High-Intensity Radiated Fields
A changing electric field produces an induced magnetic field. Similarly, a changing magnetic field produces an induced electric field. This relationship between electric and magnetic fields causes the propagation of transverse electromagnetic waves, in which oscillating electric and magnetic fields are oriented perpendicular to each other and also to the direction of propagation. Electromagnetic waves are radiated whenever charged particles are accelerated. RF electromagnetic waves, with frequencies extending from approximately 10 kHz to 100 GHz, may be produced by electrically oscillating free electrons back and forth within a conducting material. Higher-frequency electromagnetic waves, such as infrared, ultraviolet, visible light, X-rays and Gamma rays result from thermal excitation of orbital electrons (heat) or quantum state changes at the atomic and nuclear level. The earth is continuously exposed to electromagnetic energy from the sun, which imparts approximately 1 400 W of power per square metre (W/m2) to the upper atmosphere, attenuated to about 220 W/m2 at sea level.
Electromagnetic waves in the RF spectrum are used to convey analogue and digital information. This information is typically encoded by modulating the frequency, amplitude, or phase characteristics of the carrier wave. Some RF signals, such as those normally produced by radar systems, are pulsed to facilitate the measurement of distance between the transmitter and the target. Modern aircraft radiate and receive RF signals in the atmosphere surrounding the aircraft and through electrically conductive RF cabling within the aircraft. In addition, the electrical potential difference (voltage) between wire pairs is used to support binary communication protocols between various electronic systems.
The electric field component of an electromagnetic wave exerts a force on charged particles and, particularly in electrically conductive material, may cause a net flow of free electrons along the field gradient. This mechanism facilitates the coupling of external electromagnetic energy, occurring in free space, to electrical cables within an aircraft structure. When modulated RF waveforms are being used to convey information along these cables, coupling can distort the original waveform and disrupt the normal functions of electrical and avionics systems. This is especially true in highly integrated systems, where individual LRUs are interconnected by shared data and power distribution buses. These buses provide common pathways through which coupled electromagnetic energy may concurrently influence multiple discrete electronic devices. Disruptions may also occur within the flight recorders, or LRUs, when free space electromagnetic radiation passes through an opening (aperture) in the LRU case and impinges on the junctions of semiconductor components or couples directly to the conductive paths on printed circuit boards. The reflective cavity defined by the metal housing of a typical LRU may exacerbate this effect by producing standing waves with localized field gradients that exceed the maximum field gradient of the incident radiation.
In addition to the functional disruptions discussed above, other consequences may result from the exposure of an aircraft to extreme electromagnetic field gradients, such as those associated with lightning activity. Under these conditions, the electrical field strength in some region of an aircraft may become sufficiently great to induce an electrical discharge between narrowly separated conductors, by stripping electrons from the substance that occupies the gap between the conductors (ionization). Physical damage to electrical components, pyrolysis, or both of surrounding materials may occur. As well, the thermal energy released by an electrical spark, or a sustained electrical arc, may be sufficient to ignite nearby flammable materials.
The electric field strength required to induce an electrical discharge between proximate conductors is influenced by the dielectric properties of the substance in the intervening gap, the shape and width of the gap, the shape and texture of the conducting surfaces, pressure altitude, and to a lesser extent, ambient temperature. The breakdown voltage for gaps involving pointed or irregularly shaped electrodes may also vary with frequency, although reliable data for electromagnetic waveforms above 100 MHz is not widely available.
In general, the breakdown voltage between conductors, separated by a gas, is a non-linear function of the product of gas density and electrode separation. When the product of gas density and electrode separation is less than approximately 1 000 torr cm, corresponding to a maximum gap distance of roughly 1 cm at one atmosphere, the breakdown voltage function is approximated by the following, empirically derived expression:
Vbreakdown = B x p x d / ( C + ln( p x d ) )
For air, B = 365 Vcm1 torr1, C = 1.18, p = pressure (torr), d = gap distance (cm)
This function yields a minimum breakdown voltage in air of 327 V, which occurs at 0.567 torr cm. Taking sea level pressure to be 760 torr, and the potential difference between electrodes to be 327 V, an electrical discharge would not occur until the gap separation was reduced to 7.9 x 104 cm. A more realistic gap spacing of 0.05 cm, at sea level pressure, gives a breakdown voltage of about 2.6 kV, corresponding to a field gradient of 52 kV/cm. Generally speaking, localized ionization and sparking between proximate conductors, at sea level, is unlikely to occur until the field gradient around the conductors exceeds about 31 kV/cm, irrespective of conductor geometry and surface characteristics.
Within the pressurized area of a commercial aircraft, the pressure altitude rarely exceeds 8 000 feet. Because the relationship between gap breakdown voltage and pressure is approximately linear, the corresponding minimum breakdown voltage at 8 000 feet (565 torr) is approximately 243 V and the breakdown voltage for a 0.05 cm gap is approximately 1.93 kV. The minimum field gradient necessary to produce localized ionization and sparking between proximate conductors is about 23 kV/cm.
When considering the potential consequences of an electrical discharge occurring between exposed conductors, a reasonable worst-case scenario would be when the gap between the conductors is occupied by a flammable fuel-air mixture. Theoretical and experimental results suggest that a minimum energy of approximately 0.2 mJ must be imparted to the conductors to induce ignition. However, the corresponding field strength and energy density in free space must be higher, since inherent in the process of electromagnetic coupling is a loss of energy.
