aerospace

Article
High Damping Passive Launch Vibration Isolation System
Using Superelastic SMA with Multilayered Viscous Lamina
Yeon-Hyeok Park 1 , Seong-Cheol Kwon 2 , Kyung-Rae Koo 2 and Hyun-Ung Oh 1, *

1 Space Technology Synthesis Laboratory, Department of Smart Vehicle System Engineering,

Chosun University, 375 Seosuk-dong, Dong-gu, Gwangju 501-759, Korea; wkfjf6043@chosun.kr
2 Mechatronics Groups, Hanwha Systems, 491-23, Gyeonggidong-ro, Namsa-myeon, Cheoin-gu,
Yongin-si 17121, Korea; seongcheol.kwon@hanwha.com (S.-C.K.); kr.koo@hanwha.com (K.-R.K.)
* Correspondence: ohu129@chosun.ac.kr

Abstract: Whole-spacecraft launch-vibration isolation systems are attractive for achieving the goal
of better, faster, cheaper, and lighter small satellites by reducing the design-load and vibration-test
specifications for on-board components. In this study, a three-axis passive launch-vibration isolation
system, based on superelastic shape memory alloy (SMA) technology, was developed to significantly
attenuate the dynamic launch loads transmitted to a small satellite. This provides a superior damping
characteristic, achieved by superelastic SMA blades stiffened by multilayered thin plates with viscous
lamina adhesive layers of acrylic tape. The basic characteristics of the proposed isolation system with
various numbers of viscoelastic multilayers were obtained through a static load test. In addition, the
effectiveness of the design was validated through a launch environment simulating sine and random
vibration tests.





 Keywords: superelastic shape memory alloy; whole-spacecraft vibration isolation; small satellite;
Citation: Park, Y.-H.; Kwon, S.-C.; launch environment
Koo, K.-R.; Oh, H.-U. High Damping
Passive Launch Vibration Isolation
System Using Superelastic SMA with
Multilayered Viscous Lamina. 1. Introduction
Aerospace 2021, 8, 201. https:// The start of the New Space paradigm has changed the development philosophy of the
doi.org/10.3390/aerospace8080201 worldwide space-engineering field. New Space refers to the recent commercialization of
the space sector, which is mainly led by private industries, rather than government-funded
Academic Editor: Rosario Pecora
organizations. These industries are driving a better, faster, cheaper, and lighter space-
development paradigm [1–4]. One of the major accelerators of the New Space paradigm
Received: 15 June 2021
is the emergence of small satellite constellations. The large-volume production of small
Accepted: 24 July 2021
satellites, based on a standardized satellite platform and commercially available on-board
Published: 26 July 2021
hardware, has reduced the development cost and schedule. In addition, the total launch
cost can be reduced as multiple small satellites are launched together. These advantages
Publisher’s Note: MDPI stays neutral
make the small-satellite platform attractive for various challenging missions that require
with regard to jurisdictional claims in
published maps and institutional affil-
high temporal system performance, e.g., real-time remote sensing, global internet services,
iations.
and high-speed communications [4].
All satellites undergo various dynamic loads during the launch phase. These loads in-
volve steady-state acceleration caused by the engine thrust, sinusoidal vibration caused by
the engine cutoff, and a self-excited vibration, called the pogo effect, from the combustion
instability of the launcher. Random vibrations are caused by the noise of the thrust, and me-
Copyright: © 2021 by the authors.
chanical shock is induced by the separation of the launcher stages and spacecraft. Because
Licensee MDPI, Basel, Switzerland.
these dynamic loads are extremely severe and complex, they are one of the major factors
This article is an open access article
distributed under the terms and
that cause satellites and components to malfunction [5]. Therefore, a proper structural
conditions of the Creative Commons
design is essential to ensure the structural safety of the satellite in the launch environment.
Attribution (CC BY) license (https://
Two approaches are used to enhance the structural safety of satellites during the
creativecommons.org/licenses/by/ design phase. One traditional approach is to design the spacecraft structure such that it
4.0/). has sufficient strength and stiffness to endure mechanical loading. However, this approach

Aerospace 2021, 8, 201. https://doi.org/10.3390/aerospace8080201 https://www.mdpi.com/journal/aerospace

