Rocket Flight Simulation Program

Steen Eiler Jrgensen

17. januar 2014
Indhold
1 Introduction 2
1.1 Motor data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2 Simulation 2
2.1 Thrust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.2 Atmospheric model . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2.1 Air temperature . . . . . . . . . . . . . . . . . . . . . 3
2.2.2 Air pressure . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2.3 Air density . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2.4 Air resistance . . . . . . . . . . . . . . . . . . . . . . . 3
2.3 Mass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.4 Gravitational force . . . . . . . . . . . . . . . . . . . . . . . . 4
2.5 Total force . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.6 Acceleration . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.7 Velocity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.8 Altitude . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3 Two-dimensional ight 5
4 Parachute Timing 5
4.1 General Considerations . . . . . . . . . . . . . . . . . . . . . . 5
4.2 Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
5 Todos 6
A Rocket Stability Checklist 6
1
1 Introduction
This document describes the function of the rocket ight simulation program
by Steen Eiler Jrgensen. The program consists of a single .m-le for execution
in GNU Octave. The program was written to account for lacks in other rocket
ight simulation programs:
RASAero only uses imperial units, and is cumbesome to use,
Launch only allows a total initial rocket mass of 1000 kg,
VRM only allows vertical rocket motion,
1.1 Motor data
Motor data is read in standard (XXX?) format, where the rst line denes
an array consisting of
1. the propellant mass [kg]
2. the mass of the engine after ring [kg]
after which comes another array consisting of pairwise values for time [s] and
thrust [N]. For example, the data le for a Copenhagen Suborbitals HATV
booster might look like this:
HATVdata = [74,63];
HATV = [
0.1, 14993;
2.0, 11867;
4.0, 11733;
24.0, 1400;
29.9, 7;
30.0, 0
];
2 Simulation
For each step in the simulation, the following calculations are performed. All
values pertaining to the current iteration are denoted with a subscript i.
2.1 Thrust
Engine thrust F
e,i
is derived from the engine thrust curve. Thrust values
between data points are linearly interpolated.
2
2.2 Atmospheric model
2.2.1 Air temperature
Air temperature T
i
is calculated in correspondence with the International
Standard Atmosphere from the height h
i1
as found in the previous iteration.
2.2.2 Air pressure
Air pressure is calculated according to a rough t with an exponential
function to the altitude-pressure relations given in the International Standard
Atmosphere:
p
i
= p
0
e
0.000135134 h
i1
[Pa] (1)
where p
0
is the air pressure at sea level (usually 101325 Pa) and h
i1
is the
height found in the previous iteration.
2.2.3 Air density
Air density is calculated as

i
=
p
i
M
RT
i
[kg/m
3
] (2)
where
p
i
is the pressure [Pa] as found in (1)
M is the molar mass of dry air (0.0289644 kg/mol)
R is the gas constant (8.31447
J
mol K
), and
T
i
is the temperature [K].
2.2.4 Air resistance
The air resistance is calculated as
F
a,i
=
1
2

i
C
d
Av
2
i1
[N] (3)
where

i
is the air density, from (2),
C
d
is the drag coecient of the rocket, set to a static value of 0.65,
A is the frontal area of the rocket [m
2
], and
v
i1
is the speed of the rocket [m/s] as found in the previous iteration.
3
2.3 Mass
Rocket mass is calculated as
m
i
= m
r
+ max(m
f
v
m
t, 0) [kg] (4)
where
m
r
is the initial total mass of the rocket and engine, excluding fuel
m
f
is the total fuel mass
v
m
is the fuel loss rate [kg/s], initially calculated as
v
m
=
m
f
t
b
[kg] (5)
where t
b
is the burn time. Fuel loss rate is thus assumed to be constant.
2.4 Gravitational force
The gravitational force acting upon the rocket is calculated as
F
g,i
= m
i
g
i
[N] (6)
where the gravitational acceleration g
i
is calculated as
g
i
=

(r + h
i1
)
2
[m/s
2
] (7)
where
is the gravitational parameter of the Earth (3.986004418 10
14
m
3
/s
2
),
h
i1
is the height of the rocket found in the previous iteration, and
r is the distance from the center of the Earth to the rockets initial position, given by
r =

(E
2
a
cos )
2
+ (E
2
b
sin )
2
(E
a
cos )
2
+ (E
b
sin )
2
[m] (8)
where
E
a
= 6 378.137 km is the Earths equatorial radius,
E
b
= 6 356.7523 km is the Earths polar radius, and
is the geodetic latitude of the launch site.
2.5 Total force
The total force acting upon the rocket is calculated as
F
t,i
= F
e,i
+ F
g,i
+ F
a,i
[N] (9)
4
2.6 Acceleration
The acceleration experienced by the rocket is then calculated as
a
i
=
F
t,i
m
i
[m/s
2
] (10)
2.7 Velocity
The velocity of by the rocket is then calculated as
v
i
= v
i1
+ a
i
dt [m/s] (11)
2.8 Altitude
The height of by the rocket is then calculated as
h
i
= h
i1
+
v
i1
+ v
i
2
dt [m] (12)
3 Two-dimensional ight
For two-dimensional simulation of rocket ight, velocity is given by
v
i
= v
i1
+adt
=

