Mostrando postagens com marcador anti-icing. Mostrar todas as postagens
Mostrando postagens com marcador anti-icing. Mostrar todas as postagens

sábado, 7 de setembro de 2024

VOEPASS 2283 [PASSAREDO CALLSIGN] PRELIMINARY REPORT - LOSS OF CONTROL IN-FLIGHT (LOC-I)

 




You can set this video for full screen and resolution 720p

SOURCE: CENIPA

LOC-I LOSS OF CONTROL IN-FLIGHT

Date: 9 August 2024

(UTC): 9 August 2024

Time: 16:22

City: VINHEDO - SÄO PAULO - BRASIL

Aerodrome: OUTSIDE THE AERODROME

Local: RESIDENTIAL AREA OF THE CITY

Damage to third

parties: YES

Injuries Function on Board Quantity

FATAL CREW 4

FATAL PASSENGERS 58

 

History

 

At 14:58 UTC, the aircraft took off from SBCA (Coronel Adalberto Mendes da Silva Airport, Cascavel, State of Paraná), bound for SBGR (Guarulhos - Governador André Franco Montoro - Airport, Guarulhos, State Of SOO Paulo) on a public regular passenger transport flight with 04 crew and 58 passengers on board. With the aircraft flying along the route, and after encountering icing conditions, control Of the aircraft was lost and it crashed into the ground.

 

Aircraft Involved

Registration marks: PSVPB

Location of latest takeoff: SBCA - ADALBERTO MENDES DA SILVA

Location of intended landing: SBGR - GOVERNADOR ANDRÉ FRANCO MONTORO

Type of operation: REGULAR

Phase of flight: CRUISE

Aircraft damage: DESTROYED

 

Sequence of events

Based on the information collected at the initial field Investigation, as well as recordings from the Flight Data Recorder (FDR) and Cockpit Voice Recorder (CVR), the Investigation Committee identified the sequence of events preceding the aircraft's collision with the ground. The time reference utilized is UTC (Universal Time Coordinated).

• 14:58:05 - the aircraft initiated takeoff from the runway 15 of SBCA, with 58 passengers and 04 crew on board;

• - the PROPELLER ANTI-ICING 1 and 2 were turned on;

• 15:14:56 - the Electronic Ice Detector connected to the Centralized Crew Alert System (CCAS) emitted an alert signal upon passing FL130;

• - the AIRFRAME DE-ICING was turned on;

• 15:15:42 - a single chime was heard in the cockpit. Subsequently, the crew commented on the occurrence of an AIRFRAME DE-ICING Fault, and that they would turn it Off;

• 151549 - the AIRFRAME DE-ICING was turned off,

- the Electronic Ice Detector ceased emitting the alert signal.

• 1516125

• 1517:08

- the Electronic Ice Detector emitted an alert signal.

- the Electronic Ice Detector stopped emitting the alert signal;

- the Electronic Ice Detector emitted an alert signal;

- the Electronic Ice Detector stopped emitting the alert signal.

- the Electronic Ice Detector emitted an alert signal;

- the Electronic Ice Detector stopped emitting the alert signal;

- the Electronic Ice Detector emitted an alert signal;

- the SIC (pilot Second in Command) made radio contact with the airline's operational dispatcher at

Guarulhos airport, for coordination of the aircraft arrival;

• - At the same time of the SIC's coordination with the operational dispatcher, a flight attendant called

over the intercom. The SIC asked her to hold on moment and continued speaking with the dispatcher,

• - the Electronic Ice Detector stopped emitting the alert signal. At this time, the SIC was asking the

flight attendant for information that would be passed to the operational dispatcher;

• 16:17:32 - the Electronic Ice Detector emitted an alert signal; at this time, the PIC was informing the passengers about the SBGR local conditions and estimated time of landing,

• 16:17:41- the AIRFRAME DE-ICING was turned on;