For electromagnetic energy originating from sources external to an aircraft, the amount of energy coupled to internal components is attenuated, in a two-step process, by several orders of magnitude. Attenuation first occurs at the metal skin of a conventional aircraft, which reflects some of the external electromagnetic energy back into the atmosphere. The amount of energy that penetrates to the interior of an aircraft, primarily through the windows and doors, is significantly reduced. In transport aircraft, for example, the ratio between external and internal electromagnetic field strengths ranges from 2 to 40, depending on the location within the aircraft and the RF. Finally, the residual electromagnetic energy that penetrates to the internal spaces of an aircraft may subsequently couple to electrically conductive material, such as wiring or circuit board traces within an LRU, in which case further attenuation will occur at the outer boundary of the material.
Extraneous electromagnetic energy can originate within an aircraft, from inadequately shielded aircraft components or personal electronic devices, and from sources external to the aircraft, such as lightning and military or civilian radar systems. By convention, internal sources are referred to as EMI and external sources are referred to as HIRF.
The HIRF frequency domain has been defined by the Society of Automotive Engineers AE4R committee to extend from 10 kHz to 18 GHz. The lower portion of this spectrum, extending from 10 kHz to approximately 400 MHz, is dominated by weakly directional, continuously transmitting radio and television broadcast signals, exhibiting peak and average power levels that are similar in magnitude. Above 400 MHz, the HIRF frequency spectrum is principally occupied by highly directional radar systems and satellite command and control links. These systems are capable of producing peak and average power levels that are significantly higher than those produced by emitters operating below 400 MHz.
For load conditions typical of wire runs within commercial aircraft, electromagnetic coupling to aircraft wiring is most efficient at frequencies between 1 and 400 MHz. As well, aircraft windows provide less RF shielding at frequencies greater than approximately 30 MHz. For these reasons, HIRF frequencies in the kHz and GHz range must be one or two orders of magnitude greater than a HIRF signal in the 30 to 400 MHz range to achieve the same amount of coupling onto wire runs within an aircraft. In general, electromagnetic coupling to aircraft wiring and conduction into the avionics LRUs is less efficient at frequencies above 1 GHz.
The electronic circuit susceptibility within aircraft electrical and electronic LRUs also varies with frequency. At high frequencies, in the GHz range, the electronic circuits tend to have low-pass characteristics, and reject the energy in this range. In addition, rectification of the higher frequency RF energy is not as efficient as at lower frequencies. The RF coupling, penetration, and electronic circuit susceptibility all contribute to the electrical and electronic system vulnerability. These factors tend to result in maximum vulnerability for frequencies that range from 2 MHz to 400 MHz. Below 2 MHz, the RF fields do not couple efficiently to the aircraft and aircraft wiring. Above 400 MHz, the circuits do not efficiently convert the RF energy to cause system effects.
The propensity for an aircraft to be adversely affected by HIRF is influenced by the frequency of the HIRF signal, the susceptibility of the aircraft to HIRF at the same frequency, the amount of shielding provided to internal components by the aircraft's structure and electrical design, and the RF power to which the aircraft is exposed. Since frequency susceptibility and RF shielding are essentially fixed quantities, power is the principal variable in assessing the risk associated with HIRF emissions within a given range of frequencies.
Free Space Radiation
The "far field" region of an antenna begins at a distance equal to twice the square of the maximum antenna dimension divided by wavelength (2d2/l), and it extends to infinity. When the distance between a transmitter and receiver satisfies the far field criteria and the path between them is free of obstructions, the energy is assumed to radiate isotropicallyuniformly from the transmitting antenna in a spherical pattern. For high-powered radar systems capable of generating significant HIRF environments, the far field assumption is generally valid at ranges greater than 1 nm.
The power density (S) at an aircraft within the far field region of a transmitting antenna is equal to the transmitter EIRP divided by the surface area of a sphere centred on the transmitter:
S = (10EIRP/10)/(4pr2) mW/m2
Expressed in decibels, the power density is
S = EIRP 20 log10r 10.99 dBm/m2
when EIRP is expressed in dBm and distance (r) is expressed in metres.
Electric Field Strength
The electric field strength (E) is derived from the power density through the following relationship:
E = (S 120p)½
when E is expressed in V/m, 120p is the impedance of free space in ohms (377 ohms), and S is power density expressed in W/m2.
Peak and Average Power
For a pulsed RF emitter, the average power output is equal to the product of the peak transmitter power (PT), the pulse width (PW), and the pulse repetition frequency. This value can be used to calculate the average power density and the average electric field strength.
The energy density (D) generated by a pulsed emitter in free space is defined as
D = S PW
when D is expressed in J/m2, S is expressed in W/m2, and PW is expressed in seconds.
An isotropic antenna radiates equally in all directions. The gain of an antenna (PG) in a given direction is the ratio of the power density produced by it in that direction to the power density that would be produced by an isotropic antenna. Antenna gain, relative to an isotropic antenna, is frequently expressed in decibels using the unitless designation, dBi (decibels isotropic). Antenna gain is reciprocal in nature; the transmit gain and receive gain are the same. The amount of power captured by an antenna is the product of power density and antenna gain.
The EIRP of an RF emitter is determined by the transmitter output power, the transmit antenna gain, and any losses that may occur between the transmitter and the antenna. Following adjustments for internal losses, the EIRP becomes the product of PT and PG, or the sum of PT and PG when both values are expressed in decibels.
Unlike the metal skin of an aircraft, which tends to reflect HIRF away from an aircraft, window apertures are more transparent to HIRF. The MD-11 has 71 windows on either side of the fuselage, each with an area of about 0.11 m2; the total window area is approximately 7.86 m2 on each side of the aircraft. For external RF radiation that is incident upon the side of the fuselage, the maximum energy that may enter the aircraft through the windows is the product of energy density and the total window area. The MD-11 passenger and cargo doors also tend to behave like electromagnetic apertures, because the doors do not have continuous electrical contact with the door frames around the door perimeter. These doors have a total area of approximately 17.32 m2 on each side of the aircraft, although the equivalent RF aperture size is substantially smaller.
