Design and Verification of a Robust Release

Mechanism for CubeSat Deployables

Stefan Damkjar Collin Cupido Charles Nokes Ian R. Mann Duncan G.Elliott
Dept. of Electrical and Dept. of Physics Dept. of Physics Dept. of Physics Dept. of Electrical and
Computer Engineering University of Alberta University of Alberta University of Alberta Computer Engineering
University of Alberta Edmonton, Canada Edmonton, Canada Edmonton, Canada University of Alberta
Edmonton, Canada [email protected] [email protected] [email protected] Edmonton, Canada
[email protected] [email protected]

Abstract—The design of a simple and reliable release mecha-

nism for deployables on CubeSats is presented. The design relies
on a thermal knife to sever a Dyneema thread, releasing a spring
loaded deployable. Details of the design, technical challenges, and
the results of in-orbit testing of the final design are described.
Index Terms—deployable, nanosatellite, cubesat, satellite,
burnwire, hot knife, thermal knife

I. I NTRODUCTION
Many CubeSats require systems to release deployable on-
board systems such as booms, antennas, or folding solar
panels. Traditional solutions using frangible or flammable
materials, such as frangible bolts or paraffin actuators, used
on larger spacecraft to trigger deployment, are not suitable for
CubeSat applications when deployed from the International
Space Station (ISS) [1], which requires astronauts to carry the
spacecraft through crewed compartments. In this paper, the Fig. 1. Rendering of the deployment mechanism.
design of a simple and reliable mechanism for releasing spring
loaded deployable systems on CubeSats is described. Before TABLE I
launch, the deployable feature is held in a stowed position by M ATERIAL PROPERTIES FOR THREAD SELECTION
a Dyneema thread. When the release mechanism is activated, Longitudinal Longitudinal
Melting
a thermal knife is powered, melting the thread and allowing Material Tensile Tensile
Point
deployment. Successful deployment is verified by the release Modulus Strength
Kevlar (49) [5] 154 2800 550
of a microswitch otherwise held closed by the stowed de- Kevlar (29) [5] 61 2800 450
ployable. Threads and thermal knifes are often used to secure Dyneema [5] 115 3500 150
deployable features on CubeSats, however, material selection Nylon-6 [6] 2˜4 45˜90 220
(GPa) (MPa) (◦ C)
and physical design vary somewhat [2]–[4]. Justification for
the choices made in this design are given. The final design
was tested in orbit for the application to deploy a fluxgate
deployable, and through a small stainless steel eyelet, to the
magnetometer boom on the Experimental Albertan Satellite
other pair of washers. The eyelet ensures that the tension on
#1 (Ex-Alta 1) CubeSat. Technical challenges encountered and
the thread leaving the pair of washers is primarily parallel to
potential improvements for the design are discussed.
the surface of the PCB. To add redundancy, several threads can
II. M ECHANICAL D ESIGN be used. The spring force of the deployable feature holds the
The release mechanism, illustrated in Fig. 1, holds the thread firmly against the thermal knife to ensure good thermal
stowed deployable in a machined cradle made with glass fiber contact.
reinforced polyether ether ketone (PEEK) plastic. The cradle
III. T HREAD D ESIGN
is fastened, using stainless steel machine screws, to a PCB
substrate that is mounted on the outside face of the spacecraft. Several materials were considered for the thread includ-
Each end of the Dyneema thread is held securely between two ing Dyneema, Kevlar, and Nylon. The melting point, tensile
pairs of stainless steel washers. The thread runs from the first strength, and longitudinal tensile modulus of each of these
pair of washers, underneath the thermal knife element, over the materials, summarized in Table I, were considered.
 Dyneema thread has a high tensile strength, and has a However, Nichrome wire is fragile and difficult to fasten
relatively low melting point of 150 °C. These qualities make securely because traditional solder will not bond well to
Dyneema an appropriate material for this application. In partic- Nichrome wire.