• 16:18:41 - at a speed of 191 kt., the CRUISE SPEED LOW alert was triggered. Concomitantly, the SIC was about to finish relaying some information to the operational dispatcher;

• 16:18:47 - the PIC started the briefing relative to the approach for landing in SBGR. Concomitantly, APP-SP made a radio call, and instructed him to change to frequency 123.25MHz;

• 16:18:55 — a single chime was heard in the cockpit. At this time, the communication with APP-SP was taking place;

• - the AIRFRAME DE-ICING was turned off;

• 16:19:16 - the crew made a call to APP-SP (Sao Paulo Approach Control) on the frequency 123.25 MHz;

• 16:19:19 - APP-SP requested the PS-VPB aircraft to maintain FL170 due to traffic;

• 16:19:23 - the crew replied to APP-SP that they would maintain flight level and that they were at the ideal point of descent, waiting for clearance;

• 16:19:28 - at a speed of 184 kt., the DEGRADED PERFORMANCE alert was triggered, together with a single chime. The alert was triggered concomitantly with the exchange of messages between APP-SP and the Crew;

• - APP-SP acknowledged the message and requested the aircraft to wait for clearance;

• 16:19:31 - Passaredo 2283 aircraft reported receipt of the message and thanked ATC;

• - the PIC resumed delivering the approach briefing;

• - the Second in Command (SIC) commented, "a lot of icing";

• - the AIRFRAME DE-ICING was turned on for the third time;

• - APP-SP cleared the aircraft to fly direct to SANPA position, maintaining FL170, and informed that the descent would be authorized in two minutes;



• 16:20:39 - the crew acknowledged the flight instruction received (last communication performed by the flight crew);

• - the aircraft started a right turn in order to fly to SANPA position.

• 16:20:57 — during the turn, at a speed of 169 the INCREASE SPEED alert was triggered, in conjunction with a single chime. Immediately afterwards, vibration noise was heard in the aircraft, simultaneously with the activation of the stall alert;

• 16:21:09 - control of the aircraft was lost, and it entered an abnormal flight attitude until colliding with the ground. The aircraft rolled to the left to a bank-angle of 52 degrees, and then rolled to the right to a bank- angle of 94 degrees, performing a 180-degree turn in a clockwise direction. Subsequently, the turn was reversed to an anticlockwise direction, with the aircraft completing five full rotations in a flat spin before crashing into the ground.


Click on image to see it isolated




The ICING light would blink with the detection of an Icing condition and the Anti-Icing and/or De-lcing  (AIFRAME) were not selected to ON, followed by single chime. The light would remain illuminated in a continuous fashion with the systems turned on.

 

Anti-Icing and De-Icing Systems

 

The Anti-lcing functions were energized electrically, whereas the De-icing ones were provided by means of pneumatic pressure.

 

The APM system needed to be checked by the crew on a daily basis, and in case of a failure, an amber-colored FAULT message would illuminate on the APM panel.

If the aircraft's drag increased due to ice accumulation and performance was degraded, resulting in loss of cruise speed, alerts in three levels were triggered and presented to the pilots on both alert panels Of the APM, as follows:

• 1st Level - CRUISE SPEED LOW

The blue-colored message would indicate performance degradation Of around 10%, with reduction Of the Indicated Air Speed (IAS) during the cruise phase by at least 10 kt. below the speed computed by the APM.

This alert would be triggered only during the cruise phase.

• 2nd Level - DEGRADED PERFORMANCE

The amber-colored message would be followed by a single chime and a master caution alert, indicating a significant performance degradation in the range between 22% and 28%, induced by a significant increase in aerodynamic drag, causing a drop in cruise IAS of around 15 to 20 knots below the speed computed by the APM. This alert could be triggered during climb, cruise, Or descent.

• 3rd Level - INCREASE SPEED.