When a travelling wave is reflected back upon itself, the incident and reflected wave energy will combine to form spatially stationary nodes of destructive and constructive interference, the aggregate waveform constituting a "standing wave." A closed cavity, a length of wire, or the perimeter of an aperture offer geometries where multiple reflections can occur resulting in peak amplitudes that are larger than the peak amplitude of the incident wave. Since the energy of a wave is proportional to the wave amplitude, a standing wave has the capacity to intensify the energy density of the original wave.
Reinforced wave phenomena, or resonance, will occur on a wire (or the metal shielding around a wire) when its electrical length is an odd integer multiple of the incident radiation ¼ wavelength. The critical frequency defines the lowest frequency where resonance conditions can exist. At all frequencies above the critical frequency, the wire exhibits alternating resonant and anti-resonant conditions.
Under conditions of resonance, the relatively high energy of a standing wave can distort the voltage waveform on the conductive core of a wire. Fortunately, power and signalling frequencies are much lower than the frequencies associated with HIRF interference. Low pass filters can therefore be used to shield avionics from HIRF-induced interference. In practice, most aircraft avionics are designed to operate with power input variations that will exceed the variations caused by HIRF. RTCA/DO-160C, Section 16, gives the normal and abnormal power input variations that avionics must tolerate without disruption. The avionics input power is filtered to account for these variations.
Aperture resonance first occurs when the perimeter of an aperture is equal to the incident radiation ½ wavelength. As with the wire, resonance periodically recurs at higher frequencies, exhibiting broad nodes and narrow nulls. As a general rule, electromagnetic energy is efficiently transferred through an aperture at all frequencies above the critical (½ wavelength) frequency. At lower frequencies, corresponding to wavelengths more than 10 times longer than the aperture perimeter, electromagnetic radiation is attenuated by at least one order of magnitude (10 dB) as it passes through the aperture. For this reason, aircraft windows provide effective shielding against low frequency HIRF, but are essentially transparent to HIRF radiation above 30 MHz.
Aperture effects may also be observed on LRU cases, which feature holes for cabling and ventilation and which may exhibit long, narrow seams along access panels. A 15 cm long seam has a critical frequency of approximately 1 GHz. Therefore, HIRF energy at 100 MHz will be attenuated by at least 20 db as it passes through this seam. However, electromagnetic energy at frequencies above 1 GHz will pass through the same seam with no attenuation. For this reason, and for mechanical integrity, flight-critical avionics LRU enclosures incorporate closely spaced mechanical fasteners, overlapping seams, and electrically conducting gaskets to attenuate higher frequency HIRF energy.
Cavity resonance conditions exist when the length of a cavity is equal to the incident radiation ½ wavelength. Cavity resonance for LRUs of standard dimensions occurs between 1 and 3 GHz. In theory, cavity resonance can increase the peak signal strengths within an LRU by up to 14 dB. In practice, the enclosures for avionics tend not to exhibit a strong dependency to the LRU cavity resonances. This occurs because the LRUs are normally tightly packed with electronics, which makes the cavity complex. These complex cavities do not efficiently support the fundamental enclosure cavity resonant conditions.
Antennas are designed to receive RF energy in specific frequency ranges and to conduct this RF energy to the radio or radar receivers in the aircraft. HIRF energy in the appropriate frequency range that enters an antenna is, therefore, not attenuated, but is instead subject to the gain characteristics of the antenna. Standing waves are unlikely to develop because the impedance of RF components from the antenna to the receiver are carefully matched to minimize standing waves. HIRF energy outside the appropriate frequency range for the antenna and the receiver will be attenuated significantly.
Aircraft radios are designed for operation at frequencies assigned in accordance with national and international RF spectrum allocations. These RF spectrum allocations are developed to ensure that authorized high-power RF sources will not interfere with aircraft radios and radars. If an unauthorized HIRF source were to operate within the assigned frequency range for an aircraft radio, the radio receiver would only be affected by HIRF in the frequency range to which the receiver was tuned.
HIRF energy within the appropriate frequency range will normally be demodulated and amplified, along with the intended RF waveforms, within the receiver. In this way, HIRF energy may distort or corrupt legitimate signals and generally degrade the quality of reception, although some modulation types are more resilient than others. It must be emphasized, however, that RF receivers are designed to operate effectively across a wide range of dynamic signal strength. Received signals are therefore subject to gain control, to maintain an upper bound on the amount of energy that is output from the amplification chain. For this reason, HIRF energy that enters an antenna will not be amplified to unsafe levels by the receiver.
Digital devices incorporate frequency sources, or clocks, for the timing and control of internal digital functions. Aircraft avionics use digital devices that are specifically qualified for aircraft use. These digital devices tend to have slower processor and data bus clock speeds than modern consumer electronics. For aircraft flying today, the avionics processor and data bus clock speeds range from 2 MHz to approximately 300 MHz. The bandpass region for a digital device extends from the clock speed to approximately 10 times the clock speed.
HIRF interference that appears within the bandpass of a digital device may be interpreted as a legitimate control signal, driving the device into unpredictable states. HIRF interference that is not within the bandpass of the digital device may be rectified by components of the digital circuit, such as diodes. The interference will then appear as a DC offset on the control signal, triggering uncommanded state changes or locking the device into one state. Some failure modes may not be readily apparent to the operator. It is more likely, however, that error detection circuitry will detect the corrupted control signal(s), in which case error messages will be generated and system degradation will occur in a relatively controlled manner.
In the RF spectrum, digital circuits may be disrupted by potential differences ranging from 0.4 to 1.2 V. Analog circuits can be sensitive to induced gradients as small as 50 mV, although this latter value is largely dependent on the gain characteristics of the affected circuitry. However, monitoring circuits on analog systems and error detection algorithms in digital systems are normally able to detect HIRF interference before a major upset occurs. Power supply disconnects are the most common response to HIRF interference.