Aerospace 2021, 8, 201 2 of 14

faces the technical issue on limitation in minimizing the mass and volume of the satellite,
especially in small satellites with design limitations of their volume and mass. These
factors are related to the available number of satellites that can be launched within the lift
capability of the launcher and, therefore, directly lead to an increase in launch costs. The
other method involves reducing the launch loads transmitted to the satellite by applying
a whole-spacecraft vibration isolator (WSVI). The WSVI is achieved by implementing a
low-stiffness and high-damping capability, compared to the conventional concept, which
is rigidly mounted between the satellite and launch adaptor. This makes it possible to
effectively reduce the mass and volume of the satellite by minimizing the design load of
vibration-sensitive components. In addition, it can contribute to reducing the satellite’s
development cost and schedule by optimizing the conventional verification process, which
accounts for a large portion of the time and cost of satellite-development programs. In
addition, the on-ground test process can be optimized, e.g., simplifying and skipping the
test phases of the subsystem-level verification. For example, verification can be performed
at a higher system level of the integrated test because using the WSVI greatly reduces
the potential technical risks to the vibration-sensitive components at the subsystem level.
Another possible test schedule mitigation might also be expected on the notching process
in the launch-vibration tests because creating a proper notched input level takes a large
portion of the time. These potential advantages of WSVIs are attractive for the New Space
based development trend to achieve the goal of cost-effective small satellite development.
Several previous studies have been conducted to realize whole-spacecraft vibration
isolators. Johnson et al. [6] proposed WSVI systems called SoftRide UniFlex and SoftRide
MultiFlex to realize launch-vibration isolation for the MINOTAUR/JAWSAT program.
SoftRide UniFlex consisted of a titanium flexure and a viscoelastic material to reduce the
dynamic loads acting on the satellite along the launcher axial direction. To isolate both axial
and lateral excitations from the launcher, SoftRide MultiFlex consisted of a pair of UniFlex
isolators integrated with each other through a central post. The vibration in the launcher
axial direction was isolated in the same manner as the UniFlex. Isolation in the lateral
directions was achieved by vibrational-energy dissipation from the shear deformation of
the constrained layers, as the dynamic bending of the flexure occurred. To enhance the
performance of SoftRide MultiFlex, SoftRide OmniFlex was developed by Johal et al. [7].
The major difference from the previous version was that the constrained layer was installed
at the sides of the titanium flexure. This enabled the isolator to be more compact in volume,
as well as enhancing the damping capability. Jun et al. [8] proposed an isolation system
using two serially connected flexure elements and a shear damping unit laminated with
a metal plate and viscoelastic material. Isolation in the axial and lateral directions was
achieved by the bending behavior of the titanium flexure and the shearing behavior of
the damping unit, respectively. Mastroddi et al. [9] proposed a multi-frequency dynamic
absorber using a two-spring mechanism arranged in each axial and lateral direction. These
mechanisms are connected with the oscillating mass for tuning the specific target frequency
for the spacecraft during the launch phase. Rittweger et al. [10] proposed an active payload
adaptor compliant with the Ariane 5 space-launch vehicle, which was able to reduce the
interface loads to the payload in the 5–100 Hz low frequency domain by more than a factor
of four.
In actualizing a novel high damping three-axis passive launch vibration isolation
for small satellite, we focused on the following two technical aspects. One is using the
superelasticity of the shape memory alloy (SMA) material, and the other is applying
multilayered thin plates with viscous lamina tapes to the SMA blades. The superelasticity
is a unique characteristic of SMA material occurred by stress-induced phase transformation
from the austenite to the martensite phase when the material is at above austenite finish
temperature. It can be deformed considerably without being plastically deformed and
recovers its original shape upon unloading. This makes it possible to ensure the structural
safety of a satellite under the launch environment, even if the satellite is supported by
Aerospace 2021, 8, 201 3 of 14

a low-stiffness SMA application. In addition, this characteristic is associated with large

hysteretic damping, owing to the phase transformation [11].
The effectiveness of using superelastic SMA for vibration isolation in space applications
has been demonstrated in several previous studies [12,13]. For example, Kwon et al. [12]
proposed a superelastic SMA gear wheel that applied a two-axis gimbal-type X-band
antenna to enhance the micro-jitter isolation capability without undergoing plastic defor-
mation under excessive loading conditions. Kwon et al. [13] also proposed a blade-type
cooler vibration isolator using a superelastic SMA material as a technical solution over
ordinary titanium material. Then, the effectiveness of the superelastic SMA blade design
was experimentally assessed and compared with that of an isolator made of titanium.
If the superelastic SMA is applied to the WSVI, it is possible to ensure the structural
safety of the satellite in a launch environment, while implementing a low-stiffness isolator.
However, the SMA exhibits effective hysteretic damping only when the stress exceeds the
critical point at which a phase transformation occurs. This means that an isolator with
SMA material may exhibit insufficient damping performance, if the deformation is not
sufficiently large. Therefore, in this study, we proposed to apply multilayered thin plates
with viscous lamina tapes to maximize the damping capability, even under a relatively
small deformation.
The effectiveness of viscoelastic multilayered thin plates in enhancing the damping
capability has been investigated in many studies [14–17]. Minesugi et al. [14] proposed
damping mechanisms of polyimide tape with a viscous lamina to reduce the vibration
transmitted to the battery panel of the MUSES-A satellite. The experimental results showed
that a five times higher damping performance was obtained when viscous lamina was
applied, as compared to that without the lamina. Bhattarai et al. [15] proposed a highly
damped deployable solar-panel module using a multilayered stiffener with viscoelastic
acrylic tape. The design effectiveness of the solar-panel module was validated through a
launch-vibration test.
Park et al. [16] proposed a high damping printed circuit board (PCB) with multi-
layered viscoelastic acrylic tapes for use in wedge lock applications. This PCB concept
was effective in increasing the fatigue life of electronic packages, owing to the highly
increased damping capability, as well as minimizing the volume and mass of the elec-
tronics. Stoudt et al. [17] hypothesized that using nanoscale multilayer coatings could
significantly increase the fatigue durability. To confirm this hypothesis, cyclic-loading
fatigue tests were performed with Cu-coated films and different surface treatments, in-
cluding a nanoscale Cu-Ni multilayer. The results of the fatigue tests showed that the
Cu-Ni multilayer film had a fatigue life more than six times greater than that of the other
general films under cyclic-loading conditions because the slip between each layer acted as
a stress-energy–dissipation mechanism.
In this study, the basic characteristics of the passive WSVI with various number of
interlaminated layers on the superplastic SMA blade, designed for vibration isolation
of 40 kg class satellite, were obtained through static load tests. In addition, to validate
the effectiveness of the design in terms of the launch-load attenuation, sine and random
vibration tests were performed using a mass-simulating dummy satellite. These test results
demonstrated that the proposed WSVI is effective for achieving a novel design goal of both
superelastic and high damping capability.

2. WSVI Design Description

Figure 1a,b show isometric and internal views of the proposed WSVI. This WSVI was
developed to reduce the launch loads above the range of the 28 Hz target cut-off frequency
for a 40 kg class satellite. The design strategy of the isolator involves implementing a
high damping capability by applying a superelastic SMA blade with a multilayered thin
plate with viscous lamina tapes. The SMA material can be applied because it provides
superelastic characteristics, high damping capability, and fatigue durability, compared
to ordinary metal materials [12]. Superelastic behavior is known to have complete re-
Aerospace 2021, 8, x FOR PEER REVIEW 4 of 14

with viscous lamina tapes. The SMA material can be applied because it provides supere-
Aerospace 2021, 8, 201 4 of 14
lastic characteristics, high damping capability, and fatigue durability, compared to ordi-
nary metal materials [12]. Superelastic behavior is known to have complete reversibility
for strains of up to 10–12% without being plastically deformed, which is a very uncommon
feature in ordinary
versibility metal
for strains of upmaterials
to 10–12%[11]. Therefore,
without being the SMA blade
plastically makeswhich
deformed, it possible to
is a very
ensure the structural
uncommon feature insafety of themetal
ordinary low stiffness
materialsblade
[11]. of the isolator
Therefore, theunder the launch
SMA blade makes en-it
vironment.
possible to ensure the structural safety of the low stiffness blade of the isolator under the
launch environment.

