Article Information
John A. Redmond

Department of Mechanical Engineering, University of Michigan, 2350 Hayward Street, 2250 G.G. Brown, Ann Arbor, MI 48109-2125jredmond@umich.edu

Diann Brei

Department of Mechanical Engineering, University of Michigan, 2350 Hayward Street, 2250 G.G. Brown, Ann Arbor, MI 48109-2125dibrei@umich.edu

Jonathan Luntz

Department of Mechanical Engineering, University of Michigan, 2350 Hayward Street, 2250 G.G. Brown, Ann Arbor, MI 48109-2125jluntz@umich.edu

Alan L. Browne

Research & Development, General Motors, MC 480 106 256, 30500 Mound Road, Warren, MI 48090-9055alan.l.browne@gm.com

Nancy L. Johnson

Research & Development, General Motors, MC 480 106 256, 30500 Mound Road, Warren, MI 48090-9055nancy.l.johnson@gm.com

Kenneth A. Strom

Research & Development, General Motors, MC 480 106 256, 30500 Mound Road, Warren, MI 48090-9055kenneth.a.strom@gm.com

J. Mech. Des 132(6), 061007 (May 25, 2010) (14 pages) doi:10.1115/1.4001393 History: Received September 25, 2009; Revised February 12, 2010; Published May 25, 2010; Online May 25, 2010

## Abstract

Latches are essential machine elements utilized by all sectors (military, automotive, consumer, manufacturing, etc.) with a growing need for active capabilities such as automatic release and reset, which require actuation. Shape memory alloy (SMA) actuation is an attractive alternative technology to conventional actuation (electrical, hydraulic, etc.) because SMA, particularly in the wire form, is simple, inexpensive, lightweight, and compact. This paper introduces a fundamental latch technology, referred to as the T-latch, which is driven by an ultrafast SMA wire actuator that employs a novel spool-packaged architecture to produce the necessary rotary release motion within a compact footprint. The T-latch technology can engage passively, maintain a strong structural connection in multiple degrees of freedom with zero power consumption, actively release within a very short timeframe ($<20 ms$, utilizing the SMA spooled actuator), and then repeat operation with automatic reset. The generic architecture of the T-latch and governing operational behavioral models discussed within this paper provide the background for synthesizing basic active latches across a broad range of applications. To illustrate the utility and general operation of the T-latch, a proof-of-concept prototype was designed, built, and experimentally characterized regarding the basic functions of engagement, retention, release, and reset for a common case study of automotive panel lockdown. Based on the successful demonstration and model validation presented in this study, the T-latch demonstrates its promise as an attractive alternative technology to conventional technologies with the potential to enable simple, low-cost, lightweight, and compact active latches across a broad range of industrial applications.

<>

## Figures

Figure 1

Diagram of the T-latch shown with its main components. The latch connects two bodies, one body is connected rigidly to the gate and the other is connected rigidly to the upper plate. The torsional reset spring is mounted between the T-shaft and the upper plate such that it opposes the torque applied by the SMA wire. Right of the figure, nomenclature and variable definitions affecting operation behavior models are defined.

Figure 2

Graphical representation of moments on the T-latch’s SMA actuator throughout operation. The pathway of relative moments and positions of the actuator throughout the latch operation is represented qualitatively. The path between the release and austenite equilibrium points is complicated by dynamics, heating, and stress dependency of phase fraction, and is thus represented as a region.

Figure 3

T-latch operation cycle. The diagram illustrates the state of engagement, the T rotation angle ϕ, and the material phase of the wire (blue for martensite during engagement, retention, and reset, and red for austenite during release). The reset spring and upper plate are omitted from the diagram. The top row of diagrams shows side views while the bottom row shows top views.

Figure 4

Forces and moments that the T encounters during operation, and nomenclature for the key loading directions. The loads shown occur due to engagement, release, and reset operations and do not necessarily occur at the same time.

Figure 5

Schematic of the T-latch with a general applied load for the plane of shoulder and normal to shoulder loading cases. The key applied loads and locations of stresses are included.

Figure 13

Example of typical release data. The latch was configured with a 300 N separation force and 180 deg predeflection on the reset spring. In this trial, the actuator was heated for 18 ms before the phase began to transition, rotated fully 6 ms later, and released in a total of 26.9 ms.

Figure 14

Latch speed test results. The T-latch meets the specified 30–50 ms release time with 18 ms fastest measured release. The data indicate that a nearly constant 24 J of energy is required for release regardless of current, voltage, or average power supplied.

Figure 15

Release time as a function of pull-up force. The release and preheat times increased due to the separation forces while the transition time declined. Tests were conducted with 9 A excitation current and a 0.5 N mm/deg torsional reset spring.

Figure 16

Range of motion results. The motion angle was measured for varying loads on the latch which appear to be most sensitive at low tensions and level off at higher tensions.

Figure 17

Effect of reset spring on reset time. The reset time was observed to have a decreasing relationship with the preload on the reset spring and full reset occurring within 1 min. Data bars represent 1 standard deviation from the mean and each point represents data averaged from 10 to 20 trials.

Figure 6

Free body diagram of a differential element of SMA in sliding contact with the spool. The key forces are shown with the direction of friction dependent on the material phase of the SMA wire.

Figure 7

Photograph and diagram of the T-latch proof-of-concept prototype

Figure 8

Simplified constitutive model and experimental stress-strain data for selected SMA wire

Figure 9

Effect of reset spring preload on engagement force. The data at each spring load were averaged with the data bars spanning a standard deviation. The slope of the linear fit corresponds to a ramp coefficient of friction μr=0.43. The error between the model and data fit may have been caused by error in locating the zero deflection position of the reset spring.

Figure 10

T-latch retention test specimens. Three specimens are shown: (a) a typical specimen prior to loading, (b) a longitudinal specimen failed due to transverse shear in the shoulders, and (c) a combined load specimen failed due to transverse shear in the shoulders and at the edge of the shaft.

Figure 11

Retention test fixtures

Figure 12

Example of typical data collected for a single T-latch operation cycle. The trial demonstrated all four operation stages with full reset taking about 1 min.

## Discussions

Some tools below are only available to our subscribers or users with an online account.

### Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related Proceedings Articles
Related eBook Content
Topic Collections