Given the electromagnetic coupling effects and the potential avionic system susceptibility, aircraft flight-critical avionics systems and installations must be designed to withstand the effects of HIRF. The designs incorporate HIRF protection within the LRU, on the interconnecting wires, and in the interconnecting wire routing. Avionics designs include filtering and circuit designs that reduce the effects of HIRF energy that may couple into the wiring or LRUs. The LRU enclosures are designed to minimize HIRF coupling directly through openings and seams in the enclosure. The interconnecting wires are shielded and routed through areas that are protected from direct HIRF illumination. The standards for demonstrating the HIRF protection are described below.
Since the mid-1980s, regulatory authorities have required that newly certified aircraft and modified aircraft demonstrate an acceptable level of aircraft systems protection from the effects of HIRF. The MD-11 certification was subject to special conditions imposed by both the US FAA and the multinational JAA. An operational HIRF environment and associated test procedures were developed to satisfy the FAA and JAA special conditions for MD-11 certification.
These conditions and test procedures applied to aircraft systems that were designated as either critical or essential to the safe operation of the aircraft.
The MD-11 was designed to function in the specified HIRF environments. This design was successfully demonstrated to comply with the FAA/JAA HIRF requirements during certification testing.
Swissair had installed an IFEN system in their fleet of 16 MD-11 aircraft, and 5 Boeing 747 aircraft beginning in 1996. The IFEN system provided passengers with a touch screen, which could be used to select "on-demand" movies, audio, interactive PC games, a moving map display, advertising, shopping, safety videos, news updates and secure interactive gambling. The system was installed in the first- and business-class sections of the accident aircraft in August/September 1997; the economy-class section of the aircraft was not configured for IFEN system services.
The installation of the IFEN system into the Swissair MD-11 aircraft was one element of a four-part aircraft interior re-configuration project that Swissair referred to as Product '97. The IFEN system was certified and installed under the authority of the FOCA by their acceptance of the FAA-approved STC, number ST00236LA-D. (STI)
Environmental conditions and test procedures for airborne equipment are contained in the RTCA/DO-160. EMC testing of the IFEN system LRUs was conducted in accordance with RTCA/DO-160C, sections 16.0, 17.0, 18.0, 19.0, 20.0 and 21.0. Section 20.0 of RTCA/DO-160C specified RF susceptibility tests to determine the performance of the LRUs in the presence of RF voltages coupled into the equipment by a radiated field or by direct conduction into the system by the power input or the interconnect circuit configuration. The IFEN system LRUs were classified as "Category V" equipment and tested to a maximum RF field strength of 50 V/m.
An EMI ground and flight test was conducted on the IFEN system after it was installed in the MD-11. The purpose of the test was to establish that the IFEN system performed its intended function and did not adversely affect the operation and integrity of other systems on the aircraft, or vice versa. The EMI/RF tests were conducted and, according to the records, no EMI/EMC problems were observed during the testing. However, because the IFEN system was not considered to be a critical or essential system, HIRF testing of the installed IFEN system, in accordance with the original aircraft certification special conditions, was not required and was not performed.
The electromagnetic field strength of an aircraft is primarily influenced by the distance between the aircraft and the emitter and, to a lesser extent, by the EIRP of the emitter. The HIRF hazard is greatest when an aircraft is operating near one or more high-power RF emitters, which may be land-based, ship-borne, or installed on another aircraft. In 1998, the US Navy completed an assessment for the FAA of the peak and average field intensities to which aircraft operating in US civil airspace could be exposed. The assessment considered the minimum separation distances that could exist between civil aircraft and HIRF emitters on surface and airborne platforms, excluding mobile and experimental transmitters, as well as transmitters located inside restricted, prohibited, and danger areas. This information was combined with HIRF data from Western Europe to establish combined FAA/JAA certification guidance regarding the operation of aircraft systems in external HIRF environments. Advisory Circular/Advisory Material Joint 20.1317 defines severe, normal, and certification HIRF criteria for 17 discrete frequency bands within the RF spectrum. Canadian certification requirements are similar.
The severe HIRF environment represents the strongest electric field in airspace where fixed-wing flight operations are permitted. Field strengths are calculated for peak transmitter power at a slant range of 500 feet, the minimum altitude permitted under visual flight rules, from surface emitters and airborne intercept radars. Reduced separation distances, ranging from 50 to 500 feet, are assumed for emitters located at airports and heliports.
The certification HIRF environment assumes that aircraft are operating under instrument flight rules, at a minimum altitude of 1 000 feet above ground level, outside of the airport environment. For this reason, the certification environment establishes a minimum slant range, from en route surface and ship-borne emitters, of 1 000 feet. The minimum slant ranges for airborne intercept radars and airport radars are identical to those specified for the severe HIRF environment.
The normal HIRF environment represents an estimate of the electric field strength in the airspace on and about airports and heliports in which routine departure and arrival operations take place. Similar to the criteria specified for the severe HIRF environment, separation distances ranging from 50 to 500 feet are assumed for emitters located at airports and heliports. Ship-borne and air-intercept radars are not included in the normal HIRF environment.
Operating in the 2 to 4 GHz band, the SPY-1 is part of the US Navy AEGIS. At a slant range of 1 000 feet, the SPY-1 produces a power density of 12 885 W/m2 and a peak electric field strength of 2 204 V/m. At a slant range of 33 000 feet, the power density and peak electric field strength diminish to 11.83 W/m2 and 66.8 V/m, respectively.