(a)

(b)
Configurationof
Figure1.1.Configuration
Figure ofthe
theProposed
ProposedWSVI
WSVI((a)
((a)Isometric
IsometricView,
View, (b)
(b) Inside
Inside View).
View).

In addition,
In addition, multilayered
multilayered thin
thin metal
metal plates
plateswith
withviscous
viscouslamina
laminatapes
tapeswere
wereutilized to
utilized
increase the damping capability, which also helped distribute the stress release
to increase the damping capability, which also helped distribute the stress release actingacting on
thethe
on thin SMA
thin SMAblade. The
blade. proposed
The WSVI
proposed WSVI is mainly composed
is mainly composed of of
SMASMA blade modules
blade modules to
achieve the required target cut-off frequency in the axial and lateral directions, a moving
plate, upper and bottom plates, inner and outer brackets, and displacement limiters to
limit the movement of the blades to within their allowable deflection range. The stiffness
of the vibration isolator for each axis is significantly governed by the SMA blade modules.
Aerospace 2021, 8, 201 5 of 14

To achieve the required isolator stiffness, the SMA blade modules for the axial direction
are connected between the inner and outer brackets at 120◦ intervals. The SMA blade
modules for the lateral direction are arranged between the moving plate and inner bracket,
in the same manner as those of the axial direction. The blade is made of superplastic
SMA, and it provides a mechanical interface to attach viscoelastic adhesive tapes with thin
constrained layers made of FR-4 material. Two constrained layers were applied to each
side of the SMA blade. The physical contact between the constrained layers and adhesive
tapes reduces the resultant stress acting on the SMA blade because the viscoelasticity
resists additional deformations of the SMA blade. Furthermore, each boundary layer
between the adhesive tapes and constrained layers experiences shear deformation when
the multilayered SMA blade is deformed. This design contributes to the excellent damping
performance and enhanced structural safety of the SMA blades. To limit the deformation
of the blades within the allowable range of ±5 mm, displacement limiters made of high
damping plastic material of Delrin [18], with space heritages are included on the design,
which is helpful to dissipate and mitigate launch vibration loads when slip and contact
occur between the plastic and plastic materials. The mechanical properties of the SMA
materials are summarized in Table 1 [12]. The viscoelastic tape used in this study was
3M966 double-sided acrylic tape (3M) [19], as listed in Table 2.

Table 1. Material Properties of the Superelastic SMA.

Characteristic Value
Martensite Finish Temperature (M f , ◦ C) −21
Martensite Start Temperature (Ms , ◦ C) −12
Austenite Start Temperature (As , ◦ C) −5
Austenite Finish Temperature (A f , ◦ C) 15
Martensite 75
Young’s Modulus (GPa)
Austenite 80
Tensile Strength (MPa) 1300
Elongation at Break (%) 45
Density (g/cm3 ) 6.45
Poisson’s Ratio (ρ) 0.33

Table 2. Specifications of Viscoelastic Adhesive Tape (3M966).

Item Specification
Type Double-sided Acrylic Tape
Thickness (mm) 0.06
58 (20-min Dwell)
Adhesion Strength to Steel (N/100 mm) 85 (72-h Dwell)
159 (Ultimate Bond)
Outgassing (%, TML/CVCM) 0.93/0.01

A static load test was performed at room temperature to evaluate the basic charac-
teristics of the proposed WSVI. Two WSVI cases with different SMA blade thicknesses
(1) 0.8 mm and (2) 1.5 mm were used in this test. Figure 2 shows an example of a static
load test setup for the lateral axis. The WSVI was connected to the load cell using a grip
and a connector, which was mounted on the upper side of the static testing machine. In
addition, the Delrin parts of the WSVI were intentionally removed to check the internal
status of the blades during the test. Repeated translational loadings of three cycles in the
positive and negative directions were applied to the WSVI in each axis to measure its
 Aerospace 2021, 8, x FOR PEER REVIEW 6 of 14

test setup for the lateral axis. The WSVI was connected to the load cell using a grip and a
Aerospace 2021, 8, 201 connector, which was mounted on the upper side of the static testing machine. In addition,6 of 14
the Delrin parts of the WSVI were intentionally removed to check the internal status of
the blades during the test. Repeated translational loadings of three cycles in the positive
and negative directions were applied to the WSVI in each axis to measure its basic char-
basic characteristics.
acteristics. The translational
The translational displacements
displacements acting onacting on the
the SMA SMA
blade blade modules
modules in Cases in
(1)
Cases (1) and (2) are ± 4 mm and ± 3 mm, respectively, with a velocity of 2 mm/min.
and (2) are ±4 mm and ±3 mm, respectively, with a velocity of 2 mm/min. The correspond- The
corresponding
ing displacement displacement
ranges areranges are the estimated
the estimated maximummaximum displacements
displacements derived
derived from the
from the launch vibration
launch vibration analysis. analysis.

Figure2.2.Example
Figure Exampleof
ofthe
theStatic
StaticLoad
LoadTest
TestSetup
Setupfor
forthe
theLateral
LateralAxis.
Axis.