ular, a low melting point makes it easier for the thermal knife
to cut the thread without risking thermal damage to nearby B. Self-limiting PTC Heater
components on the spacecraft. A higher tensile modulus such
Self limiting PTC heaters passively converge to a calibrated
as that of Kevlar (49) is desirable to minimize stretching of
temperature and are typically used for prolonged heating
the thread while the deployable is stowed, but the high melting
applications [8]. In nominal operation, a PTC heater should
point of Kevlar make it an unsuitable choice. Dyneema has
naturally maintain it’s temperature within a safe operating
some sensitivity to degradation from prolonged atomic oxygen
range, reducing the risk of damage to the spacecraft from
(AO) exposure [7]. In most cases deployable systems such as
excessive heating. However, the gradual convergence to a set
antennas are released promptly after the initial deployment of
temperature results in unacceptably slow deployment on the
the spacecraft, minimizing the risks of premature deployment
order of 10 ∼ 20 minutes. This will result in additional heat-
due to AO effects. In applications where significant AO
ing of sorrounding components that are otherwise protected
exposure will be a major concern, alternative thread materials
by interfaces with low thermal conductivity. It may also be
should be investigated.
problematic if the energy required to power the heater over
A. Release Verification a long period is significant compared to the small battery
capacity of the CubeSat.
A microswitch is fastened directly behind the PEEK cradle
with stainless steel machine screws, underneath the stowed
C. Epoxy Coated Helical Metal Film Resistor
deployable. The microswitch, which supplies digital signaling
to a microcontroller. is held closed by the deployable until it is Thru-hole helical metal film resistors are an attractive option
released. After receiving the signal from the opened switch or because they are widely available, easily mounted on a PCB
a time-out, the microcontroller turns off the thermal knife. This substrate, and are conveniently shaped for wrapping a thread
event can be recorded in the telemetry data. The microswitch around. The epoxy coating on the outside surface of the
should be carefully positioned to maximize travel of the switch resistor increases the thermal resistance between the resistor
lever, reducing the risk of the switch opening erroneously due and the thread. To quickly achieve the desired temperature at
to small movements of the deployable prior to its release. the surface of the epoxy coating within an acceptable time
period, the helical metal film should be sufficiently hot. This
IV. T HERMAL K NIFE D ESIGN high heat can cause warping or melting of the epoxy coating
To release the stowed deployables, electrical current is making this option unsuitable.
supplied to the thermal knife. Ideally the thermal knife should
quickly become hot enough to cut the thread, but never get hot D. Silicone Coated Helical Metal Film Resistor
enough to risk damage to the rest of the spacecraft. An ideal This solution is similar to epoxy coated helical metal
system should quickly achieve the melting temperature needed film resistors, but with a silicone coating more tolerant of
to sever the thread, avoid failure from prolonged actuation, and high temperatures. These resistors are typically intended for
not cause thermal damage to surrounding components. Four high temperature applications and in some cases are rated to
different thermal knife solutions were considered including operate up to 230 ◦ C [9]. This increased temperature tolerance
Nichrome wire, epoxy coated helical metal film resistors, reduces or eliminates warping or melting of the coating due
silicone coated helical metal film resistors, and self limiting to excessive heating of the internal helical metal film. The
positive temperature coefficient (PTC) heaters. The strengths operating point (resistor value and supply voltage) should be
and weaknesses of each of theses options are discussed in this carefully chosen to avoid excessive heating.
section. The specific component we used was CPF182R000FKE14,
Out of the four options considered, the best solution was an 82 ohm resistor manufactured by Vishay with a power
the silicone coated helical metal film resistor because it is rating of 1 W. To ensure rapid melting of the thread, allowing
easily mounted on a PCB substrate, is convenient for wrapping a short duty cycle, we exceed the 1 W rating. With a maximum
the thread around, and minimizes the potential for thermal supply voltage of 16.8 V, the maximum power dissipation was
damage. The power supplied to the resistor should be suffi- 3.5 W.
cient to ensure prompt deployment but low enough to avoid
damage. This may require some experimentation in individual V. T ECHNICAL C HALLENGES AND C ONSIDERATIONS
applications.
Deployable release mechanisms should be simple and reli-
A. Nichrome Wire able. Thorough testing is needed to uncover weaknesses in
Nichrome wire is widely used as a thermal knife for the design. In this section, some problems and significant
deployables. The lack of insulating coating allows excellent considerations concerning that arose throughout the evolution
thermal contact between the Nichrome wire and the thread. of the design are discussed.
