Experimental Investigation of a Barrier Structure

Experimental Investigation of a Barrier Structure Based on a Shape Memory Alloy Actuator 1

Sonia Degeratu1, Daniela Tarnita1, Costel Caramida2, Irina Boncea1, Laurentiu Alboteanu1, Monica Staicus1 University of Craiova, Romania, 2National Institute for Research, Development and Testing in Electrical Engineering, Craiova, Romania {sdegeratu, lalboteanu}@em.ucv.ro, {tarnita.daniela, monica.staicus}@gmail.com, {irina.boncea, costica_caramida}@yahoo.com

Abstract-The use of Shape Memory Alloy (SMA) materials for actuation represents a technological opportunity for the development of innovative applications. The issue of using the SMA actuators is an enticing solution that has attracted the interest of specialists in recent years, due to their numerous advantages (i.e. reliability, compactness and flexibility, reduced cost, linear or angular movement etc.). In addition, their unmatched energy density, easy integration into various structures and the simplicity in terms of design, have all motivated the authors to develop a barrier structure model, actuated by three SMA helical springs. The actuator used in this structure works as a linear actuator, contracting with great strength and speed, thus exerting the necessary force to lift the barrier arm in the SMA springs heat-activated, austenitic state. The first part of this paper provides a description of the accomplished model and of its functioning principle. The second part of the study presented in this paper focuses on the authors’ investigation on how the performances of the actuator inserted in the barrier structure—namely the actuator’s heating time, cooling time, actuation time and stroke—are affected by the SMA spring activating direct current values. The tests were conducted while keeping the geometrical parameters of SMA springs and the weight of the barrier arm constant.

I.

INTRODUCTION

In recent years, the need for miniaturization has led to the creation smart material-based actuators, which allowed for the construction of devices on a smaller scale. The actuators employing shape memory properties represent one of the most promising new actuator technologies developed in the last few years [1]. The interest for the SMA actuators derives from their advantages, as compared to the traditional actuators: high plastic deformation, the amount of force generated, and the production of mechanical work, high power-to-weight ratio, small size, clean silent operation and mechanical simplicity [2]-[8]. On the other hand, they also display some disadvantages–including hysteresis and slow speed–which must be taken into account in any application [2]. All the above mentioned advantages are results of the SMA’s unique thermo-mechanical properties, such as the shape memory effect (SME), where large residual strain can be fully recovered upon raising the temperature after the loading and unloading cycle, and the pseudo-elasticity or super-elasticity, where a very large strain is fully recovered after loading and unloading at constant temperature [6], [8]–[13]. The cause is a martensitic phase transformation

between a high temperature parent phase, austenite (A), and a low temperature phase, martensite (M). In the absence of stress, the start and finish transformation temperatures are denoted Ms, Mf (martensite start and martensite finish) and As, Af (austenite start and austenite finish) [10] and [12]–[14]. Their unique properties allowed SMAs to be utilized in exciting, innovative and numerous applications: robotics, the automotive industry, high precision engineering (positioning systems), biomedical engineering, machine craft, electromechanical engineering etc. [10], [11], [13] and [15]–[20]. The innovative ideas regarding the applications of SMAs, the number of SMA products on the market in continual growth and the advances in the manufacturing of shape memory alloys are elements that have encouraged the authors of this paper to develop a new barrier structure actuated by SMA helical springs, dedicated to private parking systems. This barrier structure is an improved version of the experimental model presented in the papers [21] and [22]. The authors intensified the research in this direction given the great interest shown by some Romanian companies that produce such barriers. The study presented in this paper is focused on how the performances of the actuator inserted in the barrier structure—namely the actuator’s heating time, cooling time actuation time and stroke—are affected by the SMA spring activating direct current values. The influence of direct current on reaction time values of the SMA spring actuator has been studied by other specialists, e.g. by Yates and Kalamkarov in [23], but their investigation focused on the case of an actuator using only one SMA spring which was not inserted in a particular structure. In the current study, the actuator consists of three SMA springs integrated into the barrier structure, thus allowing the designer to control the direction of actuation, the amount of force generated and the stroke of the actuator, not only by the various combinations that can be achieved with these three SMA springs but also by the choice of an adequate structure of the entire driving mechanism block. The three SMA springs in question are connected in a series circuit from an electrical point of view and are part of a parallel connection from a mechanical point of view. The authors also provide a brief description of the accomplished barrier structure in order to understand the functioning principle and the possibilities of command and control of this experimental barrier.

