Content uploaded by Márton Géczi
Author content
All content in this area was uploaded by Márton Géczi on Sep 24, 2020
Content may be subject to copyright.
71th International Astronautical Congress, xx. Copyright c
2020 by Delft Aerospace Rocket Engineering (DARE). Published
by the IAF, with permission and released to the IAF to publish in all forms.. All rights reserved.
IAC–20–D2,6,8,x56366
Development of a low cost, low altitude test vehicle for high dynamic
pressure parachute testing
M. G´eczi, L. Pepermans G. Kandiyoor, O. Dvorak
Parachute Research Group (PRG) of Delft Aerospace Rocket Engineering (DARE), the Netherlands,
prg@dare.tudelft.nl
In order to increase the reliability of parachute systems, Delft Aerospace Rocket Engineering (DARE) re-
quired a test vehicle that could reach dynamic pressures comparable to a Stratos-type mission. This project
became known as the DARE Parachute Investigation Project or PIP.
The initial PIP launcher flies to an apogee of about 1000 meters where the test section separates from the
engine section. The test section follows a ballistic trajectory until the parachute is deployed. The engine
section has a parachute that allows for safe recovery and reuse. When launched to one kilometre, the
deployment conditions are expected to reach dynamic pressures of about 5 kPa.
Currently, there are four versions of PIP considered for flight. The first version is the proof of concept
that allows for demonstrating the capabilities of the launcher. The second version is meant to be used
for parachute tests. This version allows for modifications to the mass of the test section to tailor the test
conditions to the requirements of the user. The third version has a second parachute in the nose cone to
allow for testing of smaller parachutes that do not decelerate the test section enough to allow for safe landing
and reuse. The fourth version can reach an altitude of around 2500 meters to increase the test envelope.
The paper focuses on the feasibility of low altitude sounding rockets for testing main parachutes for future,
more demanding missions. The first test flight of PIP was completed on in September 2019 and was partially
successful.
I. Introduction
Parachute testing has been a notoriously difficult
element of parachute recovery systems. Either one
can test limited parts of the system, or tests are
costly. Delft Aerospace Rocket Engineering (DARE),
a student team with the goal of reaching space, has
been facing similar problems.
During the Stratos III project, many experiments
were done to prove the capabilities of the parachute
system. These experiments included wind tunnel
testing in the Open Jet Facility of TU Delft, ma-
terial property testing, tensile testing, and tabletop
systems. These tests gave the team a proper overview
of the limits of the systems, but significant safety fac-
tors were applied. These safety factors were mainly
applied to the inflation loads of the parachutes, lead-
ing to over-engineered parachute canopies and lines.
Two missions were proposed to be able to de-
crease the required safety factor on the parachute
systems. These are the Supersonic Parachute Exper-
iment Aboard REXUS (SPEAR) and the Parachute
Investigation Project (PIP). SPEAR focuses on the
supersonic inflation of the drogue parachute and the
loads involved in that process. PIP focuses on the
inflation behaviour of the main parachutes.
The parachute research group had performed ear-
lier flight test experiments onboard the DARE Cansat
v7 launcher. These experiments were done to demon-
strate the main parachute stability in free flight and
to investigate the main parachute inflation. The
major drawback of testing parachutes on board the
Cansat v7 launcher is that the parachute is deployed
from the nose cone. This method of parachute de-
ployment means that one, the parachute can only be
deployed at apogee as the dynamic pressure prevents
a proper ejection from the nose cone. Second, there
is an unpredictable interaction between the aerody-
namically unstable rocket body and the parachute.
However, the modality of the Cansat v7 launcher did
allow for it to serve as a technical basis.
A Cansat v7 flight using an upgraded engine,
dubbed Cansat Heavy, demonstrated the ability of
DARE to fly to 2.5 km in the Netherlands. Dur-
ing this flight, the drogue parachute did not function
leading to the main parachute shredding to pieces at
IAC–20–D2,6,8,x56366 Page 1 of 8
71th International Astronautical Congress, xx. Copyright c
2020 by Delft Aerospace Rocket Engineering (DARE). Published
by the IAF, with permission and released to the IAF to publish in all forms.. All rights reserved.
inflation. This flight, accidentally, demonstrated the
ability of DARE to fly a mission within the Nether-
lands that exceeds the dynamic pressure limits of the
main parachutes. During parachute deployment, the
vehicle was travelling with 140 m/s and the parachute
was inflated at a dynamic pressure of 11.3 kPa. The
parachute onboard was the engineering model of the
SPEAR and Stratos disk gap band main parachute.