A. Powering Resistors Beyond their Power Rating
In general, the power rating of resistors is chosen by the
manufacturer to ensure safe operation in normal ambient ther-
mal conditions. To achieve sufficient heating (which would be
excessive in the typical application of resistors as components
in a circuit) using regular through-hole resistors, it is typically
necessary to power the resistor beyond the manufacturer’s
recommended power rating. This increases the likelihood of
the component failing. This problem is reduced with the
use of Silicone coated resistors that are specifically intended
for demanding, high heat applications and therefore have a
greater tolerance for excessive heating and therefore typically
have much higher recommended power ratings. Depending
on the application, it may still be necessary to exceed the
recommended power rating, but the risk of failure is still Fig. 2. Cross-section of the stowed elbowed boom
significantly reduced by using components designed for high
heat applications. In our design, using an 82 ohm resistor with
a power rating of 1 W, the maximum power dissipated by the in a release of the microswitch, implying boom deployment,
resistor was 3.44 W, assuming a maximum supply voltage of whilst the boom in fact remained stowed. The tension on the
16.8 V. thread arising from the deployment springs on the magne-
tometer boom was higher than that of the other deployables,
B. Thread Beading and Catching on Eyelets and securing the two elements of the boom required a longer
Our initial design had an eyelet between the resistor and thread. These factors would have increased the possibility of
the deployable, minimizing the mechanical load on the re- the thread stretching, potentially resulting in the microswitch
sistor, but causing deployment failures during testing. As the falsely reporting that deployment had occurred. One way to
Dyneema thread is melted by the thermal knife, the molten mitigate this issue in the future is to add a feature that will
polymer accumulates on the severed end of the thread and maintain tension on the thread, absorbing any slack caused by
hardens. If the thread then passes through an eyelet, this excess stretching. Dyneema fiber has a negative coefficient of thermal
material may catch on the eyelet, preventing deployment. This expansion [11] so lengthening due to heating is not a likely
issue also poses a risk to deployables that are very thin, or cause of this issue.
that have sharp edges that could easily catch or hook onto the
VI. P OTENTIAL I MPROVEMENTS
excess material. The configurations shown in Fig. 1 and Fig. 2
are designed to be immune to this potential problem. Anyone A. Solder and Substrate Selection
attempting to reproduce this work should take care to do the Selecting a thread material with a suitably low melting
same. point does reduce the risk of thermal damage to surrounding
components by reducing the heat required from the hot knife.
C. Heat Damage to PCB Substrate However, if actuation is prolonged (on the order of tens of
During prolonged heating of the hot knife, thermal damage minutes) due to environmental factors or malfunction, it is
to the surrounding area is a major concern. The PCB substrate possible that thermal damage will still occur at the mounting
surrounding the hot knife, and the solder joints holding it in points of the thru-hole resistor. To avoid this problem, it would
place are vulnerable to excessive heating. To minimize this be ideal to select solder (such as pure lead) and PCB sub-
risk, materials with a higher melting point such as polyimide strate material (such as polyimide [12]) with sufficiently high
PCB substrate are worth considering. maximum operating temperatures. These maximum operating
temperatures should be equal to or higher than the 230 ◦ C
D. Failure of Release Verification Microswitch
maximum operating temperature of the silicone coated hot
This deployment release mechanism was tested in orbit knife resistor.
aboard the Ex-Alta 1 CubeSat. This satellite had nine deploy-
ables including four antennas, four needle Langmuir probes, B. Locking Resistor Leads
and an elbowed fluxgate magnetometer instrument boom [10] If excessive heating causes the solder to melt, it is possible
which were all held in place whilst stowed using the deploy- that the resistor will separate from the PCB substrate. To avoid
ment system described here. Deployment occurred nominally this risk, a resistor with formed locking leads, as shown in Fig.
except for issues with the microswitch which was intended 3 could be used to increase the force needed for separation,
to detect deployment of the elbowed fluxgate magnetometer making the design more reliable. Alternatively, the resistor
boom. Fig. 2 shows a cross-sectional view of the elbowed leads could be looped through a second set of mounting holes
boom in its stowed configuration. One possible cause for this as shown in Fig. 4. Although we haven’t tested this, it may
failure was stretching of the Dyneema thread which resulted also improve durability if the resistor is secured using high
temperature epoxy as shown in Fig. 5 on both end faces of likely as a result of stretching of the thread holding the boom in
the resistor. place before deployment. This weakness could be mitigated be
including a feature to maintain tension on the thread. Further
improvements are possible with more exhaustive consideration
of material properties and mechanical design.