Operating in the 1 to 2 GHz band, the ARSR-3 has a peak power output of 5 MW, significantly greater than the more modern ARSR-4 radar. At a slant range of 1 000 feet, the ARSR-3 produces a power density of 1 759 W/m2 and a peak electric field strength of 814 V/m. At a slant range of 33 000 feet, the altitude at which SR 111 was cruising, the ARSR-3 power density and peak electric field strength diminish to 1.62 W/m2 and 24.7 V/m, respectively.
Operating in the 12 to 18 GHz band, the ASDE-3 has a peak power output of 3 kW and a nominal pulse duration of 0.04 microseconds. The ASDE-3 is typically located near the aircraft manoeuvring surface at many airports, near taxiing aircraft. At a slant range of 100 feet, the ASDE-3 produces a power density of 3 412 W/m2 and a peak electric field strength of 1 134 V/m, corresponding to an energy density of 0.136 mJ/m2.
The electromagnetic environment along the flight path of the accident aircraft was studied to determine the peak field intensity that could have been present, at the airframe, owing to external RF emitters. Information concerning fixed and mobile RF emitters in the vicinity of the accident aircraft was provided by the US Department of Defense and the Canadian Department of National Defence. Radar tracks and air traffic records were examined to determine the location of airborne RF emitters.
Various RF emitters were in operation in the vicinity of SR 111 on the night of the accident. Transmitters operating within 60 nm of the route of flight of SR 111 and having a peak radiated power of 1.0 MW or greater were classified as HIRF emitters. Transmitters having a peak radiated power of 1.0 MW or greater, that were displaced from the route of flight of SR 111 by more than 60 nm, were classified as background emitters. Fewer than five background emitters were illuminating the accident aircraft at any time during its flight. Transmitters were ignored if they produced a power density at the aircraft of less than 0.025 W/m2, corresponding to an electric field strength of less than 3.0 V/m.
The accident aircraft was not illuminated by radars associated with Canadian Navy vessels and there were no military assets operating in Canadian coastal restricted areas CYR734 through CYR752. United States military aircraft were operating along the northeast coast of the United States on the night of the accident. A KC135 air-to-air refuelling aircraft from McGuire Air Force Base, New Jersey, was orbiting over northern Vermont, New Hampshire, and Maine. Other aircraft of unknown type and origin were emitting Mode 2 transponder codes; this mode is mandatory on military transponder systems and is not normally operative on civilian aircraft. One such aircraft passed directly underneath SR 111 in the vicinity of Nantucket Sound. The unknown aircraft was flying at an altitude of approximately 5 000 feet asl and maintaining a heading of approximately 223°T. SR 111 was cruising at FL270 on a heading of 040°T. The two aircraft passed each other at 0043:37, at position 41°25'20" N, 69°45'43" W. The unknown aircraft then proceeded to Otis Air National Guard Base, Massachusetts, where it landed at approximately 0122:00. Boston ARTCC and New York TRACON (area N90) have no record of this aircraft. The FAA senior military liaison was contacted to provide additional information about the unknown aircraft's type and activities. The subsequent search of military records did not reveal any such information as the Otis air base aircraft movement records are only kept for two years and no other information was provided.
The closest point of approach between SR 111 and a US military aircraft probably occurred over Nantucket Sound, when SR 111 passed over an unknown aircraft, presumed to be military. The vertical separation between the two aircraft was approximately 3.6 nm (22 000 feet). As the aircraft approached each other, SR 111 was not in radio contact with ATC. Radio contact with ATC was re-established at approximately the same time that SR 111 passed behind the unknown aircraft. Other commercial aircraft in the vicinity of SR 111 as it passed over Nantucket Sound did not report any communication anomalies. Nearby commercial flights were Delta Airlines Flight 124, AirTours International Flight 38, and Alitalia Airlines Flight 611.
It is possible that the unknown aircraft in question could have been equipped with a radar and other high-powered RF transmitters. However, it is unlikely that an airborne HIRF emitter could have produced unsafe levels of HIRF energy from a distance of 22 000 feet. Owing to size and weight constraints, airborne RF emitters are typically less powerful than land or ship-based systems. Assuming a peak power of 100 kW, which is unusually high for an airborne system, and a nominal antenna gain of 30 dBi, the equivalent EIRP is 110 dBm. At a range of 22 000 feet (6707 m) the corresponding power density is 177 mW/m2 and the field gradient is approximately 8.2 V/m. In the 1 to 12 GHz range, where virtually all airborne HIRF emitters operate, the MD-11 HIRF certification criteria specified field gradients in the order of thousands of volts per metre.
If the unknown aircraft had RF transmitters on board that could interfere with VHF communications, nearby receivers operating on the same frequency would be expected to experience similar problems with reception. Furthermore, RF interference from an external source would not degrade the crew's ability to transmit VHF communications on the selected RF. When communications with Boston ARTCC resumed, the flight crew made no mention of any such phenomena interfering with their ability to communicate. No feasible link was established with any airborne emitter.
The accident aircraft was tracked for all or part of its flight by air traffic control radars associated with New York ARTCC, Boston ARTCC, Moncton Centre, Halifax Terminal, and the airports at Yarmouth, Sydney and Greenwood, Nova Scotia. In addition, the aircraft was tracked by eight military radars belonging to the joint surveillance system of the US Northeast Air Defense Sector and the Canadian eastern region (see "HIRF Emitters of Significance to SR 111"). The aircraft passed within 60 nm of three such military radars, two in the United States and one in Canada.
The US sites were equipped with ARSR-4 systems. The Canadian site was equipped with an AN/FPS-117 radar system. The RF parameters of the ARSR-4 and AN/FPS-117 radars are similar.