v
x,i1
v
y,i1

a
x,i
a
y,i

dt (13)
4 Parachute Timing
4.1 General Considerations
Parachutes should not deploy before there is a tangible atmosphere with
which they can react. In order to minimize possible landing area, parachutes
should be deployed as late as possible. On the other hand, they should
be deployed soon enough to allow for a comfortable safety margin. If the
spacecraft is equipped with ejection seat(s), this margin should be large
enough to allow the astronaut to assert the situation and make the decision
to eject in due time.
Another consideration is speed. Parachutes should not be deployed at speeds
greater than a certain maximum threshold. The spacecraft shall be desig-
ned for a maximum free-fall speed no larger than the maximum parachute
deployment speed plus a safety margin.
The drogue parachute(s) shall slow the spacecraft to below the maximum
threshold of the main parachute(s).
5
4.2 Findings
Using the available data, the following parachute deployment data is found:
t/[s] v/[m/s] a/[g] Event
620 45.7 2.0 Drogue chute deploy
660 24.7 9.0 Main chute deploy
791 7.3 1.7 Splashdown
5 Todos
This section contains a list of features that should be developed, in no
particular order. A wish list, so to speak.
Support for external engine thrust data les
2D support for calculating downrange distance
Support for wind proles
Support for varying C
d
with velocity
Better modeling of transsonic and supersonic speeds
Rocket stability determination
Variable mass loss rate
More detailed model of air pressure variation with altitude
3D support for calculating rockets complete path through space
Eects of Earths rotation
More detailed geoid model
Porting of this program to a platform-independent language
GUI
A Rocket Stability Checklist
(From http://www.apogeerockets.com/Education/newsletter42.asp)
1. Crooked or canted ns. If you have ns on your rocket that are not
perfectly straight, they have the potential to cause unexpected lift
forces to be generated.
6
2. Fins where the airfoils are dierent. If each n on your rocket has a
dierent airfoil, this would have the same eect as a crooked or canted
ns. It generates non-uniform lift forces. The best airfoil on all the ns
would be the teardrop shape (symmetrical). But if it isnt uniform on
both sides, you have what is called a cambered airfoil. This is the
same type of airfoil that is on the wing of an airplane; whos purpose
is specically to generate lift.
3. Forward ns. These are any ns placed on the model in front of the
Center-of-Gravity (CG). They are always destabilizing if they generate
lift. So it is critically important that they be made as small as possible,
and that they be perfectly straight on the model. If they arent, the
model is probably going to be unstable.
4. Asymmetrical n arrangements. The word asymmetrical means not
symmetrical in other words ns that are not placed or spaced equal
distances around the tube. It would also include having some ns on
the rocket being bigger than others. In either case, what happens is
that the lift force on one side of the rocket can be bigger than on the
other side. This can cause the model to do loops if it is hit by a sudden
gust of wind (on the wrong side of the model).
5. Fins that pop o during ight. When this happens, the result is that
the lift forces around the rocket are not uniform. This makes the rocket
do loops. And it is pretty easy to gure out after the ight if youre
fortunate to nd the parts afterward.
6. Loose ns. Even if the ns dont pop o during ight, the reason we
dont tolerate loose ns on the rocket is that they can vibrate back
and forth. This disrupts the airow on one side of the model, and can
cause it to go unstable. So never tape a n onto a rocket, or permit
someone else to do so. This is just asking for trouble.
7. Fin utter. This condition is a lot like loose ns. But the dierence is
that the root edge of the n is securely attached. Typically, it is caused
by ns that are made from very thin material, or material that can
ex. During ight, the ns twist. When this happens, the n tip is at
an angle of attack. That generates lift, and can cause the model to go
unstable. If you ever hear of a rocket that buzzes as it goes up into the
air; this is n utter.
8. Protrusions on the side of the rocket that act like ns. It doesnt have
to look like a n to act like one. Anything on the side of the rocket
body tube can generate lift or drag forces when the model is at an
angle of attack. It may be a canopy on a model that looks like a jet; or
maybe a parasite glider that is there just to be boosted into the air.
7
These forces are very dicult to predict; which is the main reason that
RockSim does not allow things to be strapped to the sides of the tube.
9. Parachute that isnt fully inserted into the rocket, and utters along-
side of the rocket. This disrupts the smooth ow of air over the ns.
Ive seen this happen to a lot of younger modelers rockets.
10. Loose nose cones that are canted in the tube. This is similar to the
number 8 above, because a canted nose cone can generate more lift
forces on one side of the rocket than the other.
11. CG shift during ight. This can happen on really light high perfor-
mance models. On these models, the parachute sliding rearward in the
tube can be enough to move the CG behind the Center-of-Pressure.
This has happened to me several times during competition.
12. Air being ducted through the rocket, and out one side. This one isnt
common, except for rockets that have jet intake scoops. You just have
to watch out for the direction of the air coming out the back.
13. Putting the rocket high on the launch rod, instead of near the bottom.