II.

DESCRIPTION OF THE EXPERIMENTAL BARRIER STRUCTURE

In this section, the authors provide a brief description of the barrier structure experimental model. Also, they comprehensively explain how the automatic motion of the barrier arm is controlled, and how an SMA spring actually works while emphasizing the advantages that such a device displays when used as actuator. The SMA spring boasts a number of advantages, among which we mention the active shape-change control of the spring, the possibilities for miniaturization, ease of integration into a system structure, the automatic control of the barrier arm motion (by employing a programmable logic controller). Thus a visible increase in the efficiency of such barrier structures is made obvious. The so achieved model allows the structure behavior analysis to be made in various drive conditions and control of the SMA springs. The experimental model of the accomplished barrier structure is shown in Fig. 1. The main positions marked in Fig. 1 are described in detail in [24]. This section provides only the information regarding the main components of the experimental barrier structure that is necessary for understanding the functioning principle and the possibilities of command and control of the analyzed structure. A. The Control Unit The control unit block with its outputs is shown in Fig. 2. This block (position 2 in Fig. 1) contains the following sources: for the supply of the SMA springs (5V), for LOGO (24V) and for powering the barrier control transmitter (24V). B. Data Acquisition System The Data Acquisition System (position 3 in Fig. 1) used in the accomplished barrier structure is a Velleman 4 CHANNEL SIGNAL RECORDER. In this Data Acquisition System (DAS), the signals can have four separate channels, but only two were used to conduct the experiments. Channel 1 indicates the signal from

Fig. 1. Latest variant of the experimental arrangement of Ni–Ti SMA-based barrier: 1-seating base; 2-control unit; 3-Signal Recorder (Data Acquisition System); 4-remote control; 5-control receiver; 6-Logo!Power; 7-barrier arm; 8–Siemens Logo!24Co (Programmable Logic Controller-PLC); 9-SMA spring-actuated mechanism unit.

Fig. 2. The control unit of the accomplished model: 1, 2- outputs for the stroke transducer; 3, 4-outputs for the supply of the SMA springs; 5, 6- LEDs for signaling barrier position (red - closed, green - open); 7,8- terminals for connection to an independent source (variable voltage) for powering the SMA springs.

the race transducer, and channel 2, the signal from the power supply of the SMA springs. The main features, the hardware and software specifications of the Data Acquisition System and the system requirements were detailed in [24]. C. AD-IR-DRIVER04 Module for the Remote Control This component (position 5 in Fig. 1) is an electronic module with a microcontroller allowing the command of 4 independent channels using infrared remote controls. The module is provided with 4 relays of 5A/250V AC, which can operate various elements of execution. The command of the relays, the remote programming and the main technical characteristics of this module were presented in the paper [24]. D. Programmable Logic Controller-PLC The PLC used in the analyzed barrier structure (position 8 in Fig. 1) is the Siemens LOGO!24Co module, an ideal controller for simple automation tasks in the industry and building services. PLC LOGO! 24Co (with 8 inputs, 4 outputs and 24 V DC input/supply voltage) is compact, easy to use and provides a low cost solution for controlling tasks of low complexity. Together with the LOGO! Soft Comfort software, the configuration of the logical module is simply intuitive: program generation, project simulation and documentation are accomplished using drag and drop functions, allowing maximum ease of operation. Description of the command LOGO! Soft of the accomplished model is as described below. The LOGO!Soft program for the barrier command is presented in Fig. 3. The digital input I3 controls the barrier (up/down) depending on the signal from the receiver output which in turn is driven by the remote control. The signal from the stroke transducer is connected to the AI1 analog input. Through B004 Analog Threshold Trigger block, the output Q1 is set or reset depending on the threshold triggers On / Off corresponding to the position of the barrier, the SMA springs being powered up or not. Barrier position is determined by the stroke transducer (Tc) positioned on the axis of the arm. The transducer supply voltage, UTc, is 5V. The transducer stroke is 3600. Lifting barrier arm angle was set at 860, resulting in:

Fig.3. LOGO!Soft barrier command.

U Tc * On = U Tc * 86 = 1.2 V.

(1)

The amplification parameter of the Analog Threshold Trigger block is 100. Resultantly, the output Q1 is 1 if the actual value of the input functions Ax is greater than Threshold On.

