• 검색 결과가 없습니다.

Development of a non-explosive release actuator using shape memory alloy wire

N/A
N/A
Protected

Academic year: 2023

Share "Development of a non-explosive release actuator using shape memory alloy wire"

Copied!
9
0
0

로드 중.... (전체 텍스트 보기)

전체 글

(1)

Development of a non-explosive release actuator using shape memory alloy wire

Young Ik Yoo, Ju Won Jeong, Jae Hyuk Lim, Kyung-Won Kim, Do-Soon Hwang et al.

Citation: Rev. Sci. Instrum. 84, 015005 (2013); doi: 10.1063/1.4776203 View online: http://dx.doi.org/10.1063/1.4776203

View Table of Contents: http://rsi.aip.org/resource/1/RSINAK/v84/i1 Published by the American Institute of Physics.

Related Articles

High sensitivity microchannel plate detectors for space extreme ultraviolet missions Rev. Sci. Instrum. 83, 083117 (2012)

Invited Article: Electric solar wind sail: Toward test missions Rev. Sci. Instrum. 81, 111301 (2010)

Reflections of ions in electrostatic analyzers: A case study with New Horizons/Solar Wind Around Pluto Rev. Sci. Instrum. 81, 114501 (2010)

Modeling extreme ultraviolet suppression of electrostatic analyzers Rev. Sci. Instrum. 81, 045111 (2010)

Recent advances in direct solar thermal power generation J. Renewable Sustainable Energy 1, 052701 (2009)

Additional information on Rev. Sci. Instrum.

Journal Homepage: http://rsi.aip.org

Journal Information: http://rsi.aip.org/about/about_the_journal Top downloads: http://rsi.aip.org/features/most_downloaded Information for Authors: http://rsi.aip.org/authors

(2)

Development of a non-explosive release actuator using shape memory alloy wire

Young Ik Yoo,1Ju Won Jeong,1Jae Hyuk Lim,2Kyung-Won Kim,2Do-Soon Hwang,2 and Jung Ju Lee1

1Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology, 305-701, Guseong-dong, Yuseong-gu, Daejeon, South Korea

2Korea Aerospace Research Institute, 305-333, Eoeun-dong, Yuseong-gu, Daejeon, South Korea

(Received 21 August 2012; accepted 31 December 2012; published online 22 January 2013) We have developed a newly designed non-explosive release actuator that can replace currently used release devices. The release mechanism is based on a separation mechanism, which relies on seg- mented nuts and a shape memory alloy (SMA) wire trigger. A quite fast and simple trigger operation is made possible through the use of SMA wire. This actuator is designed to allow a high preload with low levels of shock for the solar arrays of medium-size satellites. After actuation, the proposed device can be easily and instantly reset. Neither replacement, nor refurbishment of any components is neces- sary. According to the results of a performance test, the release time, preload capacity, and maximum shock level are 50 ms, 15 kN, and 350 G, respectively. In order to increase the reliability of the actua- tor, more than ten sets of performance tests are conducted. In addition, the proposed release actuator is tested under thermal vacuum and extreme vibration environments. No degradation or damage was observed during the two environment tests, and the release actuator was able to operate success- fully. Considering the test results as a whole, we conclude that the proposed non-explosive release actuator can be applied reliably to intermediate-size satellites to replace existing release systems.

© 2013 American Institute of Physics. [http://dx.doi.org/10.1063/1.4776203]