I see this one all the time. Most time it is because the launch rod is
bent near the base, and you are trying to avoid that area. Or maybe
there is crud on the rod. And it happens a lot on gliders, which have
long tails that stand on the launch pad. The result is the same there
isnt enough length of rod for the rocket to travel while it builds up
to that critical ight speed. And as we all know, there is a minimum
speed the rocket must be traveling at before the ns become eective
at keeping it ying straight.
14. Rocket binding on the launch rod. This is similar to the one above.
The rocket hangs up for just a moment; decreasing its speed. When it
lets go, now it isnt traveling fast enough.
15. Getting entangled in the igniter clips; preventing it from lifting o
smoothly. Again, anything that slows the rocket while it is on the rod
may be a detriment to the stability of the ight.
16. Piston launchers come with their own problems. When the rocket is
traveling upward on the piston and it reaches the stop before popping
o is the most common problem. In eect, the rocket comes to rest for
a brief instant, and then pops o the piston assembly. Because it is
basically starting from rest, the ns of the rocket are not eective at
all. So if you are using a piston launcher, just be careful. Better yet,
mount the rocket inside a tower launcher too. That way, when it pops
o the piston, it has the guidance of the tower while the rocket builds
up speed again.
8
17. Insucient thrust level. If your rocket is heavy, and youre using a
low thrust motor, you need to be extra cautious. Run your RockSim
simulations, and see what the lift-o speed is when the rocket clears
the rod. If it isnt up to the minimum lift-o speed; try a motor with
more initial thrust.
18. A strong gust of wind right when the rocket clears the rod. We blame
a lot of unstable ights on this; although it is preventable. Just like
number 17 above, run your RockSim simulations; but with a high wind
speed. See what happens to the ight.
19. A nose cone that pops o in ight. This one is obvious when it occurs,
and usually happens on larger diameter models. And there is a reason
nose cones suddenly pop o during high speed ights. It is because
the internal pressure inside the rocket is greater than the outside air
pressure. So there is a force trying to push the nose cone out of the
tube. It is easily solved with a pressure relief hole.
20. Rod whip. We blame this one for a lot of unstable ights too. If there is
wind, we can see the rod waving about. This movement adds a velocity
component that isnt anticipated. But a lot of times, we use this excuse
when there isnt any wind at all. What we think my be occurring is
that as the rocket is traveling upward, it is somehow exing the rod.
There needs to be some high-speed movies made to actually determine
what is happening, and if this is a legitimate reason for the model to
go unstable.
21. Canted thrust line. This may occur when the motor mount is cocked
inside the rocket body tube. Because the motor is pushing in a dierent
direction than the rocket wants to move; it causes it to go unstable.
22. O-axis thrust lines. This is dierent than a canted thrust line, but
the eect can be the same. One kit in particular uses an o-axis thrust
line. That is the Estes skill level 5 Space Shuttle. The motor is straight
along the length of the model, but is oset from the centerline. This is
because the orbiter is creating lift forces that counteract the o axis
thrust. But if you tried to y the External Tank by itself, it will do
cartwheels across the sky.
23. CG that is o-axis. This one is rare, unless the CG is way o to one
side. It is usually caused by something heavy on one side of the model
(maybe a payload inside the rocket that has a heavy battery against
one side of the tube).
24. Nozzle Erosion. I have a good friend that thinks this is the cause of a
lot of unexplained unstable ights. The theory is that the gases coming
9
out the rocket are eating at the sides of the nozzle. If the erosion of the
nozzle isnt uniform, then youll get a vectored thrust. This would cause
the rocket to go unstable. I think there is a lot of legitimacy in this
theory, particularly for black powder motors that have clay nozzles. I
dont know if it occurs on composite motors that use a phenolic nozzle.
The problem can be alleviated by using a longer launch rod. Because
if the model is traveling fast enough, the ns can cancel out the eect.
25. Side wall failure. This is actually a motor cato. A tiny pin-hole develops
near the base of the motor, which vents hot exhaust gases out one
side. It always leaves physical evidence that can be seen after the ight.
There isnt anything you can do to prevent it, but it can be used to
explain why the ight went unstable.
26. Short Rockets. This is where it gets really tricky. Short rockets are less
dynamically stable. That is, they are more easily disturbed from the
ight path, and they take longer to correct back to straight ight. In
my opinion, there is something (one of the other causes listed above),
that triggers the model onto its course of instability. Figuring out what
that is will be dicult because the design of the model makes it tend to
be less stable to begin with. See the related article in e-zine newsletter
#86.
27. Very long rockets can ex. This means that the nose is ying in one
direction, while the tail and motor are going in a dierent direction.
The NAR super-roc competition can really be fun to watch because of
this fact.
Litteratur
[1] Hans Olaf Toft: 2D Baneberegning for passivt stabile raketter, DARK,
1998.
10