II. Mission
The main parachute of the Stratos III mission was
a 2m2parachute that would deploy at about 1000 me-
ters while flying Mach 0.2. The parachute envelope
limits for Stratos III were set to Mach 0.4 and 34 kPa
dynamic pressure. These values were the inputs for
the PIP project.
The trajectory of the PIP rocket on a standard
parachute test flight can be found in Fig. 1. The PIP
rocket takes off from the standard DARE tower used
to launch other DARE sounding rockets. During the
first few seconds, a solid rocket engine propels the
rocket. The predicted apogee of a standard PIP ve-
hicle is about 1 km but is influenced by the take-off
mass of the vehicle. At apogee, the test section of
the vehicle separated the engine section. The engine
section has a standard cross parachute ensuring safe
landing and reusability of the engine section. The
test section has its own fins and is thus aerodynami-
cally stable and is allowed to freely fall back to earth
up to the point where the experimental parachute is
deployed. The deployment of the parachute is done
on timers in the current version, but a version using
pressure sensors inside a timer window is considered.
Main mission requirements for the PIP mission are
as follows:
1. Cost shall not be higher than e500
2. Shall test parachutes at Mach 0.3 and a dynamic
pressure of at least 34 kPa
3. Apogee shall be no higher than 2.5 km
4. Flight time shall be a maximum of 90 seconds
The first requirements originate from the fact that
PIP is part of a student project. Plus the fact that
it is desirable to launch more than once. Keeping
cost low is thus a vital requirement. The last two
requirements originate from the launch site. The de-
sired launch location, ASK ’t Harde in the Nether-
lands, has a relatively small exclusion zone limiting
the flight ceiling and flight time.
When operational PIP shall be able to provide sev-
Fig. 1: Trajectory of the PIP rocket
eral clear advantages over test flying with a Cansat v7
or performing wind tunnel experiments in the Open
Jet Facility of Delft University. The advantages are
listed below:
1. Measure the parachute inflation loads at Stratos-
like conditions
2. Deceleration of the rocket caused by the
parachute measured with a predictable body
3. Static parachute drag coefficient measured in
free flight
III. Conceptual design of PIP
The original PIP launcher relied heavily on the
experience of the CanSat v7 launcher. To reduce the
design effort the engine section of Cansat was used
with a new test section on top. This design can be
seen in Fig. 2.
IV. Performed Flights
The first flight of PIP was performed in September,
2019. The aim of this flight was to serve a proof of
concept. In line with this aim, the flight did not serve
as a testing platform for any parachutes.
In order to provide further insight into this proof
of concept flight, it is important to address the chal-
lenges faced in the manufacturing, assembly and
launch processes by the team.
IAC–20–D2,6,8,x56366 Page 2 of 8
71th International Astronautical Congress, xx. Copyright c
2020 by Delft Aerospace Rocket Engineering (DARE). Published
by the IAF, with permission and released to the IAF to publish in all forms.. All rights reserved.
Fig. 2: Original PIP rocket concept
IV.i Production challenges
One of the primary challenges faced during pro-
duction pertained to the addition of extra weight and
larger fins due to the insufficient stability margin. In
addition to a steel bar in the nose, larger fins were also
found to aid the stability margin. Due to this, the
size of the engine section fins was increased. While it
was found that this resulted in an acceptable stabil-
ity margin on the overall vehicle, the stability margin
was unsatisfactory on the test section. With the aim
of testing the vehicle at as high a dynamic pressure
as possible, the engine section fins had to be enlarged
even further. This resulted in the overall vehicle as
well as the test section satisfying the stability margin
requirement.
The second production challenge faced was fitting
the fins inside the canister. The parachute deploy-
ment device assembly contains a canister used to hold
the test section parachute during flight. Further, the
test section fins are held in place by the internal struc-
ture of this canister. As a result of errors in the 3D
process, it was not able to push the fins into their fi-
nal positions as desired. Despite being suboptimal, a
temporary solution was utilised to resolve this issue,
which allowed satisfactory positioning.
Beyond these two challenges, a few minor ones
were also encountered. This included the slight mis-
printing of the rings of the clamp band, the lack of
a component to hold the battery in place and the
penetration of screws into the parachute compart-
ment. The final one is considered as it could serve
as a threat to the parachute. Despite these, none
served as an obstacle to the progress of the proof of
concept flight and further rectifying action has been
taken to address these for future flights.