I. INTRODUCTION

After satellites are launched by launch vehicles, several appendages, such as the fairing, certain antennas, and solar arrays should be separated successfully from the main body of the satellite so that it can perform its ordinary operations while in orbit. Most existing release devices use pyrotech- nics, possessing minimum volume/weight, providing instan- taneous operation on demand, and requiring little input en- ergy. However, these systems induce a high level of shock and contaminants and have costly handling requirements due to their hazardous nature.1Therefore, a non-explosive release actuator designed to replace pyrotechnic devices (termed ex- plosive actuators) has been investigated. Its favorable features over a pyrotechnic device are its low separation shock, lack of contamination, reusability, and safety for both the payload and the astronauts. Non-explosive release devices can be clas- sified into four categories according to the mechanism and the actuator. These are referred to as the (1) electrical spool and separation nut, (2) paraffin actuator, (3) shape memory alloy (SMA) device, and (4) thermal knife types.2–6Electrical spool and separation nut mechanisms perform the release of the segments quite rapidly (in about 1 ms). However, this fast action produces a large shock (over 1000 G). Paraffin actua- tors are voluminous and very slow in their performance. They also consume a considerable amount of energy. The thermal knife type can release the load with negligible shock, but load capacity is quite limited. Moreover, this actuator requires sig- nificant amounts of actuation time and energy. Among them, release actuators using SMA have been used due to their sim- ple mechanism, easy reusability, and resettability as well as

their unique material characteristics such as their large recov- erable strain and large force-to-volume ratio.

A shape memory alloy is a material that can return to its original shape above the phase transformation tempera- ture after a large deformation. This reversible transformation is repeatable provided the alloy does not experience exces- sive strain or temperature. Due to this phenomenon (the shape memory effect (SME)), these alloys are used in many sen- sors and actuators as a type of wire or as a spring, thin sheet, circular tube or cylinder. The maximum recoverable strain is usually reported as about 8% in the literature.7

Two types of SMA actuators are used. First, the trig- ger mechanism is fundamental to TiNi’s current product line of TiNi aerospace Pin-puller and SENER (Spanish Engineer- ing Company) rotary actuators.8These actuators use the large stroke of a redundant SMA wire or spring.

Another mechanism involves the use of a SMA directly as an actuator to break a prenotched bolt in tension.9 The notched bolt stretches until it fails at the notch, providing con- trolled breakage. However, this system creates particles that could endanger the operation of delicate equipment. More- over, a considerable amount of time is required to activate the SMA by heaters due to the SMA’s large volume.

In this study, to develop a non-explosive release actu- ator, which has a fast release time, a simple mechanism, a large preload, and low levels of shock, we propose a new model that combines a separation nut and a SMA wire trigger.

The separation nut can sustain a high preload because nuts have many threads, which can distribute applied preload effi- ciently. And the SMA wire can simplify the complex trigger

0034-6748/2013/84(1)/015005/8/$30.00 84, 015005-1 © 2013 American Institute of Physics

(3)

015005-2 Yooet al. Rev. Sci. Instrum. 84, 015005 (2013)

TABLE I. Design requirement of the non-explosive release actuator.4

Performance Requirement

Preload >10 000 N

Release time <500 ms

Separation shock <1000 G

Mass <500 g

Other Resettability

Low power consumption (Pyropulse <= 5.0 A)

mechanisms such as the motor and electrical spool, and also it can be reused repeatedly without replacement.

Resettability is also considered for the accomplishment of several repetitive separation tests of the appendages. The design requirements are listed in TableI.4After its design and manufacturing, performance and environment tests are carried out to evaluate the proposed release actuator.

II. FUNCTIONAL PRINCIPLE A. Component

The proposed mechanism of a non-explosive release ac- tuator is based on a trigger of shape memory alloy wire and the separation of segmented nuts. The configuration of the proposed device is shown in Figures1(a)and1(b). The device has a cylindrical structure with dimensions of Ø35× 72 mm (diameter × height) and a weight of 0.275 kg. It consists of a housing structure, two SMA wires (Dynalloy, Inc., USA), a safety spring, a trigger block, a bushing, a lock- ing sleeve, a compression spring, a bolt, and three segmented nuts. The housing, trigger block and locking sleeve are made of titanium. As shown in Figure 1(a), the ball restricts a locking sleeve that is under an axial load induced by the compression spring, and the radial movement of segmented nuts is restricted by rollers that are attached to the locking sleeve.