The amber-colored message would appear flashing, followed by a single chime and a master caution alert, indicating that the degraded performance condition had worsened , reaching an IAS value below the ICING BUG + 10 kt. This alert could be triggered during climb, cruise, or descent.

 


The pilot has set the ICING BUG SPEED for SEVERE ICING CONDITION to 165 Knots.


In addition to the speed alerts (emitted by the APM), the airspeed indicators of the left- and right-hand cockpit stations had BUGS for reference, particularly for minimum speed maneuvers at low bank, flaps O', and icing conditions (VMLBO ICING), The said BUGS could be adjusted manually.

 

The ICING BUG needed to be adjusted by the pilots for each flight in accordance with the aircraft's weight, in order to indicate the minimum speed for a flight in icing conditions and with flaps retracted. The VMLBO ICING.




















quinta-feira, 17 de fevereiro de 2011

Have You Had Flight Simulator's Lazy Responses in Your Training?



TRANSPORT DELAY, means

The total flight simulator system processing time required for an input signal from a pilot primary flight control until motion system, visual system or instrument response. It is the overall time delay incurred from signal input until output response and is independent of the characteristic delay of the aeroplane simulated.

DEMORA DE TRADUÇÃO, significa


O tempo total exigido do processamento do sistema de simulador de voo para um sinal de entrada [feito] por um piloto através do controle de voo primário [demorar] até o sistema se mover, o sistema ser visualizado ou [obter] a reposta de um instrumento.


É a demora total de tempo implicada da entrada do sinal até a resposta de saída e é independente da demora característica do avião simulado.

For initial evaluation of flight simulators, the aeroplane manufacturer’s validation flight test data is preferred.

Para avaliação inicial de simuladores de voo, os dados de validação dos testes de vôo do fabricante do avião são preferidos.

The Problem

• Loss of control is the leading cause of fatalities in the worldwide commercial jet fleet

• In early 2010, the NTSB recommended:

– training centers develop and conduct training that incorporates stalls that are fully developed and unexpected

– simulation model fidelity requirements to support an expanded set of stall recovery training requirements be defined and codified

Today’s Upset Training Requirements

• FAA:

Requires recoveries in the simulator from approach-to-stalls in the clean, takeoff, and landing configurations

• European Aviation Safety Agency

Used to demonstrate ability to recover from full stall, although now it is typically a briefing

• Transport Canada

Upset training is required for airline operations

O Problema


• Perda de controle é a causa líder de fatalidades na frota mundial de jatos comerciais


• No início de 2010, a NTSB recomendou:
-  centros de treinamento desenvolvam e conduzam treinamento que incorporem perda de sustentação que seja totalmente avançada e inesperada
-  requisitos de modelo de fidelidade de simulação para apoiar uma série ampliada de requisitos de treinamento de recuperação de perda de sustentação

Requisitos de Treinamento Apreensivo de Hoje

• FAA:
Exige recuperações no simulador de vôo das recuperações de aproximação de perda de sustentação nas configurações limpa, decolagem e pouso


• EASA:
Acostumado a demonstrar habilidade para recuperar perda total de sustentação, embora isso seja tipicamente um briefing


• TRANSPORT CANADA
Treinamento com Transtorno é requerido para operações de linhas aéreas


TESTING FOR FLIGHT SIMULATOR QUALIFICATION

Requirements

1. GENERAL

1.1 Flight deck, a full-scale replica of the aeroplane simulated.
Direction of movement of controls and switches identical to that in the aeroplane. Equipment for operation of the cockpit windows should be included in the flight simulator, but the actual windows need not be operable.

Note - The flight deck, for flight simulator purposes, consists of all that space forward of a cross section of the fuselage at the most extreme aft setting of the pilots’ seats. Additional required flight crew member duty stations and those required bulkheads aft of the pilots’ seats are also considered part of the flight deck and shall replicate the aeroplane.

1.2 Circuit breakers that affect procedures and/or result in observable flight deck indications properly located and functionally accurate.