Figure3a,b
Figure 3a,bshow
showthe theload
loaddisplacement
displacementrelations
relationsobtained
obtainedfrom
fromthethestatic
staticload
loadtest
testof
of
theWSVI
the WSVIin inthe
thelateral
lateralandandaxial
axialdirections
directionsfor forCases
Cases(1)
(1)and
and(2).
(2).These
Thesetesttestresults
resultsshow
show
that
thatthe
theequivalent
equivalentstiffness
stiffnessof Case (2), (2),
of Case withwith
a blade thickness
a blade of 1.5 of
thickness mm, 1.5ismm,
1.7 times
is 1.7higher
times
than thatthan
higher of Case
that (1), with (1),
of Case a 0.8with
mm ablade.
0.8 mm In the test,In
blade. the boundary
the test, the layers
boundaryof thelayers
thin plates
of the
with
thin viscous lamina
plates with tapeslamina
viscous did nottapes
delaminate,
did not and no plasticand
delaminate, deformation
no plastic was observedwas
deformation on
the SMA blade
observed on thewithin
SMA the bladetested range
within the of translational
tested range of loading. Because
translational all subsequent
loading. Because all
curves completely
subsequent curvescoincided
completely with the initial
coincided withhysteresis
the initialcurve, evencurve,
hysteresis though the though
even structural
the
analysis results from the blades made of titanium and aluminum
structural analysis results from the blades made of titanium and aluminum showedshowed a negative margin
a neg-
of safety.
ative Furthermore,
margin of safety.the results showed
Furthermore, the aresults
much showed
larger hysteresis
a much area,
largerwhich cannot
hysteresis be
area,
achieved by general
which cannot metal materials,
be achieved by generalincluding superplastic
metal materials, SMA [12].
including This is because
superplastic the
SMA [12].
slip
Thisand frictionthe
is because induced
slip and byfriction
strong induced
molecular byattraction forces between
strong molecular attraction multi laminated
forces between
plates with the viscous lamina tape helped enhance the damping capability of the WSVI,
because of the larger shear deformation and strain on the multi- laminated SMA blades. In
addition, the WSVI in the axial direction showed a much higher damping characteristic
than that in the lateral direction because a 2.5 times larger area of the hysteresis curve
was obtained. This can be explained by the fact that the shear deformation of the SMA
blades arranged in the axial direction is larger than that in the lateral direction. A structural
 multi laminated plates with the viscous lamina tape helped enhance the damping capa-
bility of the WSVI, because of the larger shear deformation and strain on the multi- lami-
nated SMA blades. In addition, the WSVI in the axial direction showed a much higher
damping characteristic than that in the lateral direction because a 2.5 times larger area of
the hysteresis curve was obtained. This can be explained by the fact that the shear defor-
Aerospace 2021, 8, 201 7 of 14
mation of the SMA blades arranged in the axial direction is larger than that in the lateral
direction. A structural analysis of the SMA blade for the axial direction also showed a 2.2
times greater maximum strain than that of the blade for the lateral direction. The test re-
sults alsoofshowed
analysis the SMA that the for
blade hysteresis
the axialarea obtained
direction alsofrom
showedCasea (2)
2.2 was
timesalmost
greaterthe same or
maximum
slightly less that
strain than thanofthatthe of Casefor
blade (1).
theMoreover, a nonlinear
lateral direction. The characteristic
test results alsoinduced
showed by that
the
phase transformation
the hysteresis of the superplastic
area obtained from Case (2) SMA
wasblade
almostwasthenot
same observed within
or slightly lessthe
thantested
that
displacement range. Ona the
of Case (1). Moreover, contrary,
nonlinear the stiffness
characteristic of the WSVI
induced by the in the axial
phase direction in-
transformation of
creased as the displacement
the superplastic SMA blade was increased. This phenomenon
not observed seemsdisplacement
within the tested to be mainly range.
relatedOn to
the
the WSVI design
contrary, because of
the stiffness thethe
stiffness
WSVI in the the axial
longitudinal
directiondirection
increased ofas
the SMA
the blade be-
displacement
comes dominant
increased. in determining
This phenomenon the to
seems stiffness
be mainlyof the WSVItoasthe
related theWSVI
displacement increases.
design because the
Therefore,
stiffness in these results from
the longitudinal the static
direction load
of the testblade
SMA of thebecomes
WSVI indicated
dominantthat the design
in determining
the stiffness
strategy of the WSVI
of applying as the displacement
multi-layered viscous lamina increases. Therefore,
to the SMA bladesthese results
is a much morefrom the
dom-
static factor
inant load test of the WSVI
in enhancing theindicated
dampingthat theisolator
of the design than
strategy
the of applying
inherent multi-layered
damping perfor-
viscousoflamina
mance to theSMA.
superelastic SMA In blades is a much
addition, more dominant
the multilayered viscous factor
laminain enhancing
contributesthe to
damping of the isolator than the inherent damping performance of
reducing the stress acting on the thin SMA blade, owing to the viscoelasticity of the adhe-superelastic SMA. In
addition,
sive tapes the multilayered
[17]. Furthermore, viscous lamina
a previous contributes
study to reducing
[12] reported that athe stress acting
superelastic SMA on de-
the
thin SMA
veloped forblade, owing toSMA
high damping the viscoelasticity
gear showed higherof the adhesive tapes [17].
fatigue durability thanFurthermore,
titanium un-a
previous
der cyclicstudy [12] reported
loadings. Therefore,thatSMA a superelastic
blades maySMA developed
be useful for high damping
for launch-vibration SMA
isolator
gear showed higher fatigue durability than titanium under cyclic loadings. Therefore, SMA
applications.
blades may be useful for launch-vibration isolator applications.

1600
Case 1
1200 Case 2

800

400
Load (N)

-400

-800

-1200

-1600
-5 -2.5 0 2.5 5

Displacement (mm)
(a)

Figure 3. Cont.
Aerospace 2021,
Aerospace 8, x201
2021, 8, FOR PEER REVIEW 88 of
of 14
14

1600
Case 1
1200 Case 2

800

400
Load (N)

-400

-800

-1200

-1600
-5 -2.5 0 2.5 5

Displacement (mm)
(b)
StaticLoad
Figure3.3.Static
Figure LoadTest
Test Results
Results ((a)
((a) Lateral
Lateral Direction,
Direction, (b)
(b) Axial
Axial Direction).
Direction).