The accident aircraft passed closest to the Barrington site; the electromagnetic energy radiated from this site represents the most severe radar environment to which SR 111 could have been exposed during the en route portion of the flight. The Barrington radar files confirm that the Barrington AN/FPS-117 radar illuminated the accident aircraft as it passed overhead at an altitude of 32 969 feet and a slant range of approximately 10.5 nm (19.5 km).
The elevation angle from the Barrington radar site to the accident aircraft, at the point of closest approach, was approximately 30 degrees from the horizontal, which corresponds to the maximum elevation angle for the AN/FPS-117. The aircraft was therefore nominally within the search volume of the high-transmit beam of the AN/FPS-117 as it passed to the east of the Barrington radar site. The AN/FPS-117 and radars of similar design are optimized to achieve maximum gain at relatively low elevation angles, and cannot point, with maximum gain, at high angles. Two power calculations were completed: one using the peak rated power for the radar system as a whole, and another using the maximum power that can be radiated at an elevation angle of 30 degrees. For the AN/FPS-117 radar, far field antenna assumptions are valid for slant ranges greater than 2 337 feet (712 m).
The high-transmit beam radiates pulsed RF energy with a frequency of approximately 1.25 GHz and a pulse duration of 150 microseconds. Each pulse comprises two sub-pulses, with durations of 88 microseconds and 58 microseconds, respectively. The peak power on each sub-pulse varies between the leading and trailing edge and corresponds to a nominal peak power, averaged over the entire pulse, of 59 kW. The high-beam transmit antenna gain, PG, is approximately 36 dBi. Expressed in decibels, the peak power at an elevation angle of 30 degrees is 77.7 dBm, where PT = 10 log (59 000/0.001)). The EIRP, defined as the sum of PT and PG, is, therefore, 113.7 dBm.
Expressed in decibels, the far field power density for free space path loss is
S = EIRP 20 log10 rm 10.99 dBm/m2
where rm (the slant range in metres) is 19 478.
For an EIRP of 113.7 dBm, the power density at the aircraft is 16.92 dBm/m2 or 0.049 W/m2.
The corresponding electric field strengths are derived from the power density figures from the relationship:
E = (S 120p)½
For an EIRP of 113.7 dBm and a power density of 0.049 W/m2, E = 4.31 V/m.
The corresponding energy densities are similarly derived from the power density figures from the relationship: W = S PW, where PW is 0.150 msec.
For an EIRP of 113.7 dBm and a power density of 0.049 W/m2, E = 0.0074 mJ/m2.
Along each side of the MD-11 aircraft, the total area of the windows is approximately 7.86 m2 and the total area of the passenger and cargo doors is approximately 17.32 m2; these yield a combined area of approximately 25.18 m2 (per side). The maximum energy that could have entered the aircraft through the windows and doors is approximately 0.186 mJ.
As previously stated, fewer than five background emitters were illuminating the accident aircraft at any time during its flight. Assuming each emitter had a maximum rated power output of 1.25 MW (91 dBm) and a nominal antenna gain of 36 dBi, the corresponding EIRP for each emitter is 127 dBm. At a range of 60 nm, at the aircraft, the maximum combined power density produced by these emitters is approximately 0.161 W/m2. The aggregate electric field strength is approximately 7.8 V/m. Assuming a nominal pulse width of 0.150 msec, the corresponding energy density is approximately 0.0242 mJ/m2.
Energy produced by any single emitter may have entered the aircraft through the windows and doors on one side of the aircraft or the other, but not from both sides simultaneously. The maximum energy produced by background emitters that may have entered the aircraft is, therefore, equivalent to the product of the combined energy density (0.0242 mJ/m2) and the window/door area (25.18 m2), yielding a result of 0.6092 mJ.
The maximum combined energy penetration from foreground and background sources, as the accident aircraft passed over the Barrington radar site, is presented below.
Lightning conditions are associated with towering clouds, which cause water droplets and ice crystals to collide, creating positively and negatively charged particles. These particles separate by weight and form intense electric fields within the cloud, and between the cloud and the earth's surface. Aircraft operating near severe weather may be exposed to strong electric field gradients and an increased risk of lightning strikes.
Thunderstorm activity was forecast and observed northwest and south of JFK airport at the time SR 111 departed. After departing from Runway 13 and assuming a heading of 155°M, the crew requested and received a change of heading to 120 degrees, to avoid the weather. Cloud to ground lightning strike data indicates that the aircraft was more than 23 nm away from the closest ground strike, and much farther away from the major ground lightning activity. After departing from the JFK terminal area, there was no further lightning activity in the vicinity of the aircraft for the remainder of the flight. Following a detailed analysis of the meteorological conditions along the flight path of the aircraft, the possibility of a cloud-to-ground lightning strike on SR 111 was judged to have been unlikely. (STI)
Table: Land-Based Military Radars Tracking SR 111
The MD-11 certification process included tests to demonstrate an acceptable level of aircraft systems protection from the effects of HIRF. These test conditions were more stringent than those currently specified for the certification of new aircraft and represented field strengths that vastly exceed those actually produced by commercial and military radars at minimum range. For example, in the 2 to 4 GHz band, the MD-11 was certified to a peak field strength of 17 000 V/m, whereas the maximum field strength generated by an AEGIS class missile cruiser at 1 000 feet agl is 2 204 V/m, and the current FAA certification guidance for new aircraft is 3 000 V/m. Similar margins of safety exist in all frequency bands within the RF spectrum. No reasonable combination of known emitters and separation geometry can be shown to exceed the RF field-strength criteria used during MD-11 HIRF certification testing.