1.3 Flight dynamics model that accounts for various combinations of drag and thrust normally encountered in flight corresponding to actual flight conditions, including the effect of change in aeroplane attitude, thrust, drag, altitude, temperature, gross mass, moments of inertia, centre of gravity location and configuration.

1.4 All relevant instrument indications involved in the simulation of the applicable aeroplane to automatically respond to control movement by a flight crew member or external disturbance to the simulated aeroplane, i.e. turbulence or wind shear.

1.5 Communications, navigation, and caution and warning equipment corresponding to that installed in the applicant’s aeroplane with operation within the tolerances prescribed for the applicable airborne equipment.

1.6 In addition to the flight crew member duty stations, three suitable seats for the instructor/observer and authority inspector. The authority will consider options to this requirement based on unique flight deck configurations. The location of these seats shall provide an adequate view of the pilots’ panels and forward windows.

Observer seats need not represent those found in the aeroplane but shall be adequately secured to the floor of the flight simulator, fitted with positive restraint devices and of sufficient integrity to safely restrain the occupant during any known or predicted motion system excursion.


1.7 Flight simulator systems to simulate the applicable aeroplane system operation, both on the ground and in flight. Systems shall be operative to the extent that all normal, abnormal and emergency operating procedures can be accomplished.

1.8 Instructor controls to enable the operator to control all required system variables and insert abnormal or emergency conditions into the aeroplane systems.

1.9 Control forces and control travel which correspond to that of the replicated aeroplane. Control forces should react in the same manner as in the aeroplane under the same flight conditions.

1.10 Ground handling and aerodynamic programming to include:

1.10.1 Ground effect. For example: round-out, flare and touchdown.
This requires data on lift, drag, pitching moment, trim and power in ground effect.

1.10.2 Ground reaction. Reaction of the aeroplane upon contact with the runway during landing to include strut deflections, tire friction, side forces and other appropriate data, such as weight and speed, necessary to identify the flight condition and configuration.

1.10.3 Ground handling characteristics. Steering inputs to include crosswind, braking, thrust reversing, deceleration and turning radius.

1.11 Wind shear models which provide training in the specific skills required for recognition of wind shear phenomena and execution of required manoeuvres. Such models shall be representative of measured or accident derived winds, but may include simplifications which ensure repeatable encounters. For example, models may consist of independent variable winds in multiple simultaneous components. Wind models should be available for the following critical phases of flight:

1) prior to take-off rotation;

2) at lift-off;

3) during initial climb;

4) short final approach.

Note - The United States FAA Wind Shear Training Aid, wind models from the United Kingdom Royal Aerospace Establishment (RAE), the Joint Airport Weather Studies (JAWS) project or other recognized sources may be implemented and shall be supported and properly referenced in the QTG. Wind models from alternative sources may also be used if supported by aeroplane related data and such data are properly supported and referenced in the QTG.


Use of alternative data must be coordinated with the authority prior to submission of the QTG for approval.

1.12 Representative crosswinds and instructor controls for wind speed and direction.

1.13 Representative stopping and directional control forces for at least the following runway conditions based on aeroplane related data:

1) dry;

2) wet;

3) icy;

4) patchy wet;

5) patchy icy;

6) wet on rubber residue in touchdown zone.

1.14 Representative brake and tire failure dynamics (including antiskid) and decreased braking efficiency due to brake temperatures based on aeroplane related data.

1.15 A means for quickly and effectively conducting daily testing of flight simulator programming and hardware.

1.16 Flight simulator computer capacity, accuracy, resolution and dynamic response to fully support the overall flight simulator fidelity.

1.17 Control feel dynamics which replicate the aeroplane simulated.

Free response of the controls shall match that of the aeroplane within tolerance given in Appendix B. Initial and upgrade evaluations will include control-free response (pitch, roll and yaw controllers) measurements recorded at the controls. The measured responses shall correspond to those of the aeroplane in take-off, cruise and landing configurations.