Threshold On = U Tc *100 = 120 ,

(2)

Threshold Off = 118 .

(3)

The B002 and B003 Analog Threshold Trigger blocks command, at the outputs Q2 and Q3, the LEDs functioning which indicates the barrier position, respectively the red LED for the lowered position of the barrier and the green LED for the raised position of the barrier. These LEDs function as elements which signal when it is possible or not to go through the space of the barrier. E. LOGO!Power Module The mini power supply devices designed into the LOGO!POWER module (position 6 in Fig. 1) provide great performance in the smallest space and excellent efficiency over the complete load range. Its technical characteristics were presented in [24]. F. Driving Mechanism Block The mechanism block driving the barrier (position 9 in Fig.1) is displayed in Fig. 4. The actuator of this block relies on three SMA Electric Pistons (3) rigidly set at their pistonfree ends on the fixed support (4). The opposing ends of the SMA Electric Pistons have been mounted on a mobile support (5), with the traction wires (6) departing in the direction of the driving barrier arm system (8) endowed with a rotating stroke transducer, Tc (7). This transducer is

powered by 5V and it displays a maximum stroke of 3600. The accomplished mechanism enables an 860 rotation of the barrier arm at a complete stroke of 16mm of the SMA Electric Pistons (when electrically activated). The central element of this experimental arrangement is the represented by the SMA Electric Pistons (Fig. 5), an SMA driving element that is frequently used [25]. The SMA Electric Piston was produced by Mondo-Tronics, Inc. [26] and [27]. 1. Functioning description of the SMA Electric Piston The SMA Electric Piston is a linear actuator mechanism that shortens in length with great strength and speed when it is activated by a direct electric current. An SMA spring placed inside makes all these possible. The SMA spring displays two entirely different forms or "phases" at the distinct temperatures Mf and Af. At the "low" temperature (Mf), the SMA spring is extended, and can be stretched easily or deformed by a small force. When raised to the "high" temperature Af, by applying a direct electric current, the SMA spring changes to a much harder form. In this phase, it shortens in length and exerts the necessary force to lift the barrier arm. The SMA Electric Piston, used in our model, implements an SMA spring working in tension to provide a force of up to 4.5N, while the SMA Electric Piston itself weighs only 0.1N. Nonetheless, the total force developed by the driving mechanism is 13.5N, due to the way the mechanical coupling of the three SMA Electric Pistons. 2. SMA springs operating lengths determination In this section the operating lengths of the SMA spring are calculated at low temperature (in martensite phase) and high temperature (in austenite phase). In the analyzed model, a linear force-deflection behavior is assumed in order to simplify the analysis [10], [13] and [14].

Using these values, it is possible to calculate [10], [13], and [25]: − the spring rates at high and low temperature, Kh and Kl

Kh =

Kl =

Gh ⋅ d 4 8⋅ n ⋅ D3 Gl ⋅ d 4 8⋅n ⋅ D3

= 1.933 N/mm ,

(4)

= 0.429 N/mm .

(5)

− the spring deflections at high and low temperature, δ h and δ l

δh =

F = 2.55 mm , Kh

δ l = δ h + S = 18.55 mm . Fig. 4. Driving mechanism block of the experimental barrier structure.

(7)

− the high and low temperature lengths, Lh and Ll The length of the spring when it’s fully compressed at high temperature, Lf, is given by:

L f = d ⋅ ( n + 1 ) = 6.4 mm .

Fig. 5. The SMA Electric Piston.

The three SMA springs must develop, at high temperature, the necessary force F to lift the barrier arm with an angular experimental displacement of 860, which corresponds to an SMA springs linear stroke S of 16mm, in the direction of the SMA actuator slider axis. By using the mechanical model of the barrier structure presented in [21], we have adapted it to the experimental arrangement in Fig. 1, resulting in a value of 4.92N for this force F in static regime. Consequently, there is enough reserve for the transition to a barrier structure closer to the real dimensions, given the fact that the force developed by the three SMA springs is 13.5N. It is known that for any SMA element its phase transitions during heating and cooling regimes and, correspondingly, its contracting and relaxing regimes are, in thermodynamic terms, non-quasi-static processes. Therefore, it is considerably challenging to issue any sort of thermodynamicbased constitutive equation having the capabilities to directly connect the transversal shear modulus to thermodynamic temperature [16]. That is why, in what follows, we shall calculate the SMA spring operating lengths only for Af and Mf temperatures. The differential scanning calorimetry (DSC) method was used to determine these temperatures, and the experimental results presented in [22], indicate that their values are: Af=66.80C and Mf=27.10C. For these temperatures, the values of shear modulus are Gh=16920MPa and Gl=3753MPa, respectively. The SMA spring data values Catalog (© Mondo-Ttronics, Inc.) are: wire diameter d=0.8mm; average diameter D=4mm, number of active turns n=7.