The estimated
The estimated values
values ofof the
the equivalent
equivalent damping
damping ζ𝜁eq for foreach
eachcase
caseare summarized
are summarized in
Table 3. The values of in each direction were obtained by the following
in Table 3. The values of 𝜁 in each direction were obtained by the following equivalent
ζ eq equivalent lin-
earization method,
linearization method,in in
which
whichthethe
nonlinear
nonlinear stiffness andand
stiffness damping
damping coefficients are are
coefficients translated
trans-
into linear ones [20].
lated into linear ones [20].
∆E(α0 )
ζ eq (α0 ) = (1)
∆𝐸(𝛼
2πa 0 k eq)
2
𝜁 (𝛼 ) = (1)
where, k eq is estimated from the linear-curve 2𝜋𝑎 𝑘 of the overall slope of the load-
fitting
displacement
where, curve. ∆E
𝑘 is estimated is the
from thearea of the closed
linear-curve fittingloop ofoverall
of the the hysteresis
slope of curve and a0
the load-dis-
is the amplitude of the displacement.
placement curve. ∆𝐸 is the area of the closed loop of the hysteresis curve and 𝑎 is the
amplitude of the displacement.
Table 3. Static Load Test Results of the Proposed WSVI.
Table 3. Static Load Test Results of the Proposed WSVI.
Equivalent Damping Ratio (ζ eq ) Equivalent Stiffness (keq , N/mm)
Equivalent Damping
Case Axial Ratio (𝛇𝒆𝒒Lateral
) Equivalent Stiffness (𝒌𝒆𝒒 , N/mm)
Axial Lateral
Case Axial (z-Axis) (x-
Lateral and y-Axes) (z-Axis)
Axial (x- and y-Axes)
Lateral
1 (z-Axis) 0.095(x- and y-Axes)
0.044 (z-Axis)219.5 (x- and 175.8
y-Axes)
1 2 0.095 0.087 0.044 0.032 219.5 386.6 175.8
325.4
2 0.087 0.032 386.6 325.4
3. Design Validation Test
3. Design Validation Test
To verify the effectiveness of the proposed WSVI design under a launch vibration
To verify i.e.,
environment, the performance
effectiveness of
oflaunch
the proposed WSVI design
load reduction under asafety
and structure launch vibration
of the WSVI,
environment, i.e., performance
sine and random of launch
vibration tests load reduction
were performed and
at the structure safety
qualification level.ofLow
the WSVI,
Level
sine and random vibration tests were performed at the qualification level. Low Level Sine
Aerospace 2021, 8, 201 9 of 14
Aerospace 2021, 8, x FOR PEER REVIEW 9 of 14

Sine Sweep (LLSS) tests were performed before and after the vibration test to confirm the
Sweep (LLSS)variations
characteristic tests wereof performed
the WSVI.before and after the vibration test to confirm the char-
acteristic
Figurevariations
4 shows of anthe WSVI.of the launch vibration test setup for the WSVI condition
example
on the Figure 4 shows
z-axis. In thean example
test setup, oftothe launch
achieve a vibration test setup
28 Hz cut-off for thefor
frequency WSVI
a 40condition
kg class
on the z-axis.
satellite, four In the test
WSVIs setup,
with the to achieve
Case a 28 Hz
(1) design cut-off
were frequency
integrated withforaamass
40 kgsimulating
class satel-
lite, foursatellite.
dummy WSVIs with The the Case (1)
dummy design
satellite were
was integrated
configured aswith
a flata plate-type
mass simulating dummy
to implement
satellite. The dummy satellite was configured as a flat plate-type
the design concept of a small synthetic aperture radar (SAR) technology experimental to implement the design
concept(S-STEP)
project of a small synthetic
satellite [21]. aperture
The S-STEP radar (SAR) istechnology
satellite experimental
an 80 kg class project (S-
small SAR-satellite
STEP)
that satellitea [21].
provides high The S-STEP1satellite
resolution m stripmapis animage.
80 kg class
It wassmall SAR-satellite
designed that
with a flat provides
plate-type
a high resolution
structure 1 m stripmap
for mechanical designimage. It was
simplicity anddesigned with a flat stability
high dimensional plate-typein structure for
orbit. In the
mechanical
test, the rigiddesign
mounted simplicity and was
condition highalso
dimensional
exposed stability in orbit.test
to the vibration In the
fortest, the rigid
comparison
mounted
with condition
the WSVI. Thewas also exposed
vibration to thethe
input from vibration test shaker
vibration for comparison with the
was obtained fromWSVI.
the
The vibration
reference input fromThe
accelerometer. the vibration
vibration responses
shaker wasfor obtained
each axis from theWSVI
of the reference
wereaccelerom-
obtained
from a three-axis
eter. The vibration accelerometer
responses for placed
each on
axisthe
ofdummy
the WSVI satellite, as shownfrom
were obtained in Figure 4. The
a three-axis
qualification
accelerometerlevelsplaced ofon
thethesine and random
dummy satellite,vibration
as showntest specifications
in Figure applied to lev-
4. The qualification the
design
els of theverification of the WSVI
sine and random are listed
vibration in Tables 4 and
test specifications 5, andtothe
applied theaxes of the
design test are
verification
shown in Figure
of the WSVI 4.
are listed in Tables 4 and 5, and the axes of the test are shown in Figure 4.

Figure 4.
Figure 4. Example
Example of
of the
the Launch
Launch Vibration
VibrationTest
TestSetup
Setupfor
forthe
thez-axis.
z-axis.

Table4.4.Specification
Table Specificationof
ofthe
theSine-vibration
Sine-vibrationTest.
Test.

Item
Item Specification
Specification
Direction
Direction x, y,
x, y,zz
Duration
Duration 22 oct/min
oct/min
Frequency [Hz]
Frequency [Hz] Acceleration [g]
Acceleration [g]
55 11
Profile
Profile
15
15 1.25
1.25
100
100 1.25
1.25

Table 5. Specification of the Random-vibration Test.

Item Specification
Direction x, y, z
RMS Acceleration 11.64 grms
Aerospace 2021, 8, 201 10 of 14

Table 5. Specification of the Random-vibration Test.