An IFEN had been installed in the first- and business-class sections of the accident aircraft. Installation of the IFEN system represented a significant modification to the original aircraft configuration, which was authorized by the issuance of an STC. The IFEN LRUs were tested, in accordance with RTCA/DO-160C, to withstand RF radiated fields to a maximum strength of 50 V/m. Electromagnetic compatibility and interference tests were also conducted on the installed IFEN system. There was no regulatory requirement to repeat the original HIRF certification test procedure on the aircraft once the IFEN system was installed, and HIRF testing was not performed. However, the RTCA/DO-160C test condition of 50 V/m, which makes no allowance for the RF shielding provided to an installed IFEN system by the aircraft skin and structure, exceeds the worst-case en route field strength of 12.1 V/m, representing the aggregate effect of the Barrington radar site and background HIRF emitters. It is unlikely that the IFEN system was significantly affected by HIRF.
After leaving the airspace of JFK airport, the most significant source of known HIRF in the vicinity of SR 111 originated from an AN/FPS-117 air route surveillance radar near Barrington, Nova Scotia. The power density and field strength of the HIRF produced by the Barrington radar, in the external environment near SR 111, were 0.049 W/m2 and 4.31 V/m, respectively, corresponding to an energy density of approximately 0.0074 mJ/m2. The maximum power density and field strength that can be produced by the Barrington radar and background emitters, at a distance equivalent to the closest point of approach between SR 111 and the radar site, are approximately 0.21 W/m2 and 12.1 V/m, respectively. By comparison, the power density of solar radiation at sea level is 1 000 times greater, and the estimated peak field intensity encountered by aircraft during normal approach and landing conditions is approximately 100 times greater. It is, therefore, probable that the normal operating environment around JFK airport was the most severe HIRF environment encountered by the aircraft during any portion of the accident flight.
The normal HIRF environment at JFK airport does not represent a hazard to aviation, as shown by the uneventful arrival and departure of many aircraft each day, including previous flights by the aircraft involved in this accident. The peak electric field strengths associated with normal airport HIRF environments are well within the MD-11 HIRF certification criteria. The minimum field strength required to induce an electrical discharge between exposed conductors (31 kV/cm) is 1 033 times greater than the peak airport HIRF, which is 3 000 V/m in the 2 to 6 GHz band, and 430 times greater than the most severe HIRF that commercial aircraft should encounter in any phase of flight: 7 200 V/m in the 4 to 6 GHz band.
The preceding observations have not addressed the electromagnetic shielding effects of the metal skin and structure of the MD-11, which significantly reduces the amount of electromagnetic radiation that penetrates an aircraft. As previously described, the amount of external electromagnetic energy that is conveyed to the internal wiring or avionics of a conventional aircraft is typically reduced by factors of 2 to 40.
The aircraft antennas can receive HIRF energy that is in the frequency range of the antennas and radio receivers. Electromagnetic energy that impinges upon an antenna will be conducted to the receiver circuits and may distort or corrupt legitimate signals. However, the radio receiver front-end saturation and gain control circuitry will prevent external HIRF from being amplified to unsafe levels. HIRF energy at frequencies outside the radio receiver frequency ranges will be rejected by the antennas and receiver bandpass filters. For these reasons, antennas will not significantly contribute to an accumulation of HIRF energy levels sufficient to cause an electrical discharge event within the internal wiring of an aircraft.
Approximately 15 minutes after taking off, radio communication with the accident aircraft was lost. The communication interruption continued for approximately 13 minutes, during which time several attempts were made by air traffic controllers to communicate with the aircraft on the assigned VHF radio frequency. During this time, aircraft systems appeared to function normally and no anomalies were reported by the aircrew. No communication anomalies were reported by the crew of other commercial aircraft in the vicinity of SR 111.
An unknown aircraft, presumed to be military, was passing underneath SR 111 during the period that radio communications were lost. RF interference from an external source, with radio communications between the accident aircraft and ATC is considered unlikely. Interference of this sort should have been received by other radios tuned to the same frequency and would not have prevented the crew of SR 111 from transmitting on the assigned frequency. A detailed analysis of this event attributed the cause of the 13-minute communications lapse to human factors.
A maximum of 0.795 mJ of energy, from the Barrington radar and background emitters, could have penetrated the aircraft windows and doors. A small per centage of this energy may have coupled to the aircraft wiring, although a reliable estimate of the coupled energy cannot be given. It is unlikely that the coupling of HIRF energy to aircraft wiring could have produced a localized field gradient of sufficient strength to induce an electrical discharge across a narrow gap separating exposed conductors. More importantly, if a HIRF-induced discharge occurred, the associated energy release would have been insufficient to ignite any flammable material. As previously discussed, approximately 0.2 mJ of energy is required to ignite a highly volatile mixture of jet fuel and air. This worst-case scenario defines a lower bound for the amount of energy required to ignite a fire. In the accident aircraft, flammable material in the area of the fire was far less volatile than a fuel/air mixture and substantially more energy would have been necessary to ignite the material. While it is clear that the material did ignite, it is equally clear that sufficient ignition energy could not have been produced solely by the penetration of HIRF energy from the Barrington radar site and other en route RF emitters, into the interior of the aircraft.
A rigorous analysis of the energy levels associated with emitters situated on or near JFK airport was not completed. However, the energy levels associated with a representative airfield emitter, the ASDE-3, are relatively benign (0.136 mJ/m2) and are significantly lower than the energy densities produced by the Barrington radar and associated background emitters. Also, as previously observed, the peak HIRF field strength for a severe airport environment (7 200 V/m) is two orders of magnitude lower than the minimum field strength (31 kV/cm) required to induce an electrical discharge between exposed conductors.