B. Release mechanism

The trigger operation for the bolt release is performed by the contraction of the SMA wire. The SMA wire shrinks to its original shape by heating it to more than 100C (the austenite finish temperature) using a power source. This activation is fully redundant, as it is provided with two SMA wires electri- cally connected to different circuits. Each wire has their tips, which are fixed to the insulated contacts at the housing lid.

The wire is disposed in a loop over the trigger block and is preloaded to obtain the wire actuation stroke. This preload is given by the reset bolt that puts back the release device into initial state. The original length and diameter of the SMA wires are 120 and 0.38 mm, respectively. The wire can pro- vide up to 110N with a stroke of about 4 mm. It is enough to overcome the safety spring.

Once the SMA wires contract, the trigger block moves in the upward direction, from the position shown in Figure1, after which the balls move radially in a grooved slot in the trigger block due to the annular surface formed on the inside of the locking sleeve loaded by the compression spring. The locking sleeve and rollers then move upwardly so that the seg- mented nuts are separated into three pieces and the bolt is re- leased. Figure1(b)shows the released position of the device.

A safety spring is inserted to prevent unintended trigger activation before the release operation and to minimize the shock from the impact generated by the locking sleeve and compression spring after the release operation.

The proposed non-explosive actuator was shown to be fully reusable after all tests. It provides a very easy resetting mechanism without a change of any of its components. Thus, there is no need to use disposable components. Figure2shows the resetting configuration. After the released actuator is fixed inside the compression fixture, a reset bolt is inserted in the cylindrical groove of the trigger block through the top of the housing. When the trigger block moves downwardly toward its untriggered position through the rotation of the reset bolt, the contracted SMA wire can be strained again. At this time, the balls are loaded outwardly by the beveled grooves on the outside of the trigger block, going back to their unreleased

FIG. 1. Configuration of release device.

(4)

FIG. 2. Experimental setup for the resetting of the release actuator.

FIG. 3. Experimental setup for the preload test.

FIG. 4. Preload test up to 15 kN.

position. The rollers attached to the locking sleeve then push the segmented nuts inwardly toward its unreleased position.

Finally, the release actuator returns to its initial state as the nuts are closed and the bolt is preloaded.

III. PERFORMANCE TEST

To check the performance of the proposed non-explosive release actuator, several performance checks are done. These were a static load (preload) test, a separation time test, and a shock level test.

A. Static load test

Based on the launching experience equation, the static load from the relationship between the acceleration and the mass can be calculated.10

a= 50/m0.3[G], (1)

where, m is mass in kg.

From Eq. (1), if we assume that the mass of the so- lar panel is 20 kg, 20.4G of static acceleration can be ob- tained. Consequently, the panel will be under a static load of 3.994 kN. Besides the mass of the solar panel, parameters such as those pertaining to safety factors and binding points should be considered.11 To check the structural integrity of the release actuator using a high preload, a 0.25-inch heat- treated steel bolt was prepared. Its yield strength is 20 kN.

The experimental system is shown in Figure3. The static load test was carried out using a materials test system (MTS 810).

FIG. 5. Release time test. Video sequence.

(5)

015005-4 Yooet al. Rev. Sci. Instrum. 84, 015005 (2013)

FIG. 6. Configuration of the preload generator.

Figure4 shows the results of the preload test. A maximum tensile force of 15 kN is applied to the proposed model for about 10 min. No loss of preload was detected during the test, and the threads of the bolt and the nuts were not damaged after the static load test. This indicates that the proposed re- lease system can sustain the solar arrays of a medium-sized satellite.

B. Separation time test

The separation time is monitored by a high-speed video camera (FASTCAM SA4, Photron, Inc.), as shown in Figure5, and it is measured through an investigation of the video clips. To apply a preload to the release actuator, a preload generator was designed, as shown in Figure6. It con- sists of a compression spring, a bolt, an aluminum plate, and a fixation nut. The preload is applied to the release actuator via the bolt, and reactive force acts on the aluminum plate.

