Once we learned the basic operation of the CVR, we had to determine how
it records a signal. Analyzing
a signal can be thought of as a puzzle that must be solved in reverse, as the
engineer is first given the final result and then works to discover the size and
shape of each piece. The sources of
the signal represent the individual pieces of the puzzle, and to properly
analyze the signal it is necessary to know the number and behavior of the pieces
that make up the signal. Only after
all the pieces of the puzzle are in place, can the contribution of each piece be
determined and a proper analysis performed.
To show that the analysis technique used in this paper is valid, we start
in this section by building the puzzle, piece by piece, or in this case, source
by source. Aside from the
microphones each wire is attached to, there are two main sources of input to the
CVR: electro-magnetic fields and the triboelectric effect [10].
The
relationship between a current carrying wire and magnetic fields has been known
since Hans Christian Oersted made the discovery in 1820 [11].
Oersted noted that any electric current causes a magnetic field.
The magnetic field lines around a straight current carrying wire are a
set of concentric circles around the wire, running the length of the wire [12].
The direction of the field is based on the right hand rule, demonstrated
by grasping the wire with the right hand so that the thumb is in the
direction of the current and the direction
of the field is the same as the direction of the fingers [12].
This is represented in Figure 2.
Figure 2: Magnetic Field around Current Carrying Wire [11]
Equation 1 can be
implemented to give the strength of the field (B). In Equation 1, I is the current in the wire,
is a constant known as the permeability of free space equal to
, and r is the radial distance from the wire [12].
Equation
1:
Strength of Magnetic Field [12]
Just as current in a wire causes a magnetic field, a magnetic field passing over a wire also induces a current in the wire [12]. The force induced by the magnetic field always acts perpendicular to the velocity of the particle and the magnetic field itself [12]. This can be visualized by pointing the fingers of your right hand in the direction of the velocity of the current, curling your fingers to point in the direction of the magnetic field, and noting that your thumb now points in the direction the force will act. Equation 2 gives the amount of force on a particle of charge q (coulombs) and velocity v (m/s) in the magnetic field B (T) [12].
Equation 2: Force on a Charged Particle in Magnetic Field B [12]
In
this section we have shown that a current flowing through a conductor produces a
magnetic field around that conductor, and that a magnetic field also produces a
current in any conductor it passes through.
Any electromagnetic field passing through the wires of the cockpit voice
recorder would generate a current that would be recorded equally on all four
tracks.
Starting
power to the engines on the Beech 1900C is provided through batteries, but once
the plane is started, the electrical power is supplied through two generators
located in the nacelles of the plane [13].
Gas turbine engines power the propellers, which in turn spin the magnets
in each generator [13]. The faster
the propeller spins, the more current the generator can output.
Any change in the power to the propeller would result in a change in the
current produced by the generator, resulting in a change in the magnetic field
produced by the generator.
Mounted in the cockpit of every plane is a voltage regulator meter [13]. The amount of voltage the generator is allowed to hold is limited to a set value depending on the needs of the specific plane [13]. If the power to an airplane engine were suddenly increased, the increased voltage would be shed through the ground into the frame of the plane. The voltage bleed-off would be recorded as it affected the current in the CVR wires.
The wires running within the insulation also act as capacitors, holding a charge and creating a magnetic field around each wire [10]. If the wire were to be crimped or pinched, the relative displacement in the wires would cause an interaction between the magnetic fields, creating a current that is visible to the CVR [10].
Consider a sound wave passing through and around a wire. A sound wave is simply a pressure wave traveling through air. As the pressure wave passes over the wires of the CVR, it is possible that the pressure change would be enough to induce a voltage similar to that seen when the wire is pinched.
The appearance of a magnetic field in the signal recorded on the CVR would differ from that of the mechanical effect. Even if the field around the wire were to change suddenly, the change in the current through the wires would represent a steady increase due to the capacitating effects of the wires. The current would also slowly return too normal as the field around the wire leveled off. Field changes resulting from the increasing and decreasing magnetic field around the generator would also maintain the same frequency pattern regardless of the magnitude of the change. For these reasons, we should be able to separate magnetic variations from mechanical events such as the triboelectric effect. The linear property of the strength of the field given in Equation 1 shows that any change in the power of the engine would result in a linear change in the current of the CVR.
The triboelectric effect is mechanically induced input generated by movement in the cable’s components [10]. The wires running from the CVR in the tail of the plane to the microphones at the front of the plane are composed of dozens of small wires wound around each other, and covered in a protective coating. As the wire is bent, the coating and the wires themselves rub against each other as in Figure 3 [10]. It should be noted that the inner conductor in Figure 3 is not one single wire, but several wires twisted together. As the insulation rubs against the inner wires, it rubs off electrons, creating a static charge similar to that produced by someone rubbing their feet across a carpeted floor [12].
Figure 3: Triboelectric Effect [7]
The term “tribo” means “rubbing”. In order to generate a triboelectric effect, the inner wires and outer coating must rub against each other. For this to happen, the wire must either be bent or pulled [10]. The signal that manifests itself on the CVR as a result of the triboelectric effect would appear as a series of spike like transients resulting from the intermittent behavior of the rubbing of the wires.
The impedance of the circuit in which the triboelectric effect is present is very important. Impedance is the AC (alternating current) version of DC (direct current) resistance, and high source impedances accentuate the triboelectric effect [10]. The impedance in a wire increases as the length of the wire increases, with the highest possible impedance resulting from an open circuit such as when a cable is plugged into an amplifier with no instrument at the sending end [10]. A similar situation exists in the four wires of the CVR. The wires running from the CVR to the pilot, copilot, and cockpit area microphone act as closed circuits with the end microphones acting as sources. While the public address system is inoperative, the silent track acts as a virtual open circuit, making it very susceptible to triboelectric effects. For this reason, we expect the triboelectric effect to be more predominantly on the silent track.
While useful for this project, the triboelectric effect is a problem in many fields, such as the music industry. In an effort to reduce the triboelectric effect, wires are often routed in twisted pairs [10]. When a twisted pair is bent, the voltages created in the wires are mirror images of each other [10]. This mirroring lowers the overall amplitude of the charge created by the triboelectric effect [10]. This effect is called the common-mode rejection ratio (CMRR), and is improved by increasing the number of twists in the pair [10].
The twisted pair does have a drawback.
The two wire conductors wrapped around each other form a “loop
antenna” that amplifies stray magnetic fields passing through the wires [10].
The further apart the wires and the more wires twisted together, the
higher the amplification of the magnetic field [10].