IV.ii Assembly challenges
In addition to manufacturing challenges, certain
challenges were also faced during assembly. The first
one pertained to the nose cone, where the placement
of the steel bar required the nose cone to be cut, and
subsequently loosely joined to the test skin section.
Furthermore, the cabling was not taken into ac-
count during the design, requiring slight modification
to allow these into the stringer.
Finally, the alignment of the clamp band holes pre-
sented a challenge. The assembly team did not feel
that the final assembly was particularly simple and
changes have been made in this regard. In addition,
the lack of parachute attachment points in either sec-
tion and the weak servo posed additional difficulties.
Both of these challenges were overcome and did not
act as a barrier to the flight.
IV.iii Challenges associated with launch
During the proof of concept flight launch, two
main challenges were faced. The first one was that
the enlarged fins meant the rocket could not fit into
the launch tower, if all the engine section fins were
present. This is due to the manner in which the
launch lugs are placed. The lower launch lug is po-
sitioned in the rail, following which the rocket is ro-
tated and finally the top launch lug is inserted. This
was rectified by adjusting the number of fins, which
allowed a successful flight.
Secondly, the weather conditions during the launch
IAC–20–D2,6,8,x56366 Page 3 of 8
71th International Astronautical Congress, xx. Copyright c
2020 by Delft Aerospace Rocket Engineering (DARE). Published
by the IAF, with permission and released to the IAF to publish in all forms.. All rights reserved.
meant that the rocket was not visible after entering
the clouds. It was noted that after losing visual track-
ing, only the engine section was recognisable, due to
the parachute being bright orange. No visual tracking
of the test section was successful.
IV.iv Post flight visual inspection
Following the proof of concept flight, where an
apogee between 750 and 800 metres was achieved,
various parts of the rocket were successfully recov-
ered and inspected.
Firstly, the test section showed that the nose cone
was destroyed, along with which one stringer was
bent. Due to the impact the fins detached. Fur-
thermore, the parachute was still present in the can-
ister, and the electronics compartment was damaged.
These issues have been analysed and improvements
had been made in subsequent flights.
Secondly, the clamp band was recovered, despite
expectations and exceeding requirements. It was
found to be mostly intact, except for one piece that
was missing, likely as a result of impact with the
ground.
Finally, it was noted that the parachute of the en-
gine section opened. The parachute deployment was
successful with no line tangling. The skin as well as
one fin were damaged suggesting the skin was too thin
and the fins too large causing them to fail under the
impact loads. Furthermore, video footage from the
flight suggests the landing velocity may have been too
high. This was likely linked to the changes made due
to the stability margin and has since been analysed
for areas of improvement. Despite this, the engine
section can generally be considered successful as it
was recovered and the parachute deployed in a nom-
inal fashion.
Overall, the proof of concept performance flight
in September 2019 fulfilled the aim of establishing a
proof of concept.
V. Vehicle overview
During the design of the rocket, it was prioritised
to keep the cost as low as possible. This require-
ment was necessary to make it a viable alternative
to wind tunnel testing. The vehicle is to be reused
several times with little to no refurbishment between
launches. A new vehicle would only be needed when
non-repairable damage is encountered due to test
parachute failure.
V.i System breakdown
A render of the rocket can be seen in Figure 3. On
the bottom, the engine section is located that houses
the motor, the pin puller mechanism and the engine
section parachute. On top of it lies the test section
with the parachute deployment device (PDD), the
flight computer and a black box. At the very top
beneath the nose cone, an adjustable weight system
is located for stability reasons. The main dimensions
of the rocket can be seen in Table 1.
Fig. 3: Render of PIP
IAC–20–D2,6,8,x56366 Page 4 of 8
71th International Astronautical Congress, xx. Copyright c
2020 by Delft Aerospace Rocket Engineering (DARE). Published
by the IAF, with permission and released to the IAF to publish in all forms.. All rights reserved.
Table 1: Main dimensions of the PIP rocket
Parameter Value
Engine section (ES) length 860 mm
ES diameter 120 mm
ES fin area 14000 mm2
Clamp band max diameter 155 mm
Test section (TS) length 750 mm
TS diameter 120 mm
TS fin area 4250 mm2
Nose cone length 200 mm
Furthermore, the aim was to make the system as
easy to produce and assemble as possible. This helps
easy reproducibility as also new members of DARE
can build the rocket without any prior machining
knowledge.
Engine Section
The ES houses the flight-proven DARE DX1 mo-
tor also used in the Dutch CanSat competition. The
characteristics of the engine can be seen in Table 2.