(6)

(8)

Under this condition, the following expressions result for Lh and Ll:

III.

Lh = L f + δ h = 8.95 mm ,

(9)

Ll = Lh + S = 24.95 mm .

(10)

EXPERIMENTAL RESULTS AND DISCUSSION

Because the shape memory alloys are metal materials, they certainly possess a great resistivity (especially Ni-Ti), which means they can be heated and, therefore, activated by passing a direct current through them. The direct current value can be empirically evaluated for the specific ambient conditions in which the actuator must operate. For a specific type of alloy and under still and forced air conditions, in [13] Waram presents a more precise method for estimating the direct current value, both for the activation and cooling time values. Since the thermo-mechanical properties of shape memory materials vary greatly from alloy to alloy and with the specific ambient conditions, it is best recommended to evaluate the activation current and the reaction times by means of testing [25]. That is why the purpose of the present investigation is to determine, via experimentation, how actuator’s reaction times are affected by the SMA springs activating current value. In our experimental model, the electric current for powering the SMA springs can come from two sources: an internal power source or an external variable source.

This paper presents the test results obtained with the experimental arrangement, using an external 40V and a 5A stabilized power source, connected to the terminals (7) and (8) of the control unit shown in Fig. 2. In our tests, the voltage varied between a maximum and a minimum. The maximum voltage that can be applied to springs corresponds to the value at that, with the current through springs not exceeding a maximum limit of 4.2A, as imposed by the provider. The minimum value of the applied voltage is chosen so that the current that runs through the springs can be large enough to activate them, so as to ensure the angular displacement of the barrier arm by 860. The experiments consisted of determining the SMA spring reaction time values (tsc, ta, trel, and tr), at different values of the activating electric current. The race transducer converts the angular motion into electric voltage signal, as follows: 5V corresponds to 3600. DAS made it possible to emphasize the displacement of the barrier arm as a function of time for all activating currents. As an example, in Fig. 6 and Fig. 7 the obtained results using a supply voltage of the SMA springs of 1.48V and 2.79V respectively, are presented. As previously mentioned, in DAS the signals can have four separate channels, but only two were used to conduct the experiments. Channel 1, the curve marked with yellow in

Fig. 6. A complete up-down cycle of the barrier arm in the case of the voltage supply of the SMA springs of 1.48V: signal from Tc transducer (V); signal from power source of the SMA springs (V).

Fig. 7. A complete up-down cycle of the barrier arm in the case of the voltage supply of the SMA springs of 2.79V: signal from Tc transducer (V); signal from power source of the SMA springs (V).

Fig. 6 and Fig. 7, indicates the signal from the race transducer, and channel 2, the curve marked with green, indicates the signal from the power supply of the SMA springs. By using the two markers from the transducer race signal we were able to determine the values for all SMA spring functioning time periods: tsc, ta, trel, and tr. The values of the voltages are indicated by the voltage markers. As an example, in Fig. 8 the two transducer markers indicate the value for the time to reset, tr=13.56s, and the two voltage markers indicate the value of voltage applied to the springs by 2.79V. Fig. 9 displays an up-down functioning cycle of the barrier arm in the case of powering the SMA springs at the maximum voltage of 2.98 V, with a delay in the return command of the barrier arm. One can observe the voltage impulse which ensures the retention of the barrier arm around the 860 position, within the On-Off threshold positions of the Analog Threshold Trigger block. All results are detailed in Table 1, where the parameters have the following meaning: tsc, time to start contracting, or the necessary time from the start of current application to reach the temperature As; ta, time to actuate, or the contraction time, or the necessary time for the barrier arm to reach the angular displacement of 860; and trel, relaxing time, or the necessary time for the SMA spring to cool from a temperature greater than or at least equal to Af to the

Fig. 8. The value for the time to reset (tr=13.56s) indicated by the transducer markers, and the value for the applied voltage (2.79V) indicated by the voltage markers: signal from Tc transducer (V); signal from power source of the SMA springs (V).