Before the separation test, a tensile force of 10 kN was ap- plied to the release actuator, and the units sustained the load without any damage. The separation time from the activation of the power switch to the release of the bolt is 50 ms on av- erage, and the power is only 10.4 W (2 ampere, 5.2 V). In order to avoid current leakage to release actuator housing or

FIG. 7. Experimental setup for the shock level test.

FIG. 8. Shock response spectrum (SRS) results of the shock level test.

FIG. 9. Experimental setup for the thermal vacuum test.

(6)

FIG. 10. Evolution of the thermal vacuum test.

other parts, a passivation process such as anodization of the structural components is needed.

C. Shock level test

The shock output was measured to characterize the typ- ical shock impact of the release actuator onto the satel- lite structure. The test components are mounted on a 1×1 m Al honeycomb test panel with 1 mm Al sheets on both sides. It is vertically suspended with elastic ropes to simu- late zero gravity. The accelerometers (Bruel & Kjaer 4371) are installed at 10 (the source point), 100, 200, and 300 mm from the release actuator position, as shown in Figure7. The preloads applied to the release actuator are 5, 10, and 15 kN, respectively.

Figure8shows the result of the shock level test. The max- imum shock level of about 350 G at 6350 Hz was observed at a shock source with a preload of less than 15 kN. Compared to other developed release actuators, especially pyrotechnic de- vices, the shock level of the proposed release actuator is very low. Pyrotechnics typically cause accelerations of more than

FIG. 11. Release operation test at extreme temperatures.

FIG. 12. Experimental setup for the random vibration test.

1000 G at 1000 Hz and up to 10 000 G between 5000 Hz and 20 000 Hz.12,13Thus, the proposed actuator reduces the shock by at least one order of magnitude.

IV. ENVIRONMENTAL TEST

Environmental tests are used to verify whether a piece of equipment can withstand the rigors of harsh environments.

The proposed release actuator is tested under extremely high and low temperatures and vibration environments.

A. Thermal vacuum cycling test

The mechanism was tested under thermal vacuum condi- tions with P < 10−5 mbar to verify the non-operation of the proposed actuator between−90C and 90C. The unit is in- stalled in the chamber as shown in Figure9. The preload is 10 kN. The preload generator is covered with a safety cover to prevent accidents caused by malfunctioning.

Figure 10 shows the evolution of the thermal cycling test. During the test, the actuator was not released and no problems were generated. After nine cycles in the thermal vacuum, a post-release performance test was performed, as shown in Figure11. The actuator was successfully released

TABLE II. Input loads used for the random vibration test.

Frequency (Hz) Acceptance PSD (G2/Hz)b

X, Y, Za 20 0.026

20∼ 50 +6dB/Octave

50 0.16

800 0.16

800∼ 2000 −6dB/Octave

2000 0.026

Grms 14.1

aX,Y: out of plane, Z: in-plane.

bAcceptance: 1 min per axis as noted and qualification: 3 min per axis+3dB form values in the table.

(7)

015005-6 Yooet al. Rev. Sci. Instrum. 84, 015005 (2013)

FIG. 13. Comparison of all pre- and post-frequencies.

without anomalies at ambient temperatures of −90C and 90C.

B. Random vibration test

The actuator is subjected to a random vibration test. The acceleration is measured by four three-axis accelerometers at- tached to a jig. The actuator is also preloaded to 10 kN using a preload generator. Figure12shows the setup for the exci- tation test. The environmental conditions are summarized in TableII.

Figures 13 and 14 depict a comparison of the pre- and post-vibration test results and the power spectrum den- sity (PSD)-frequency curves, respectively. As shown in Figure13, these results reveal little shift in the frequency or

FIG. 14. Results of the random vibration test.

amplitude in all directions of excitation. This indicates that the release actuator is not deformed and that it survived with- out any degradation and without experiencing an unintended release after the random vibration test. The results of the out- of-plane (X, Y) vibration test are similar to each other due to the axisymmetric structure (Figure14). However, the re- sult of the in-plane (Z) test shows a difference compared to the other results because the components of the springs, trig- ger block, and locking sleeve are positioned in the Z-axis direction. The detailed results are summarized in TableIII.