1.17.1 For aeroplanes with irreversible control systems, measurements may be obtained on the ground if proper pitot static inputs are provided to represent conditions typical of those encountered in flight. Engineering validation or aeroplane manufacturer rationale shall be submitted as justification to ground test or to omit a configuration.

1.17.2 For simulators requiring static and dynamic tests at the controls, special test fixtures will not be required during initial evaluations if the QTG shows both test fixture results and alternate test method results, such as computer data plots, which were obtained concurrently. Repeat of the alternate method during initial
evaluation may then satisfy this requirement.

1.18 Relative response of the visual system, flight deck instruments and initial motion system coupled closely to provide integrated sensory cues. Visual scene changes from steady state disturbance (i.e. the start of the scan of the first video field containing different information) shall occur within the system dynamic response limit of 150 milliseconds (ms). Motion onset shall also occur within the system dynamic response limit of 150 ms. While motion onset should occur before the start of the scan of the first video field containing different information, it must occur before the end of the scan of the same video field. The test to determine compliance with these requirements shall include simultaneously recording the output from the pilot’s pitch, roll and yaw controllers, the output from the accelerometer attached to the motion system platform located at an acceptable location near the pilots’ seats, the output signal to the visual system display (including visual system analog delays) and the output signal to the pilot’s attitude indicator or an equivalent test approved by the authority. The following two methods are acceptable means to prove compliance with the above requirement:

1.18.1 Transport delay: A transport delay test may be used to demonstrate that the flight simulator system response does not exceed 150 ms. This test shall measure all the delays encountered by a step signal migrating from the pilot’s control through the control loading electronics and interfacing through all the simulation software modules in the correct order, using a handshaking protocol, finally through the normal output interfaces to the motion system, to the visual system and instrument displays. A recordable start time for the test should be provided by a pilot flight control input. The test mode shall permit normal computation time to be consumed and shall not alter the flow of information through the hardware/software system. The transport delay of the system is then the time between the control input and the individual hardware responses. It need only be measured once in each axis.

1.18.2 Latency: The visual system, flight deck instruments and initial motion system response shall respond to abrupt pitch, roll and yaw inputs from the pilot’s position within 150 ms of the time, but not before the time, when the aeroplane would respond under the same conditions. The objective of the test is to compare the recorded response of the flight simulator to that of the actual aeroplane data in the take-off, cruise and landing configuration for rapid control inputs in all three rotational axes. The intent is to verify that the simulator system response does not exceed 150 ms (this does not include aeroplane response time as per the manufacturer’s data) and that the motion and visual cues relate to actual aeroplane responses. For aeroplane response, acceleration in the appropriate corresponding rotational axis is preferred.

1.19 Aerodynamic modelling, that includes, for aeroplanes issued an original type certificate after June 1980, low altitude level flight ground effect, Mach effect at high altitude, normal and reverse dynamic thrust effect on control surfaces, aeroelastic effect and representations of non-linearities due to side-slip based on aeroplane flight test data provided by the aeroplane manufacturer.

1.20 Modelling that includes the effects of airframe and engine icing.

1.21 Aerodynamic and ground reaction modelling for the effects of reverse thrust on directional control.

1.22 Realistic implementation of aeroplane mass properties, including mass, centre of gravity and moments of inertia as a function of payload and fuel loading.

1.23 Self-testing for simulator hardware and programming to determine compliance with the simulator performance tests as prescribed in Appendix B. Evidence of testing must include flight simulator number, date, time, conditions, tolerances and the appropriate dependent variables portrayed in comparison to the
aeroplane data. Automatic flagging of “out-of-tolerance” situations is encouraged.

1.24 Timely permanent update of flight simulator hardware and programming subsequent to aeroplane modification.

1.25 Daily pre-flight documentation either in the daily log or in a location easily accessible for review.


2. MOTION SYSTEM

2.1 Motion cues perceived by the pilot representative of aeroplane motions (e.g. touchdown cues should be a function of the simulated rate of descent).