Fig. 9. A complete up-down cycle of the barrier arm in the case of the SMA springs supply voltage of 2.98 V, with a delay in the return command of the barrier arm.

temperature Ms; tr, time to reset, or the necessary time for the arm to come back to its initial position. In this status, the SMA temperature is below Mf. In all cases, the environment temperature was of 23.20C. TABLE I REACTION TIME VALUES OF THE BARRIER ARM U [V]

I [A]

tsc [s]

ta [s]

trel [s]

tr [s]

1.48

2.11

21.56

42.81

7.5

13.54

1.83

2.57

10.16

8.75

7.5

13.57

2.13

2.96

9.31

6.13

6.94

13.58

2.79

3.83

3.75

1.03

5.93

13.56

2.98

4.02

3.69

1.00

5.88

13.55

By analyzing the results, it becomes obvious that, because the SMA spring activates through electric heating, the contracting time varies in a quite significant range with the applied current; the higher the current, the faster the heating, and the faster the contraction and so, the faster the stroke of the barrier arm. The relaxing time values are not much different from each other and that is because the values of final heating temperatures ((Tfinal>Af)) do not differ much, either, in each and every case. Resetting times are approximately the same, because in all cases the conditions of reset, i.e. the weight of the barrier arm and the ambient temperature, were the same. As far as the barrier arm behavior is concerned, the optimal choice is that of a power supply with a value of 2.79 V for the SMA springs. IV.

CONCLUSIONS

The actuator used in our experimental barrier structure, is based on three SMA Electric Pistons. The SMA Electric Pistons actuator provides efficiency in terms of energy, weight (as it is always lighter), space and cost savings, and noise-free operating as compared to a wide variety of alternative products, all while developing a force of up to 4.5N. The value of this force seems to be large for microscale applications, while it is smaller for macro applications in the loading realm [25]. The total force developed by the actuator inserted in our barrier structure is 13.5N, due to the mechanical coupling of the three SMA Electric Pistons. This value of the force has enabled the transition to a barrier model that is closer to the average scale as displayed in Fig.1. The authors manifest confidence that in the future a real scale model will be accomplished, which will become possible, on the one hand, due to the various combinations that can be achieved with the SMA springs and, on the other hand, due to the choice of an adequate structure for the entire driving mechanism block capable of providing the desired level of actuation. Applications such as the NASA Pathfinder rover [28] and the drive mechanism (using SMAs) for use in a solar tracking device [29] come in support of this belief. Upon analyzing the experimental results, it has become obvious that our model behaves quite well in both cases of large and small activating currents. For a given barrier

structure, the choice of a specific value for SMA activating current will be made according to the specific requirements of each customer, relative to the values of SMA actuator reaction times. This new barrier structure may prove potential usefulness in: toys, parking spaces, toll gates, bridge barriers, apartment buildings access systems etc. ACKNOWLEDGMENT This work was supported by the Grant number 7C/2014, awarded in the internal grants competition by the University of Craiova. REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]

[12] [13] [14] [15] [16]

[17]

[18] [19]

[20]