Frequency shifts of all directions are less than 1%; hence, it can be said that this system would be stable in the launching environment.

After the random vibration test, a release test under ambi- ent conditions was performed; the actuator was released dur- ing this test without any problems.

(8)

TABLE III. Results of random vibration test.

Amplification factor @ natural frequency

Sequence Overall Grms (control) Test duration (min) 1st 2nd 3rd Overall Grms (response)

Pre 0.943 Grms 1 6.339 10.40 17.26 2.908 Grms

@801.2 Hz @916.2 Hz @1149 Hz

X Main 20.03 Grms 3 7.534 13.84 65.12 Grms

@908.7 Hz @1176 Hz

Post 0.968 Grms 1 6.561 9.683 15.70 2.950 Grms

@805 Hz @921 Hz @1148 Hz

Frequency shift (%) 0.474 0.415 0.087

Pre 0.965 Grms 1 12.16 37.08 4.211 Grms

@915 Hz @1180 Hz

Y Main 20.05 Grms 3 10.94 12.62 66.21 Grms

@897.5 Hz @1166 Hz

Post 0.938 Grms 1 11.04 44.67 4.604 Grms

@916.2 Hz @1185 Hz

Frequency shift (%) 0.131 0.424

Pre 0.985 Grms 1 1.648 6.223 63.97 2.966 Grms

@435 Hz @781.2 Hz @1548 Hz

Z Main 20.13 Grms 3 1.581 1.914 15.56 49.50 Grms

@427.5 Hz @766.2 Hz @1546 Hz

Post 0.974 Grms 1 1.648 6.081 62.80 3.079 Grms

@435 Hz @783.8 Hz @1550 Hz

Frequency shift (%) 0.0 0.333 0.129

V. CONCLUSION

A non-explosive release actuator is designed, manufac- tured, and tested. To meet the design requirements pertaining to the separation time, shock level and preload, segmented nuts and a trigger combined with a SMA wire that provides a very simple operating mechanism are chosen. In addition, the proposed actuator is equipped with a redundant system to increase the reliability of the release operation.

Through a performance test, we confirmed that the re- lease actuator can be released within 50 ms and that it can sustain a preload of 15 kN due to the SMA wires and the seg- mented nuts, respectively. An inserted safety spring prevents unintended releases and reduces the shock induced by the re- lease operation. The shock output is considerably less than the shock of typical release devices. The comparison of com- mercial release actuators is summarized in TableIV.4,5 The proposed release actuator satisfies fast releasing time, high preload capacity, and low shock level at the same time. In ad- dition, the advantage of this device is that the units intended to

be flown are the same one, which successfully passed accep- tance testing. It is possible to have rapid and repeatable so- lar array release and deployment test without disassembling of spacecraft hardware since the refurbishment or replace- ment of unit is not required in the process of reassemble of release device. Namely, this release actuator can be applied readily in spacecraft intended to be flown. For this reason, the cost savings can be also obtained in integration and test schedules.

The proposed release actuator was tested under a space environment, which had a high vacuum level and extreme temperatures. It was also tested in an extreme vibration en- vironment, which simulated an launch condition. No degra- dation or damage was observed during the two environmental tests, and the release actuator operated successfully.

Based on the results obtained from the tests, the proposed device is a good and economic substitute for other release de- vices due to its high level of performance and quite simple mechanism.

