361 lines
17 KiB
TeX
361 lines
17 KiB
TeX
|
||
|
||
\chapter{Basics of model rocket flight}
|
||
\label{chap-basics}
|
||
|
||
|
||
As rockets and rocket motors come in a huge variety of shapes and
|
||
sizes, different categories are defined for different levels of
|
||
rocketry. {\it Model rocketry} itself is governed by the NAR Model
|
||
Rocket Safety Code~\cite{nar-safety-code} in the U.S. and other
|
||
similar regulations in other countries. The safety code requires that
|
||
the model rockets be constructed only of light-weight materials
|
||
without any metal structural parts, and have a maximum lift-off weight
|
||
of 1.5~kg. They may only used pre-manufactured motors of classes A--G
|
||
(see Section~\ref{sec-motor-classes} for the classification).
|
||
|
||
{\it High power rocketry} (HPR) is basically scaled up model
|
||
rocketry. There are no weight restrictions, and they can use
|
||
pre-manufactured solid or hybrid rocket motors in the range of H--O.
|
||
The combined total impulse of all motors may not exceed 81\s920~Ns.
|
||
|
||
{\it Experimental} or {\it amateur rocketry} includes any rocketry
|
||
activities beyond model and high power rocketry. This may include
|
||
for example using motor combinations that exceed the limits placed by
|
||
high power rocketry, building self-made motors or utilizing liquid
|
||
fueled motors. Finally there is {\it professional rocketry} which is
|
||
conducted for profit, usually by governments or large corporations.
|
||
|
||
Even though rockets come in many different sizes, the same principles
|
||
apply to all of them. In this thesis the emphasis will be on model
|
||
rocketry, but the results are just as valid for larger rockets as long
|
||
as the assumptions of for example the speed range remain valid. In
|
||
this chapter the basics of model rocketry and differences to high
|
||
power rocketry are explained.
|
||
|
||
|
||
\section{Model rocket flight}
|
||
|
||
A typical flight of a model rocket can be characterized by the four
|
||
phases depicted in Figure~\ref{fig-model-flight}:
|
||
%
|
||
\begin{enumerate}
|
||
\item Launch: The model rocket is launched from a vertical launch
|
||
guide.
|
||
\item Powered flight: The motor accelerates the rocket during the
|
||
powered flight period.
|
||
\item Coasting flight: The rocket coasts freely until approximately
|
||
at its apogee.
|
||
\item Recovery: The recovery device opens and the rocket descends
|
||
slowly to the ground.
|
||
\end{enumerate}
|
||
|
||
\begin{figure}
|
||
\centering
|
||
\epsfig{file=figures/model-flight,scale=0.8}
|
||
\caption{The basic phases of a typical model rocket flight:
|
||
1.~Launch, 2.~Powered flight, 3.~Coasting and 4.~Recovery.}
|
||
\label{fig-model-flight}
|
||
\end{figure}
|
||
|
||
Model rockets are launched from a vertical launch guide that keeps the
|
||
rocket in an upright position until it has sufficient velocity for the
|
||
fins to aerodynamically stabilize the flight. The NAR safety code
|
||
forbids launching a model rocket at an angle greater than
|
||
$30^\circ$ from vertical. A typical launch guide for small rockets is
|
||
a metal rod about 3-5~mm in diameter, and the launch lug is a short
|
||
piece of plastic tube glued to the body tube. Especially in
|
||
larger rockets this may be replaced by two extruding bolts, the ends
|
||
of which slide along a metal rail. Use of a launch lug can be avoided
|
||
by a tower launcher, which has 3--4 metal bars around the rocket
|
||
that hold it in an upright position.
|
||
|
||
After clearing the launch guide, the rocket is in free, powered flight.
|
||
During this phase the motor accelerates the rocket while it is
|
||
aerodynamically stabilized to keep its vertical orientation. When the
|
||
propellant has been used, the rocket is typically at its maximum
|
||
velocity. It then coasts freely for a short period while the motor
|
||
produces smoke to help follow the rocket, but provides no
|
||
additional thrust. Finally, at approximately the point of apogee, a
|
||
small pyrotechnical ejection charge is fired upwards from the motor
|
||
which pressurizes the model rocket and opens the recovery device.
|
||
|
||
High-power rocket motors usually have no ejection charges incorporated
|
||
in them. Instead, the rocket carries a small flight computer that
|
||
measures the acceleration of the rocket or the outside pressure change
|
||
to detect the point of apogee and to open the recovery device.
|
||
Frequently only a small drogue parachute is opened at apogee, and the
|
||
main parachute is opened at some pre-defined lower altitude around
|
||
100--300 meters.