ACKNOWLEDGMENT
The authors acknowledge the contributions of Miroslaw
Ciurzynski, David Barona, David Miles, Dan Sameoto, Carlos
Lange, David Milling and Chris Robson in developing the
design of the CubeSat fluxgate magnetometer boom deployed
with the release mechanism described here. The authors are
grateful to the many University of Alberta AlbertaSat members
who participated in the design and construction of the Ex-Alta
1 satellite, deployed from the International Space Station in
Fig. 3. Through-hole resistor with formed locking leads May 2017. Funding from the Canadian Space Agency (CSA)
and the Natural Sciences and Engineering Research Council
(NSERC) of Canada made this work possible. The Ex-Alta 1
project received funding from the European Union’s Seventh
Framework Programme for Research and Technological De-
velopment under grant agreement 284427 through the QB50
mission. This publication reflects the views only of the authors,
and the European Union cannot be held responsible for any
use which maybe made of the information contained therein.
R EFERENCES
[1] T. Prejean, “NanoRacks cubesat deployer (NRCSD) interface definition
Fig. 4. Through-hole resistor with additional mounting holes to improve document (IDD),” NanoRacks, Tech. Rep. NR-NRCSD-S0003, Jun.
durability in-case excessive heat causes the solder joints to melt 2018.
[2] G. F. Brouwer, W. J. Ubbels, A. A. Vaartjes, and F. T. Hennepe,
“Assembly, integration and testing of the DELFI-C3 nanosatellite,” in
59th IAC, no. IAC-08-D1.5.6, Glasgow, Scotland, Oct. 2008.
[3] S. Jeon and T. W. Murphey, “Design and analysis of a meter-class
cubesat boom with a motor-less deployment by bi-stable tape springs,”
in 52nd AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics
and Materials Conference, Denver, CO, 2011, pp. 1–11.
[4] K. Nakaya, K. Konoue, H. Sawada, K. Ui, H. Okada, N. Miyashita
et al., “Design & development of flight model and future plan,” in 21st
International Communication Satellite Systems Conference and Exhibit,
Yokohama, Japan, 2003, pp. 1–10.
[5] J. Pilling. (2006, Apr.) Mechanical properties data. Dept.
of Metallurgy, University of Manchester. [Online]. Available:
http://www.mse.mtu.edu/ drjohn/my4150/props.html
[6] (2003) Young’s modulus - tensile and yield strength for
common materials. Engineering ToolBox. [Online]. Available:
https://www.engineeringtoolbox.com/young-modulus-d 417.html
[7] M. M. Finckenor, “Comparison of high-performance fiber materials
Fig. 5. Through-hole resistor with high temperature epoxy on both end faces properties in simulated and actual space environments,” Marshall Space
(adapted from [13]) Flight Center, techreport, Jul. 2017.
[8] S. Azechi and T. Nakamura, “Electrically conductive silicone rubber
composition,” US patentus 6 734 250B2, May, 2004.
[9] “Metal film resistors, axial, industrial power, precision, flameproof,”
VII. C ONCLUSION Vishay Dale, Tech. Rep. 31021, Dec. 2016.
The design presented here represents a low cost and low [10] D. M. Miles, I. R. Mann, M. Ciurzynski, D. Barona, B. B. Narod,
J. R. Bennest, I. P. Pakhotin, A. Kale, B. Bruner, C. D. A. Nokes,
risk, space validated, design for the release of deployable C. Cupido, T. Haluza-DeLay, D. G. Elliott, and D. K. Milling, “A
systems on nanosatellite and CubeSat applications. The design miniature, low-power scientific fluxgate magnetometer: A stepping-
may have widespread utility for low cost CubeSat projects stone to cube-satellite constellation missions,” Journal of Geophysical
Research: Space Physics, vol. 121, no. 12, pp. 11,839–11,860, 2016.
such as those being developed by University student teams. [11] A. Yamanaka, T. Kashima, M. Tsutsumi, K. Ema, Y. Izumi, and S. Nishi-
The risk of thermal damage to surrounding components on jima, “Thermal expansion coefficient of unidirectional high-strength
the spacecraft is minimized by using a thread with a relatively polyethylene fiber reinforced plastics at low temperature,” Journal of
Composite Materials, vol. 41, no. 2, pp. 165–174, 2007.
low melting point. [12] “85n polyimide laminate and prepreg,” Arlon EMD, Tech. Rep., 2009.
One release verification sensor comprising a microswitch [13] Workmanship Standard for Polymeric Application on Electronic Assem-
was prematurely triggered on-orbit prior to deployment, most blies, NASA Std. NASA-STD-8739.1A, Rev. 2, Mar. 2011.