The conductive core of an electrical wire is not normally exposed, although chafing and similar physical damage to the wire can produce small regions of exposed conductor. In the accident aircraft, most of the electrical wiring was insulated with aromatic polyimide film, a lightweight thermoplastic material with a dielectric strength of approximately 2 756 kV/cm. Under certain conditions, an electrical discharge originating from a region of exposed conductor can generate sufficient energy to cause nearby wire insulation to decompose, producing additional arc sites and resulting in a cascade effect. The amount of energy necessary to thermally decompose modern aircraft wire insulation has been conservatively estimated at 690 J. In the event that a HIRF-induced spark occurred, the energy release would not have been sufficient to degrade the dielectric properties of adjacent wire insulation material on the same wire or on other wires.
 Ambient electric field strengths in lightning areas typically exceed 50 kV/m.
 This relationship was first identified in 1889 by F. Pashchen and is known as Pashchen's Law.
 E. M. Bazelyan and Yu P. Raizer, Spark Discharge (Boca Raton, Fla.: CRC Press, 1998), p. 32.
 M. S. Naidu and V. Kamaraju, High Voltage Engineering, 2nd ed. (McGraw-Hill, 1995), p. 27; and E.C. Jordan, Reference Data for Radio Engineers: Radio, Electronics, Computer, and Communications, 7th ed. (Howard W. Sams & Co., Inc., 1985), p. 484.
 Franklin A. Fisher, Some Notes on Sparks and Ignition of Fuels (Pittsfield, Mass.: Lightning Technologies, Inc., NASA/TM-2000-210077, March 2000).
 These conditions are not believed to have existed on the SR 111 aircraft.
 Franklin A. Fisher, J. A. Plumer and R. A. Perala, Lightning Protection of Aircraft (Pittsfield, Mass.: Lightning Technologies, Inc., 1990), p. 174.
 J. J. Ely, T. X. Nguyen, K. L. Dudley, S. A. Scearce, F. B. Beck, M. D. Deshpande and C. R. Cockrel, Investigation of Electromagnetic Field Threat to Fuel Tank Wiring of a Transport Aircraft (NASA TP-2000-209867, March 2000).
 Letter from David B. Walen, Chief Scientific and Technical Advisor, Electromagnetic Interference, Aircraft Certification Service, 12 March 2001 (United States Federal Aviation Administration file number ANM-110N:01-01).
 In general, electromagnetic attenuation will occur at the boundary between media with dissimilar impedances.
 The general expression for a decibel ratio is 10 log10(ratio A/B).
 Peak power is root mean square unless otherwise stated.
 Antenna gain can also be defined with respect to a half-wave dipole, denoted dBd. Gain relative to an isotropic antenna is assumed throughout this paper.
 The effective RF aperture for windows and doors is typically less than the physical opening, which is used in this paper as a worst-case assumption.
 Owing to the high energy associated with standing waves, the critical frequency also establishes a lower bound for the optimal coupling of HIRF energy to aircraft wiring.
 Instrumentation and control frequencies are below a few kHz.
 Zero lead length capacitors must be used to prevent inductance on the capacitor leads from going into series resonance with the filter capacitor (normally below 100 MHz).
 R. W. Ziolkowski and J. B. Grant, "Scattering from Cavity-Backed Apertures: The Generalized Dual Series Solution of the Concentrically Loaded E-Pol slit Cylinder Problem," IEEE Transactions on Antennas and Propagation, Vol. Ap-35, No. 5, May 1987.
 Bruce T. Clough, "Microwave induced upset of a Digital Flight Control Computer" (USAF Wright Laboratory).
 Outside of the specified frequency range, antenna gain will be uncontrolled and normally poor. In addition, the receiver will employ bandpass filters for frequency selective tuning and for adjacent channel rejection.
 Flight critical avionics use aerospace-qualified circuit components and digital devices. Owing to the expense and time required for this qualification, avionics components tend to lag consumer electronics by a couple of generations.
 Federal Aviation Administration Proposed Special Condition, McDonnell Douglas MD-11, Radio Frequency (RF) Energy Protection, 12 April 1988.
 Joint Aviation Authorities Special Condition JAA/MD-11/05, Lightning Strike Indirect Effects and External Radiation Protection, 2 May 1989.
 The system installed on Swissair's MD-11 fleet was known as the "IFEN-2 shipset."
 The installation of IFEN components into the first- and business-class seats was accomplished by the seat manufacturers under a separate STC, numbered ST01373AT.
 Swissair Document 20034, revision A, "EMI/RF Test Plan/Report Ground and Flight MD-11 Swissair", 22 January 1997.
 F. W. Heather, High Intensity Radiated Field External Environments for Civil Aircraft Operating in the United States of America (US Navy technical memorandum NAWCADPAX-98-156-TM, November 1998).
 Proposed Advisory Circular/Advisory Material Joint 20.1317, Figure 8, page 54, 8 November 1998.
 The effect of vertical sidelobes at a range of 500 feet was also considered.
 F. W. Heather, Table 16, p. 50.
 Derived from Douglas Aircraft Company, Report Number MDC K5336, "MD-11 High Energy Radiated Fields (HERF) Certification Report", 21 September 1990 using standard equations.
 Table: FAA/JAA Operational HIRF Environment for MD-11 Certification, 1-2 GHz band, normal peak HIRF during approach and landing is estimated to be 1 300 V/m.
 The typical attenuation in a large transport aircraft varies depending on the location within the aircraft. The data in draft AC 20.1317 gives attenuation that ranges from 6 to 32 dB or electric field reduction factors of 2 to 40.
 The IFEN system included electrical wire insulated with cross-linked ethylene-tetrafluoroethylene copolymer, another thermoplastic material, with a dielectric strength of approximately 1377 kV/cm. S.S. Schwartz and H.G. Sidney, Plastics Materials and Processes (Van Nostrand Reinhold Company Inc., 1982), p. 518.
 P. Ladkin and W. Schepper, "EMI, TWA 800 and Swissair 111" (University of Bielefeld, Germany, 10 October 2000).