|
||
|
||
The typical recovery device of a model rocket is either a parachute or
|
||
a {\it streamer}. The parachutes are usually a simple planar circle
|
||
of plastic or fabric with 4--10 shroud lines attached. A streamer is
|
||
a strip of plastic or fabric connected to the rocket, intended to
|
||
flutter in the air and thus slow down the descent of the rocket.
|
||
Especially small rockets often use streamers as their recovery device,
|
||
since even light wind can cause a light-weight rocket with a
|
||
parachute to drift a significant distance.
|
||
|
||
|
||
|
||
|
||
\section{Rocket motor classification}
|
||
\label{sec-motors}
|
||
\label{sec-motor-classes}
|
||
|
||
The motors used in model and high power rocketry are categorized based
|
||
on their total impulse. A class `A' motor may have a total impulse in
|
||
the range of 1.26--2.50~Ns. Every consecutive class doubles the
|
||
allowed total impulse of the motor. Thus, a B-motor can have an
|
||
impulse in the range 2.51--5.00~Ns and a C-motor in the range
|
||
5.01--10.0~Ns. There are also classes \half A and \quarter A which
|
||
have impulse ranges half and one quarter of those of an A-motor,
|
||
respectively. Commercial rocket motors are available up to
|
||
class~O with a total impulse of 30\s000~Ns~\cite{all-certified-motors}.
|
||
Table~\ref{tab-motor-classes} lists the impulse ranges for model
|
||
and high-power rocket motors.
|
||
|
||
\begin{table}
|
||
\caption{Total impulse ranges for motor classes \quarter A--O.}
|
||
\label{tab-motor-classes}
|
||
\begin{center}
|
||
\begin{tabular}{cr@{--}l|cr@{--}l|cr@{--}l}
|
||
\hline
|
||
\quarter A & 0.0 & 0.625~Ns & E & 20.01 & 40.0~Ns & K & 1280.01 & 2560~Ns \\
|
||
\half A & 0.626 & 1.25~Ns & F & 40.01 & 80.0~Ns & L & 2560.01 & 5120~Ns \\
|
||
A & 1.26 & 2.50~Ns & G & 80.01 & 160~Ns & M & 5120.01 & 10240~Ns \\
|
||
B & 2.51 & 5.00~Ns & H & 160.01 & 320~Ns & N & 10240.01 & 20480~Ns \\
|
||
C & 5.01 & 10.0~Ns & I & 320.01 & 640~Ns & O & 20480.01 & 40960~Ns \\
|
||
D & 10.01 & 20.0~Ns & J & 640.01 & 1280~Ns & \\
|
||
\hline
|
||
\end{tabular}
|
||
\end{center}
|
||
\end{table}
|
||
|
||
Another important parameter of a rocket motor is the thrust given by
|
||
the motor. This defines the mass that may be lifted by the motor and
|
||
the acceleration achieved. Small model rocket motors typically have
|
||
an average thrust of about 3--10~N, while high-power rocket motors can
|
||
have thrusts in excess of 5\s000~N.
|
||
|
||
The third parameter used to classify a model rocket motor is the
|
||
length of the delay between the motor burnout and the ignition of the
|
||
ejection charge. Since the maximum velocity of different rockets
|
||
using the same type of motor can be vastly different, also the length
|
||
of the coasting phase varies. Therefore motors with otherwise the
|
||
same specifications are often manufactured with several different
|
||
delay lengths. These delay lengths do not apply to high-power rocket
|
||
motors, since they do not have ejections charges incorporated in them.
|
||
|
||
Model rocket motors are given a classification code based on these
|
||
three parameters, for example ``D7-3''. The letter specifies the
|
||
total impulse range of the motor, while the first number specifies the
|
||
average thrust in Newtons and the second number the delay of the
|
||
ejection charge in seconds. The delay number can also be replaced by
|
||
`P', which stands for {\it plugged}, \ie the motor does not have an
|
||
ejection charge. Some manufacturers may also use an additional letter
|
||
at the end of the classification code specifying the
|
||
propellant type used in the motor.
|
||
|
||
Even motors with the same classification code may have slight
|
||
variations to them. First, the classification only specifies the
|
||
impulse range of the motor, not the exact impulse. In principle, a
|
||
D-motor in the lower end of the range might have a total impulse only
|
||
1~Ns larger than a C-motor in the upper end of its range. Second,
|
||
the code only specifies the average thrust of the motor. The thrust
|
||
rarely is constant, but varies with time.
|
||
Figure~\ref{fig-thrust-curve} shows the typical thrust curve of a
|
||
small black powder rocket motor. The motors typically have a short
|
||
thrust peak at ignition that gives the rocket an initial acceleration
|
||
boost before stabilizing to a thrust level a little below the average
|
||
thrust. Statically measured thrust curves of most commercial rocket
|
||
motors are readily available on the
|
||
Internet~\cite{thrust-curve-database}.