TABLE IV. Comparison of commercial release actuators.4,5

Paraffin Thermal Pin-puller Frangibolt Electrical Suggested

Pyro-technic actuator knife (SMA wire) (SMA tube) spool actuator

Release time (ms) 30 180 000 60 000 30 30 000 25 50

Preload (kN) ∼40 ∼3 ∼4 ∼0.1 ∼2 ∼20 ∼15

Shock (G) 7000 250 350 500 2000 3000 350

Replacement or refurbishment Y N Y Y Y Y N

(9)

015005-8 Yooet al. Rev. Sci. Instrum. 84, 015005 (2013)

ACKNOWLEDGMENTS

This work was supported from the Korea Aerospace Re- search Institute (KARI). Additional support was obtained from the National Research Foundation of Korea (NRF) grant funded by the Korean Government (MEST) (Grant No. 2012- 0000976).

1E. R. Fosness, S. J. Steven, and W. F. Gammill, AIAA Paper No. 2001–

4601, 2001.

2P. A. Jones and B. R. Spence,IEEE Aerosp. Electron. Syst. Mag.26(8), 17 (2011).

3A. C. Peffer, E. R. Fosness, B. F. Carpenter, and K. K. Denoyer,Proc. SPIE 3991, 187 (2000).

4M. H. Lucy, R. C. Hardy, E. H. Kist, J. J. Watson, and S. A. Wise, in Proceedings of the 10th Annual AIAA/USU Conference on Small Satellites, Logan, UT, 1996 (Utah State University, Logan, UT, 1996), p. 1.

5J. H. Lim, K. W. Kim, S. W. Kim, D. S. Hwang, Y. I. Yoo, and J. J. Lee, in Proceedings of the 61th International Astronautical Congress, Prague, Czech Republic, 2010 (IAF, Paris, France, 2010), p. 1.

6M. S. Lee, J. U. Jo, W. J. Tak, and B. K. Kim,Int. J. Precis. Eng. Manuf.

12, 569 (2011).

7Y. Bellouard,Mater. Sci. Eng., A481–482, 582 (2008).

8J. Vazquez and J. I. Bueno, in Proceedings of the 9th European Space Mechanisms and Tribology Symposium, Liege, Belgium, 2001, edited by R. A. Harris (ESA, Noordwijk, Netherlands, 2001), p. 131.

9J. D. Busch and M. D. Bokaie, in Proceedings of the 28th Aerospace Mech- anisms Symposium, Cleveland, OH, 1994 (NASA Lewis Research Center, Cleveland, OH, 1994), p. 379.

10T. S. Jang, H. B. Kim, S. H. Woo, S. H. Lee, and M. R. Nam, in Proceedings of the Korean Society for Noise and Vibration Engineering Conference, Daegu, South Korea, 2005 (KSNVE, Seoul, South Korea, 2005), p. 957.

11W. J. Tak, M. S. Lee, and B. K. Kim,J. Mech. Sci. Technol.25, 1141 (2011).

12D. Wattiaux, O. Verlinden, C. Conti, and C. D. Fruytier,J. Vib. Acoust.

130, 041012 (2008).

13A. Peffer, K. Denoyer, E. Fosness, and D. Sciull, in Proceedings of the 14th Annual/USU Conference on Small Satellites, Logan, UT, 2000 (Utah State University, Logan, UT, 2000), p. SSC00-XI-5.

참조

관련 문서

In this study, we developed a non-toxic titanium alloy with low elastic modulus in order to improve biocompatibility, and the surface characteristics of

Wang et al., Smart Soft Composite Actuator with Shape Retention Capability using Embedded Fusible Alloy Structures, Composites Part B: Engineering (2015).. Shape

Introduction to shape memory effect and shape memory alloy Introduction to electro-active polymer4. Application examples of smart

The scheduling equations are mutually dependent: the release jitter of a receiver task (equation 31) depends on the arrival time of the message a m (equation

Self-regulating shape memory alloy robots controlled with soft ionic skin

• The PDEase2D evolution solver automatically and continuously adjusts its time step interval to reduce the time interval when PDEase2D encounters a region of rapid change

[r]

 The simplest method of storing a raster layer in the memory of the computer is using a data structure called an array..  We consider alternative methods for searching through