Table 2: DARE DX1 engine parameters
Parameter Value
Peak thrust 1500 N
Total impulse 2.7 kNs
Specific impulse 106 s
Burn time 3.5 s
Propellant mass 2.5 kg
The skin is rolled from 1 mm-thick laser cut alu-
minium. To which four two mm thick aluminium
bulkheads are attached to support the engine in the
horizontal direction and another ten mm-thick bulk-
head is used to transfer the thrust force. The servo
of the separation mechanism is also located on this
bulkhead. A conventional powerful metal gear servo
is used to guarantee the pin puller mechanism al-
ways works, which can be seen in Figure 4. The ES
parachute is packed tightly on top of the assembly
but also takes some space in the bottom of the TS.
Separation system
To separate the engine section and test section a
clamp band is used. The clamp band system has been
used in various DARE missions such as the Stratos
sounding rocket and the SPEAR supersonic exper-
iment. A clamp band performs well in both hold-
Fig. 4: Pin puller mechanism on top of the engine
bulkhead
ing the rocket elements together and separation upon
command.
The clamp band is used to keep the ES and TS
together during ascent. A ring is attached to both
sections. These are tightly held in place by the clamp.
The full assembly is 3D-printed.
Fig. 5: Clamp band
Test Section
On the bottom of the test section the PDD is lo-
cated and can be seen in Figure 6. The TS fins are at-
tached to the canister by an internal geometry. This
is possible as these parts are 3D-printed. The can-
IAC–20–D2,6,8,x56366 Page 5 of 8
71th International Astronautical Congress, xx. Copyright c
2020 by Delft Aerospace Rocket Engineering (DARE). Published
by the IAF, with permission and released to the IAF to publish in all forms.. All rights reserved.
ister itself contains the TS parachute. On top, the
actuation device is located. This can be seen in Fig-
ure 7. The servo pushes on an arm that releases the
push plate. A conical spring is used to shoot out the
parachute. The flight computer with the black box
is located at around two-thirds of the section. In the
nosecone, the adjustable weight system is located and
can be seen in Figure 8. In the current configuration,
1.8 kg of steel is used for stability reasons of the TS.
Fig. 6: Parachute deployment device
Fig. 7: Parachute actuation device
Similarly to ES, the TS skin is also rolled from
1 mm-thick laser cut aluminium. The PDD is 3D-
printed except for the 2 mm aluminium release rod.
Fig. 8: Adjustable weight system in the nosecone
The flight computer compartment is also 3D-printed,
while the black box is embedded in a steel casing. The
steel weight in the nosecone is resting on a ten mm-
thick aluminium bulkhead. The nosecone assembly
is 3D-printed.
V.ii Stability
The stability of the rocket is of high importance
as the test section is relatively empty as there is only
the PDD and the flight computer inside, thus it can-
not pull the centre of gravity forward to counteract
the heavy engine. This would result in an insuffi-
cient stability margin. For this reason, the previously
mentioned weight in the nosecone was implemented
as increasing the ES fin area was deemed unfeasible.
The stability margin of the full rocket is 1.66, while
for the test section it is 1.58.
V.iii Manufacturing
During the design, manufacturing was one of the
top priorities from both ease of reproducibility and
cost perspective as well.
Four categories of processes can be distinguished:
1. 3D printing: several facilities are available on
TU Delft campus at different faculties and at
the machining hall for DARE members
2. Laser cutting at TU Delft: Mechanical, Mar-
itime and Materials Engineering offers laser cut-
ting for student for up to 1 mm of aluminium
IAC–20–D2,6,8,x56366 Page 6 of 8
71th International Astronautical Congress, xx. Copyright c
2020 by Delft Aerospace Rocket Engineering (DARE). Published
by the IAF, with permission and released to the IAF to publish in all forms.. All rights reserved.
3. Laser cutting at third-party companies: as the
possibility of laser cutting at TU Delft is lim-
ited, an outside company manufactures thicker
aluminium parts
4. Lathing: lathes are available at the machining
hall for DARE
The 3D-printed parts are the one coloured green
in Figures 3 to 8. These involve the pin puller mecha-
nism, the clamp band, the PDD, the flight computer
bulkhead and the nosecone. The skin sheet of both
ES and TS are laser cut at TU Delft premises. All
bulkheads and fins are laser cut at a third-party. Self-
lubricating launch lugs and the steel black box.
V.iv Cost breakdown
The costs can be split into two categories: building
cost to manufacture a full rocket and refurbishment
cost.