|
||
|
||
\begin{figure}
|
||
\centering
|
||
\epsfig{file=figures/motors/D12-thrustcurve,width=9cm}
|
||
\caption{A typical thrust curve of an Estes D12-3 rocket motor and
|
||
its average thrust.~\cite{D12-curve}}
|
||
\label{fig-thrust-curve}
|
||
\end{figure}
|
||
|
||
Also the propellant type may affect the characteristics of the motor.
|
||
Most model rocket motors are made up of a solid, pyrotechnical
|
||
propellant---typically black powder---that is cast into a suitable
|
||
shape and ignited on launch. Since the propellant burns on its
|
||
surface, different thrust curves can be achieved by different mold
|
||
shapes.
|
||
|
||
% vesiraketit!
|
||
|
||
A significantly different motor type, {\it hybrid motors}, were
|
||
commercially introduced in 1995. These motors typically include the
|
||
propellant and oxidizer in different states, typically a composite
|
||
plastic as the fuel and a separate tank of liquid nitrous oxide
|
||
($\rm N_2O$) as the oxidizer. The plastic on its own does not
|
||
burn very well, but provides ample thrust when the nitrous oxide is
|
||
fed through its core. The nitrous oxide tank is
|
||
self-pressurized by its natural vapor pressure. However, since
|
||
temperature greatly affects the vapor pressure of nitrous oxide, the
|
||
thrust of a hybrid motor is also diminished if the oxidizer is cold.
|
||
On the other hand, the motor will burn longer in this case, and since
|
||
nitrous oxide is denser when cold, the motor may even yield a greater
|
||
total impulse.
|
||
|
||
The significance of this effect was observed when analyzing the video
|
||
footage of the launch of the first Finnish hybrid rocket,
|
||
``Haisun<75><6E>t<EFBFBD>''~\cite{haisunaata-launch}. The average thrust during the
|
||
first 0.5~seconds was determined to be only about 70~N, whereas the
|
||
static tests suggest the thrust should have been over 200~N.
|
||
Instead, the motor burned for over 10~seconds, while the normal thrust
|
||
curves indicate a burning time of 5--6~seconds. This shows that the
|
||
temperature of the hybrid motor oxidizer can have a dramatic effect on
|
||
the thrust given by the motor, and the static test curve should be
|
||
assumed to be valid only in similar operating conditions as during the
|
||
test.
|
||
|
||
One further non-pyrotechnical rocket type is {\it water rockets}.
|
||
These are especially popular first rockets, as they require no special
|
||
permits and are easy to construct. The water rocket includes a bottle
|
||
or other chamber that has water and pressurized air inside it. On
|
||
launch the pressure forces the water out of a nozzle, providing thrust
|
||
to the rocket. While simulating water rockets is beyond the scope of
|
||
this thesis, it is the aim that methods for modeling water rockets can
|
||
be added to the produced software in the future.
|
||
|
||
|
||
|
||
\section{Clustering and staging}
|
||
|
||
Two common methods used to achieve greater altitudes with model
|
||
rockets are {\it clustering} and {\it staging}. A cluster has two or
|
||
more rocket motors burning concurrently, while staging uses motors
|
||
that burn consecutively. The motor configuration of a cluster and
|
||
staged rocket is depicted in Figure~\ref{fig-cluster-stages}.
|
||
|
||
When a cluster is launched, the total thrust is the sum of the thrust
|
||
curves of the separate motors. This allows greater acceleration and
|
||
a greater liftoff weight. Staging is usually performed by using
|
||
zero-delay motors, that ignite the ejection charge immediately at
|
||
burnout. The ejection charge fires towards the upper stage motor and
|
||
ignites the next motor. High power motors with no ejection charges
|
||
can be clustered by using an onboard accelerometer or timer that
|
||
ignites the subsequent stages. Staging provides a longer duration of
|
||
powered flight, thus increasing the altitude.
|
||
|
||
|
||
\begin{figure}
|
||
\centering
|
||
\parbox{65mm}{\centering
|
||
\epsfig{file=figures/motors/cluster,width=60mm} \\ (a)}
|
||
\hspace{10mm}
|
||
\parbox{40mm}{\centering
|
||
\epsfig{file=figures/motors/staged,width=30mm} \\ (b)}
|
||
\caption{The motor configuration for (a) a cluster rocket and (b) a
|
||
two-staged rocket.}
|
||
\label{fig-cluster-stages}
|
||
\end{figure}
|
||
|
||
Clustering provides a greater acceleration at launch, but staging
|
||
typically provides greater altitude than a cluster with similar
|
||
motors. This is because a clustered rocket accelerates quickly to a
|
||
greater speed thus also increasing the aerodynamic drag. A staged
|
||
rocket has a smaller thrust for a longer period of time, which reduces
|
||
the overall effect of drag during the flight.