B. Donmez, B. Ozkan, and S. Kadioglu, “Precise position control using shape memory alloy wires,” Turk J Elec Eng & Comp Sci, vol. 18, no. 5, pp. 899-912, 2010. S. J. Yates and A. L. Kalamkarov, “Improving the speed of shape memory alloy actuators by faster electrical heating,” Springer Tracts in Advanced Robotics, vol. 21, pp. 67-76, March 2006. E. Torres-Jara, K. Gilpin, J. Karges, R.J. Wood, and D. Rus, “Compliant Modular Shape Memory Alloy Actuators,” IEEE Robotics & Automation Magazine, vol. 17, pp. 78–87, December 2010. K. Seung and K. Byungkyu, “Design parametric study based fabrication and evaluation of in-pipe moving mechanism using shapememory alloy actuators,” J. Mech Sci., vol. 22, pp. 96–102, 2008. W. Huang, “Shape Memory Alloys and Their Application to Actuators for Deployable Structures,” Ph.D. Thesis, Univ. of Cambridge, 1998. C. Mavroidis, “Development of actuators using shape memory alloys and electrorheological fluids,” J. Res Nondestr. Eval., vol. 14, pp. 1–32, 2002. G. Knowles, R. Bird, and V. Birman, “Shape memory alloy springs used as reduced power/weight actuators,” in Proc. of ASME congress, Anaheim, California, USA, pp. 17–25, November, 2004. K. Yang K, CL. Gu, “A compact and flexible actuator based on SMA springs,” J Mech. Eng. Sci., vol. 222, pp.1329–1337, 2008. H. Morawiec, "Structure formation and its effect on properties of shape memory alloys", Physics of Advanced Materials, Winterschool, Thesaloniki, Greece, pp. 14-18, January 2008. S. Degeratu and N. Bizdoac, Shape memory alloys: fundamentals, design and applications, Universitaria Press, Craiova, Romania, 2003. H. Wu Ming and L.McD. Schetky, "Industrial applications for shape memory alloys", in Proc. of the International Conference on Shape Memory and Superelastic Technolgies, Pacific Grove, California, pp. 171-182, 2000. H. Funakubo, Shape Memory Alloys, Gordon and Breach Science Publishers, 1987. T. C. Waram, Actuator Design Using Shape Memory Alloys, Ontario Press, Canada, 1993. K. Otsuka and C. M. Wayman, Shape Memory Materials, chap. 2, Cambridge Univ. Press, 1999. G. Songa, N. Ma and H.-N. Li, “Applications of shape memory alloys in civil structures,” Eng. Structures, vol. 28, pp. 1266–1274, 2006. J. Huang, X. Chen, and L. Zhong, “Analysis and Testing of MR Shear Transmission Driven by SMA Spring,” Advances in Materials Science and Engineering, vol. 2013, 6 pages, 2013. L.G. Machado and M.A. Savi, “Medical applications of shape memory alloys”, Brazilian Journal of Medical and Biological Research, vol.36, pp. 683-691, 2003. D. Tarnita, D.N. Tarnita, R. Tarnita, C. Berceanu, and F. Cismaru, “Modular adaptive bone plate connected by Nitinol staple,” Materialwissenschaft und Werkstofftechnik, vol. 41, 2010. S. Degeratu, et al., “Active Solar Panel Tracking System Actuated by Shape Memory Alloy Springs,” in Proc. of the IEEE International Conference on Applied and Theoretical Electricity, Craiova, Romania, 5 pages, October 2014. N. Bizdoaca, et al., “Robotic Finger Actuated with Shape Memory Alloy Tendon,” in Proc. of Soft Computing, Optimization, Simulation & Manufacturing Systems, Malta, pp. 897-906, 2003.

[21] S. Degeratu, et al., “Barrier structures using shape memory alloy springs,” in Proc. of the International Conference DEEE, Tenerife, pp. 367-372, December 2010. [22] S. Degeratu, P. Rotaru, S. Rizescu, and N. Bîzdoacă, “Thermal study of a shape memory alloy (SMA) spring actuator designed to insure the motion of a barrier structure,” J. of Thermal Analysis and Calorimetry, vol. 111, pp. 1255-1262, 2013. [23] S. J. Yates and A. L. Kalamkarov, “Experimental study of helical shape memory alloy actuators: effects of design and operating parameters on thermal transients and stroke,” Metals, vol. 3, pp. 123-149, 2013. [24] S. Degeratu, et al., “Structure Based on Shape Memory Alloy Actuator,” in Proc. of the IEEE International Conference on Applied and Theoretical Electricity, Craiova, Romania, pp.: 6, October 2016.

[25] J.R. Santiago Anadón, “Large Force Shape Memory Alloy Linear Actuator,” Ph. Master Thesis, Univ. of Florida, 2002. [26] Mondo-Tronics, Inc., Jameco electronics archive, Current Tech. Documents, pp. 1 and 2. ftp://ftp.jameco.com/Archive/Current-TechDocuments/141903.pdf [27] Jameco electronics catalog, vol. 624, October 2012, pp. 61. ftp://ftp.jameco.com/Archive/PreviousCatalogs/624catalog.pdf [28] P. P. Jenkins and G. A. Landis, “A rotating arm using shape-memory alloy,” NASA, 1995. [29] K. A. Harrisonburg, T. Benjamin, “Shape Memory Alloy Motor as Incorpoated into Solar Tracking Mechanism,” United States Patent, No. - US 7692091 B2, Apr.6, 2010.