2.2 A motion system which produces cues at least equivalent to those of a six-degree-of-freedom synergistic platform motion system.

2.3 A means of recording the motion response time as required.

2.4 Motion effects programming to include:

2.4.1 Effects of runway rumble, oleo deflections, ground speed, uneven runway, centre line lights and taxiway characteristics;

2.4.2 Buffets on the ground due to spoiler/speed-brake extension and thrust reversal;

2.4.3 Bumps associated with the landing gear;

2.4.4 Buffet during extension and retraction of landing gear;

2.4.5 Buffet in the air due to flap and spoiler/speed-brake extension;

2.4.6 Approach-to-stall buffet;


2.4.7 Touchdown cues for main and nose gear;

2.4.8 Nosewheel scuffing;

2.4.9 Thrust effect with brakes set;

2.4.10 Mach and manoeuvre buffet;

2.4.11 Tire failure dynamics;

2.4.12 Engine malfunction and engine damage; and

2.4.13 Tail and pod strike.

2.5 Motion vibrations: Tests with recorded results that allow the comparison of relative amplitudes versus frequency are required.

2.5.1 Characteristic motion vibrations that result from operation of the aeroplane, in so far as vibration marks an event or aeroplane state that can be sensed at the flight deck, shall be present. The flight simulator shall be programmed and instrumented in such a manner that the characteristic vibration modes can be measured and
compared to aeroplane data.

2.5.2 Aeroplane data are also required to define flight deck motions when the aeroplane is subjected to atmospheric disturbances.

General purpose disturbance models that approximate demonstrable flight test data are acceptable. Tests with recorded results that allow the comparison of relative amplitudes versus frequency are required.


3. VISUAL SYSTEMS

3.1 Visual system capable of meeting all the standards of this appendix and Appendices B and C.

3.2 Continuous, cross-cockpit, minimum collimated visual field of view providing each pilot with 180 degrees horizontal and 40 degrees vertical field of view. Application of tolerances requires the field of view to be not less than a total of 176 measured degrees horizontal field of view (including not less than ±88 measured degrees either side of the centre of the design eye point) and not less than a total of 36 measured degrees vertical field of view from the pilot’s and co-pilot’s eye points.

3.3 A means of recording the visual response time for visual systems as required.

3.4 Visual textural cues to assess sink rate and depth perception during take-off and landing.

3.5 Horizon and attitude correlated to the simulated attitude indicator.

3.6 A minimum of ten levels of occulting.

3.7 Surface resolution demonstrated by a test pattern of objects shown to occupy a visual angle of not greater than 2 arc minutes in the visual display used on a scene from the pilot’s eye point.

3.8 Light-point size — not greater than 5 arc minutes.

3.9 Light-point contrast ratio — not less than 25:1.

3.10 Daylight, twilight (dusk/dawn) and night visual capability as defined by terms in the glossary section of this document

3.10.1 Contrast ratio. A raster drawn test pattern filling the entire visual scene (three or more channels) shall consist of a matrix of black and white squares no larger than 10 degrees and no smaller than 5 degrees per channel with a white square in the centre of each channel.


4. SOUND SYSTEM

4.1 Significant flight deck sounds corresponding to those of the aeroplane which result from pilot actions.

4.2 Sound of precipitation, rain removal equipment and other significant aeroplane noises perceptible to the pilot during normal and abnormal operations and the sound of a crash when the simulator is landed in excess of limitations.

4.3 Comparable amplitude and frequency of flight deck noises, including engine and airframe sounds. The sounds shall be coordinated with the required weather.