|
||
|
||
|
||
\section{Stability of a rocket}
|
||
\label{sec-stability}
|
||
|
||
When designing a new rocket, its stability is of paramount
|
||
importance. A small gust of wind or some other disturbance may cause
|
||
the rocket to tilt slightly from its current orientation. When this
|
||
occurs, the rocket centerline is no longer parallel to the
|
||
velocity of the rocket. This condition is called flying at an
|
||
{\it angle of attack $\alpha$}, where $\alpha$ is the angle between
|
||
the rocket centerline and the velocity vector.
|
||
|
||
When a stable rocket flies at an angle of attack, its fins produce a
|
||
moment to correct the rocket's flight. The corrective moment is
|
||
produced by the aerodynamic forces perpendicular to the axis of the
|
||
rocket. Each component of the rocket can be seen as producing a
|
||
separate normal force component originating from the component's CP,
|
||
as depicted in Figure~\ref{fig-normal-forces}.
|
||
|
||
\begin{figure}
|
||
\centering
|
||
\epsfig{file=figures/aerodynamics/component-normal-forces,width=130mm}
|
||
\caption{Normal forces produced by the rocket components.}
|
||
\label{fig-normal-forces}
|
||
\end{figure}
|
||
|
||
The effect of the separate normal forces can be combined into a single
|
||
force, the magnitude of which is the sum of the separate forces and
|
||
which effects the same moment as the separate forces. The point on
|
||
which the total force acts is defined as the center of pressure or the
|
||
rocket. As
|
||
can be seen from Figure~\ref{fig-normal-forces}, the moment produced
|
||
attempts to correct the rocket's flight only if the CP is located aft
|
||
of the CG. If this condition holds, the rocket is said to be
|
||
{\it statically stable}. A statically stable rocket always produces a
|
||
corrective moment when flying at a small angle of attack.
|
||
|
||
The argument for static stability above may fail in two conditions:
|
||
First, the normal forces might cancel each other out exactly, in which
|
||
case a moment would be produced but with zero total force. Second,
|
||
the normal force at the CP might be in the wrong direction (downward
|
||
in the figure), yielding an uncorrective moment. However, we shall
|
||
see that the only component to produce a downward force is a boattail,
|
||
and the force is equivalent to the corresponding broadening of the
|
||
body. Therefore the total force acting on the rocket cannot be zero
|
||
nor in a direction to produce an uncorrective moment when aft of the
|
||
CG.
|
||
|
||
The {\it stability margin} of a rocket is defined as the distance between
|
||
the CP and CG, measured in {\it calibers}, where one caliber is the
|
||
maximum body diameter of the rocket. A rule of thumb among model
|
||
rocketeers is that the CP should be approximately 1--2 calibers aft of
|
||
the CG. However, the CP of a rocket typically moves upwards as the
|
||
angle of attack increases. In some cases, a 1--2 caliber stability
|
||
margin may totally disappear at an angle of attack of only a few
|
||
degrees. As side wind is the primary cause of angles of attack, this
|
||
effect is called
|
||
{\it wind caused instability}~\cite{galejs}.
|
||
|
||
Another stability issue concerning rocketeers is the
|
||
{\it dynamic stability} of a rocket. A rocket that is statically
|
||
stable may still be poor at returning the rocket to the original
|
||
orientation quickly enough. Model rockets may encounter several types of
|
||
dynamic instability depending on their shape, size and
|
||
mass~\cite[pp.~140--141]{stine}:
|
||
%
|
||
\begin{enumerate}
|
||
\item {\it Too little oscillation damping.} In short, light-weight
|
||
rockets the corrective moment may significantly over-correct the
|
||
perturbation, requiring a corrective moment in the opposite
|
||
direction. This may lead to continuous oscillation during the
|
||
flight.
|
||
\item {\it Too small corrective moment.} This is the case of over-damped
|
||
oscillation, where the corrective moment is too small compared to
|
||
the moment of inertia of the rocket. Before the rocket has been
|
||
able to correct its orientation, the thrust of the motors may have
|
||
already significantly affected the direction of flight.
|
||
\item {\it Roll-pitch coupling.} If the model has a natural roll
|
||
frequency (caused \eg by canting the fins) close to the oscillation
|
||
frequency of the rocket, roll-pitch resonance may occur and cause
|
||
the model to go unstable.
|
||
\end{enumerate}
|
||
|
||
By definition, dynamic stability issues are such that they occur over
|
||
time during the flight of the rocket. A full flight simulation that
|
||
takes into account all corrective moments automatically also simulates
|
||
the possible dynamic stability problems. Therefore the dynamic
|
||
stability of rockets will not be further considered in this
|
||
thesis. For an analytical consideration of the problem, refer to
|
||
\cite{advanced-model-rocketry}.
|
||
|
||
|
||
|