Firstly, the manufacturing of a full rocket is de-
tailed in Table 3. This involves all costs except the
expenses related to manufacturing of the engine and
the parachutes.
Table 3: Cost of a full rocket
Part(s) Cost (e)
Bulkheads & fins 100
ES & TS skin sheet 30
Servos 60
Flight computer 150
Black box 20
Smaller components 50
Total 410
Ideally, if a test is successful, both sections are re-
trieved and are unharmed. In this case only inspec-
tion and testing is needed between flight. Naturally,
the rocket is also to test experimental parachute that
might not be able to safely land the test section. In
this case, a full new test section has to be manufac-
tured with related costs in Table 4.
V.v Control system
During the flight the separation of the rocket sec-
tions and the deployment of the TS parachute have
to be actuated. The separation occurs at apogee and
the parachute deployment at a desired airspeed.
For simplification a 1-D trajectory model is used.
To ensure desired operation it is necessary to know
both the vertical position and the vertical velocity
of the rocket. These quantities are acquired by fus-
ing the data from a three-axis accelerometer and a
pressure sensor by means of a Kalman filter.3
Table 4: Cost of refurbishment in case of test section
loss
Part(s) Cost (e)
Bulkheads & fins of TS 50
TS skin sheet 15
Servo 30
Flight computer 150
Black box 20
Smaller components 25
Total 290
In optimal case both the separation of sections and
the parachute deployment will occur when a specific
vertical velocity is reached. However to account for
possible measurement errors or failure to reach these
velocities there are present redundant conditions that
will trigger the actuation of these events in such case
to ensure that the rocket survives and the mission
can provide at least less precise data on parachute
behaviour. For the same reasons, events can also
be actuated only in specific time windows based on
OpenRocket4simulations. System logic flow can be
seen in figure 9.
A modular set of electronics was developed by
DARE Electronics Team to be universally used in
smaller DARE rockets.5This approach minimises
workload and gives access to a well tested and reli-
able platform, where specific behaviour for individual
rockets can be easily implemented. In case of PIP it
allows to measure and process both acceleration and
static pressure, store data and actuate servos accord-
ingly. It also controls the launch sequence in a man-
ner standard for DARE organisation, ensuring that
no potentially dangerous misunderstanding of system
occurs. Another useful feature is ability to store flight
data in multiple places. To ensure that at least one
SD card with flight data survives any rocket malfunc-
tion, there is an aluminium-steel black box placed in
the TS.
VI. Outlook
At the time of the publishing of this articles, there
is one rocket in construction. Due to the COVID-
19 pandemic, it is uncertain when the team will be
able to have access to machining facilities and finish
the rocket. When it is completed, the rocket is ready
to test parachute. As a proof of concept flight has
already been performed, the next launch can be an
actual test. When DARE and TU Delft returns to
normal operations allowing to manufacture & oper-
IAC–20–D2,6,8,x56366 Page 7 of 8
71th International Astronautical Congress, xx. Copyright c
2020 by Delft Aerospace Rocket Engineering (DARE). Published
by the IAF, with permission and released to the IAF to publish in all forms.. All rights reserved.
Fig. 9: Logic flow of the system
ating more rockets at three to four launch days will
provide an immense advantage over wind tunnel test-
ing as the obtained data is more representative of ac-
tual flight conditions.
Furthermore, PIP can be upgraded for higher
apogee thus enlarging the testing envelope. This
can be done by relatively small number of changes
while still leaving the main elements of the rocket
unchanged. The team has already investigated the
process of upgrade and was deemed feasible to design
and build if needed.
Once, the launcher is mature enough after a cou-
ple of launches, the system will provide valuable and
representative data for future DARE missions.
References
[1] L. Pepermans et al., ”Flight Simulations of the
Stratos III Parachute Recovery System”, Inter-
national Astronautical Congress 2018.
[2] L. Pepermans, M. Rozemeijer, E. Menting, N.
Suard, S. Khurana, ”Systematic Design for a
Parachute Recovery System for the Stratos III
Student Build Sounding Rocket”, AIAA Avia-
tion Forum 2018.
[3] D. Schultz, ”Application of the Kalman Filter to
Rocket Apogee Detection”, 2004
[4] OpenRocket [Computer software], Retrieved
from http://openrocket.info
[5] DARE modular set of rocket electronics,
2016 https://dare.tudelft.nl/2016/06/
a-modular-set-of-rocket-electronics/
IAC–20–D2,6,8,x56366 Page 8 of 8