Aerospace 2021, 8, x FOR PEER REVIEW Item Specification 10 of 14

Direction x, y, z
RMS Acceleration 11.64 grms
Duration 2 min
Duration 2 min
Frequency [Hz] PSD [G2/Hz]
2
Frequency [Hz] PSD [G /Hz]
20 0.014
20 0.014
80 0.044
80 0.044
160 0.07
160 0.07
Profile
Profile 640 0.07
640 0.07
800
800
0.12
0.12
1150
1150 0.12
0.12
1300
1300 0.04
0.04
2000
2000 0.04
0.04

A modal analysis was performed to investigate the dynamic behavior of the dummy
A modal
satellite analysis
combined was
with theperformed
WSVI. Figureto investigate
5 shows thethe dynamic
results of thebehavior of the dummy
modal analysis. The
satellite combined with the WSVI. Figure 5 shows the results of the
first and second modes at 28 and 30 Hz, respectively, mainly represent the bending modal analysis.
mode The
first and second modes at 28 and 30 Hz, respectively, mainly represent the bending
of the blade modules for the lateral direction in the x- and y-axes. The third mode at 34 Hz mode
ofindicates
the bladethemodules
bendingfor the lateral
mode direction
of the blade in theofx-the
modules and y-axes.
WSVI The
in the third
axial mode at
direction 34 Hz
along
indicates the bending mode of the blade modules of the WSVI in the axial direction
the z-axis. The fourth mode at 40 Hz indicates the local bending mode of the center of the along
the z-axis.satellite.
dummy The fourth mode
These at 40
results Hz used
were indicates the local the
to investigate bending mode of thetest
launch-vibration center of the
results.
dummy satellite. These results were used to investigate the launch-vibration test results.

(a) (b)

(c) (d)
Figure Modal
5. 5.
Figure Analysis
Modal AnalysisResults
Results((a)
((a)1st
1stMode,
Mode, (b)
(b) 2nd Mode,
Mode, (c)
(c) 3rd
3rdMode,
Mode,and
and(d)
(d)4th
4thMode).
Mode).

Figure
Figure66shows
shows the sine vibration
the sine vibrationtesttestresults
results
forfor
thethe WSVI
WSVI during
during x-,and
they-,
the x-, y-, z-
and
z-axis excitation. In
axis excitation. Inthe
thecase
caseofofthe x-axisvibration
thex-axis vibrationresponse,
response, the
the highest
highest acceleration
acceleration of the
of the
dummy
dummysatellite
satellitewith
with the WSVI
the WSVIwaswas4.74.7
g atg 28
at Hz, which
28 Hz, corresponds
which to the
corresponds bending
to the mode
bending
ofmode
the SMA blade modules for the lateral direction of the WSVI along the
of the SMA blade modules for the lateral direction of the WSVI along the x-axis.x-axis. This value
isThis
almost similar to the estimated first eigenfrequency of 28 Hz from the modal
value is almost similar to the estimated first eigenfrequency of 28 Hz from the modal analysis
results shown
analysis in Figure
results shown 5. in In addition,
Figure the second
5. In addition, response
the was followed
second response by approximately
was followed by ap-
proximately 43 Hz, which was induced by the rotational mode of the SMA blade modules
for the lateral direction of the WSVI along the z-axis. The y-axis response shows that the
Aerospace 2021, 8, 201 11 of 14

Aerospace 2021, 8, x FOR PEER REVIEW 11 of 14

43 Hz, which was induced by the rotational mode of the SMA blade modules for the
lateral direction of the WSVI along the z-axis. The y-axis response shows that the highest
highest acceleration
acceleration of 5.3
of 5.3 g was atg28was
Hz.atThe28 Hz. Thetendency
overall overall tendency shows characteristics
shows characteristics similar tosim-
the
ilar
x-axis result, owing to the symmetric configuration of the test setup. The z-axisThe
to the x-axis result, owing to the symmetric configuration of the test setup. z-axis
response
response
shows theshows highest theacceleration
highest acceleration
of 11.1 g atof35
11.1
Hz,gwhich
at 35 Hz, whichtoisthe
is similar similar to thethird
estimated esti-
mated third eigenfrequency
eigenfrequency of 34 Hz. This of response
34 Hz. This response
is higher thanis that
higher thanx-that
of the andof the x-because
y-axes and y-
axes because of the coupling with the fourth mode of the structural
of the coupling with the fourth mode of the structural elastic mode of 40 Hz for the elastic mode of 40test
Hz
for the test dummy structure. In addition, the second and third
dummy structure. In addition, the second and third peak responses were followed at 72peak responses were fol-
lowed
and 98at Hz72from
and the
98 Hz from the
structural structural
mode mode ofsatellite.
of the dummy the dummy From satellite. From the sine-
the sine-vibration test
vibration
results, it can be seen that the highest acceleration response obtained fromobtained
test results, it can be seen that the highest acceleration response each of thefromx-,
each
y-, andof the x-, y-,
z-axes didandnotz-axes
exceeddid thenot exceed
design loadtheofdesign load of 23
23 g, derived g, derived
from the mass from the mass
acceleration
acceleration
curve (MAC)curve (MAC)
[22]. This [22]. This
indicates thatindicates
the WSVIthat the WSVI
is effective forisattenuating
effective forsineattenuating
vibration
sine vibration loads,
loads, as the design intends.as the design intends.

2
10
Input Profile (max. 2.5 g)
x-axis Res. (max. 4.7 g)
y-axis Res. (max. 5.3 g)
z-axis Res. (max. 11.1 g)

1
10
Acceleration (g)

0.1
10 100
Frequency (Hz)
Figure 6. Sine-Vibration
Sine-Vibration Test Results for x-, y-, and z-axes.

Figure 7a,b showshow the therepresentative

representativerandom-vibration
random-vibrationtest testresults forfor
results thethe
WSVI
WSVI during
dur-
the y-
ing y- z-axis
theand excitation.
and z-axis To compare
excitation. To comparethe vibration-reduction
the vibration-reductioncapabilities of the WSVI,
capabilities of the
the testthe
WSVI, results obtained
test results from the
obtained rigid-mounted
from condition
the rigid-mounted are alsoare
condition plotted in the Figure.
also plotted in the
The x-axis test result is not shown here, owing to the symmetrical configuration
Figure. The x-axis test result is not shown here, owing to the symmetrical configuration of the WSVI of
in the x and y planes, as mentioned in the sine-vibration test results. In the
the WSVI in the x and y planes, as mentioned in the sine-vibration test results. In the case case of Figure 7a,
theFigure
of first eigenfrequency of the dummyof
7a, the first eigenfrequency satellite without
the dummy the WSVI
satellite was observed
without the WSVIatwas 975 ob-
Hz
and a maximum
served at 975 Hz acceleration
and a maximum of 23.8 grms was observed,
acceleration of 23.8 grmswith
wasrespect to the
observed, withrandom
respecttest
to
input
the of 11.64
random testgrms . However,
input of 11.64 gthe
rms. maximum
However, the acceleration
maximumwas significantly
acceleration was decreased
significantly to
2.0 grms bytoapplying
decreased 2.0 grms by theapplying
WSVI, and the random-vibration
the WSVI, response was
and the random-vibration reduced
response wasbyre-a
factor of 11.9, comparing with the rigidly mounted condition. The
duced by a factor of 11.9, comparing with the rigidly mounted condition. The maximum maximum acceleration
response along
acceleration the z-axis
response along without the WSVI
the z-axis without wasthe114.8
WSVI grms
wasat114.8
260 Hz.
grmsThis
at 260response
Hz. Thisis
higher than that of the y-axis because the structural elastic modes
response is higher than that of the y-axis because the structural elastic modes of the of the dummy satellite
are dominant
dummy along
satellite arethe z-axis. However,
dominant along thethe maximum
z-axis. acceleration
However, of the dummy
the maximum satellite
acceleration of
the dummy satellite with the WSVI was also significantly reduced to 12.2, as shown in
Figure 7b. This indicates that the output response is reduced by a factor of 9.4.
Aerospace 2021, 8, 201 12 of 14

Aerospace 2021, 8, x FOR PEER REVIEW 12 of 14

with the WSVI was also significantly reduced to 12.2, as shown in Figure 7b. This indicates
that the output response is reduced by a factor of 9.4.

2
10
Input Profile (max. 11.64 grms)
with WSVI (max. 2.0 g rms)
1
10 w/o WSVI (max. 23.8 grms)

1
PSD Acceleration (g /Hz)
2

0.1

0.01

0.001

0.0001

-5
10

-6
10
100 1000
Frequency (Hz)

(a)

2
10

1
10
PSD Acceleration (g /Hz)

1
2

0.1

0.01

0.001

0.0001 Input Profile (max. 11.64 grms)

with WSVI (max. 12.2 grms)
w/o WSVI (max. 114.5 grms)
-5
10
100 1000
Frequency (Hz)

(b)

Figure
Figure7.7.Random-Vibration
Random-VibrationTest
TestResults
Results((a)
((a)y-axis,
y-axis,(b)
(b)z-axis).
z-axis).

Table66summarizes
Table summarizesthethefirst
firsteigenfrequencies
eigenfrequenciesofofthe
thedummy
dummysatellite
satellitewith
withthe
theWSVI
WSVI
inineach
eachaxis,
axis,obtained
obtainedthrough
throughthetheLLSS
LLSStests
testsperformed
performedbefore
beforeand
andafter
aftereach
each vibration
vibration
test.LLSS
test. LLSStests
testswere
wereperformed
performedtotovalidate
validatethe
thestructural
structuralsafety
safetyof
ofthe
theSMA
SMAmultilayered
multilayered
blade modules by investigating the dynamic responses of the dummy satellite with the
WSVI. The results show that the maximum first eigenfrequency shift was within 3.9%
Aerospace 2021, 8, 201 13 of 14

blade modules by investigating the dynamic responses of the dummy satellite with the
WSVI. The results show that the maximum first eigenfrequency shift was within 3.9%
throughout the test event, which was within the 5% criterion of the vibration test [23].
This indicated that no mechanical failures of the WSVI were observed, e.g., the plastic
deformation or delamination of the SMA multi-layered blades with viscous lamina. This
is because the interlaminated surfaces with the double-sided adhesive are effective in
resisting the shear force acting on the multilayered blade with energy-absorption effects.
These launch-vibration test results indicate that the SMA multi-layered blade is effective
for ensuring the structural safety of the WSVI itself and reducing the transmitted launch
loads to the dummy satellite, as the design strategy intended.

Table 6. Results of the WSVI’s Low Level Sine Sweep (LLSS) Tests Conducted before and after the
Launch-vibration Tests.

Excitation Corresponding Axis Frequency Shift

Test Status
Axis 1st Eigenfrequency (Hz) Difference (%)
x After 28.8 -
Before 28.2
y 0.35
Sine Vibration After 28.3
Before 34.2
z 0.87
After 34.5
Before 28.2
x 2.08
After 28.8

Random Before 27.9

y 3.9
Vibration After 26.8
Before 34.4
z 0.86
After 34.7

4. Conclusions
In this study, a three-axis passive WSVI was developed to significantly attenuate
the dynamic launch loads transmitted to a small satellite. To achieve a high damping
capability, the proposed WSVI applied two technical design concepts which are to use the
superelasticity of the SMA material and the other is to apply multilayered thin plates with
viscous lamina tapes on the SMA blades. The basic characteristics of the proposed WSVI
were investigated using static load tests with SMA blades of various thicknesses. The vibra-
tion test results to validate the effectiveness of the design showed great launch-vibration
isolation performance. From the sine-vibration test results, the highest acceleration re-
sponse obtained from each of the x-, y-, and z-axes are reduced 76%, 76%, 52% compared
with design load of 23 g. The results of random vibration test showed that the maximum
acceleration from each of the x-, y-, and z-axes are reduced 92%, 92%, 90% compared with
that of rigid mounted condition. Moreover, the structural safety of the WSVI was within
the qualification level of the vibration test specifications.

Author Contributions: Conceptualization, Y.-H.P. and H.-U.O.; methodology, Y.-H.P. and H.-U.O.;
software, Y.-H.P. and H.-U.O., formal analysis, Y.-H.P., S.-C.K. and H.-U.O.; validation, Y.-H.P.,
S.-C.K. and H.-U.O.; writing-original draft preparation, Y.-H.P.; writing-review and editing, H.-U.O.;
supervision, H.-U.O.; funding acquisition K.-R.K. and H.-U.O.; All authors have read and agreed to
the published version of the manuscript.
Funding: This research was funded by Hanwha Systems (U-19-001).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Aerospace 2021, 8, 201 14 of 14

Data Availability Statement: The data used to support the findings of this study are available from
the corresponding author upon request.
Acknowledgments: This research was supported by Hanwha Systems (U-19-001).
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Available online: http://trinitamonti.org/2020/07/15/the-newspace-paradigm-spacex-is-just-the-beginning/ (accessed on
15 May 2021).
2. Stephanie, K.; Martin, L. New Space: Impacts of Innovative Concepts in Satellite Development on the Space Industry. In
Proceedings of the 69th International Astronautical Congress (IAC), Bremen, Germany, 1 October 2018; pp. 1–5.
3. Ince, F. Nano and Micro Satellites as the Pillar of the “New Space” Paradigm. J. Aeronaut. Space Technol. 2020, 13, 207–221.
4. Vaios, L.; Vassilis, K. A Survey on Small Satellite Technologies and Space Missions for Geodetic Applications; IntechOpen: London, UK,
2020; pp. 1–22.
5. Wijker, J.J. Space Structures; Springer: Berlin, Germany, 2008.
6. Johnson, C.D.; Wilke, P.S.; Pendleton, S.C. SoftRide Vibration and Shock Isolation Systems that Protect Spacecraft from Launch
Dynamic Environments. In Proceedings of the 38th Aerospace Mechanisms Symposium, Williamsburg, VI, USA, 17–19 May 2006;
pp. 89–102.
7. Johal, R.; Christensen, J.; Doud, D. ORBCOMM Generation 2 Access to LEO on the Falcon 9 Using SoftRide, a Case History. In
Proceedings of the 26th Annual AIAA/USU Conference on Small Satellites, Logan, UT, USA, 13–16 August 2012.
8. Jun, Z.; Hongxing, H.; Zhiyi, Z. An evaluation of the whole-spacecraft passive vibration isolation system. Proc. Inst. Mech. Eng.
Part G 2007, 221, 67–72. [CrossRef]
9. Mastroddi, F.; Facchini, G.; Gaudenzi, P. Multi-frequency dynamic absorber for improved spacecraft comfort during the launch
phase. CEAS Space J. 2012, 3, 77–88. [CrossRef]
10. Rittweger, A.; Beig, H.G.; Konstanzer, P.; Bureo Dacal, R. Feasibility Demonstration of an Active Payload Adapter for Ariane
5. In Proceedings of the European Conference on Spacecraft Structures, Noordwijk, The Netherlands, 10–12 May 2005; pp.
149.1–149.11.
11. Otsuka, K.; Xu, Y.; Ren, X.B. Ti-Ni-Based Shape Memory Alloys as Smart Materials. Mater. Sci. Forum 2003, 426–432, 251–260.
[CrossRef]
12. Kwon, S.-C.; Oh, H.-U. Passive micro-jitter isolation of gimbal-type antenna by using a superelastic SMA gear wheel. Mech. Syst.
Signal Process. 2019, 114, 35–53. [CrossRef]
13. Kwon, S.-C.; Jeon, Y.-H.; Oh, H.-U. Micro-jitter attenuation of spaceborne cooler by using a blade-type hyperelastic shape memory
alloy passive isolator. Cryogenics 2017, 87, 35–48. [CrossRef]
14. Minesugi, K.; Onoda, J.; Ohkubo, H.; Hanawa, Y. Experimental Study on Damping 17 Capability of Thin Film with Vis-cous
Lamina. In Proceedings of the 37th Structure, Structural Dynamics and Materials Conference, Salt Lake City, UT, USA, 15–17
April 1996; pp. 1016–1023.
15. Bhattarai, S.; Kim, H.; Oh, H.U. CubeSat’s Deployable Solar Panel with Viscoelastic Multi-layered Stiffener for Launch Vibration
Attenuation. Int. J. Aerosp. Eng. 2020, 2020, 1–10.
16. Park, T.-Y.; Shin, S.-J.; Park, S.-W.; Kang, S.-J.; Oh, H.-U. High-damping PCB implemented by multi-layered viscoelastic acrylic
tapes for use of wedge lock applications. Eng. Fract. Mech. 2021, 241, 107370. [CrossRef]
17. Stoudt, M.R.; Ricker, R.E.; Cammarata, R.C. The Influence of a Multilayered Metallic Coating on Fatigue Crack Nuclea-tion. Int. J.
Fatigue 2001, 23, 215–223. [CrossRef]
18. Available online: https://www.aquariusplastics.co.uk/ (accessed on 12 May 2021).
19. Available online: https://www.3m.com/3M/en_US/company-us/all-3m-products/~{}/3M-Adhesive-Transfer-Tape-966/?N=
5002385+3293241971&rt=rud (accessed on 12 May 2021).
20. Kwon, S.-C.; Jeon, S.-H.; Oh, H.-U. Performance investigation of a novel pseudoelastic SMA mesh washer gear wheel with
micro-jitter attenuation capability. Smart Mater. Struct. 2016, 25, 55004. [CrossRef]
21. Kwon, S.-C.; Son, J.-H.; Song, S.-C.; Park, J.-H.; Koo, K.-R.; Oh, H.-U. Innovative Mechanical Design Strategy for Actualizing
80 kg-Class X-Band Active SAR Small Satellite of S-STEP. Aerospace 2021, 8, 149. [CrossRef]
22. Available online: http://www.vibrationdata.com/tutorials2/mass_acceleration_curves.pdf (accessed on 13 May 2021).
23. Oh, H.-U.; Lee, K.-J.; Jo, M.-S. A passive launch and on-orbit vibration isolation system for the spaceborne cryocooler. Aerosp. Sci.
Technol. 2013, 28, 324–331. [CrossRef]