4.4 The volume control shall have an indication of sound level setting which meets all qualification requirements.

REFERENCES

ICAO Doc 9625 AN/938, Manual of Criteria for the Qualification of Flight Simulators, 2003


segunda-feira, 10 de janeiro de 2011

Future of Deicing Technology and Effective Training for Flight in Icing Conditions


Icing effects in even the most sophisticated flight simulators are simple models that do little more than increase aircraft weight to simulate in-flight icing.

Currently, pilots only experience the effects of an ice protection system failure for the first time in a real flight situation.

Snow
Normally snow is not a hazard with respect to icing, unless it begins to adhere to aircraft surfaces. If snow does begin to stick, it should then be treated as an icing encounter because ice may begin to form under this accumulation of snow. This is a rare situation for which no aircraft is evaluated in the icing-certification process. If it occurs, the aircraft should exit the conditions as quickly as possible, coordinating with ATC as necessary.


AIRCRAFT ICING CONDITIONS
Nearly all aircraft icing occurs in supercooled clouds. These are clouds in which liquid droplets are present at temperatures below 0 ºC (32 ºF). At temperatures close to 0 C (32 ºF), the cloud may consist entirely of such droplets, with few or no ice particles present. At decreasing temperatures, the probability increases that ice particles will be found in significant numbers along with the liquid droplets. In fact, as the ice water content increases, the liquid water content tends to decrease, since the ice particles grow at the expense of the water particles. At temperatures below about -20 ºC (-4 ºF), most clouds are made up entirely of ice particles.

SUBLIMATION. A process where ice turns directly into water vapor without passing through a liquid state.

SUPERCOOLED DRIZZLE DROP (SCDD). Drizzle drops aloft within supercooled clouds. The term currently is not in wide use.

SUPERCOOLED LARGE DROP (SLD). A supercooled droplet with a diameter greater than 50 micrometers (0.05 mm). SLD conditions include freezing drizzle drops and freezing raindrops.

SUPERCOOLED LIQUID WATER (SLW). Liquid water at temperatures below 0 °C. SLW is found in clouds, freezing drizzle, and freezing rain in the atmosphere. This water freezes on aircraft surfaces. Most aircraft icing occurs in supercooled clouds, which consist of SLW, sometimes with ice crystals.

LIGHT ICING. The rate of accumulation may create a problem if flight is prolonged in this environment. Occasional use of deicing/anti-icing equipment removes/prevents accumulation.

ICING ENVELOPES. These icing envelopes, found in 14 CFR part 25, appendix C, are used for the certification of aircraft for flight in icing conditions. They specify atmospheric icing conditions in terms of altitude, temperature, liquid water content (LWC), and droplet size represented by the median volume diameter (MVD). (The envelopes use the term mean effective diameter (MED), but this equates to the median volume diameter for the instrumentation and assumptions current at the time the envelopes were established.) There are two classes of icing envelopes: continuous maximum and intermittent maximum. The continuous maximum is for stratus-type clouds, and the intermittent maximum is for cumulus-type clouds.

FREEZING RAIN. Freezing rain consists of supercooled liquid water drops with diameters of 500 micrometers (0.5 mm) or greater, and a typical representative diameter of 2 mm. The droplets tend to break up if their size exceeds approximately 6 mm. When encountered by an aircraft in flight, freezing rain can cause a dangerous accretion of icing.

FREEZING DRIZZLE. Freezing drizzle consists of supercooled liquid water drops with diameters smaller than 500 micrometers (0.5 mm) and greater than 50 micrometers (0.05 mm).

CLEAR ICE. A glossy, clear, or translucent ice formed by the relatively slow freezing of large supercooled water droplets. The terms “clear” and “glaze” have been used for essentially the same type of ice accretion, although some reserve “clear” for thinner accretions which lack horns and conform to the airfoil.

MIXED ICE. Simultaneous appearance of rime and clear ice or an ice formation that has the characteristics of both rime and clear ice.

MODERATE ICING. The rate of accumulation is such that even short encounters become potentially hazardous and use of deicing/anti-icing equipment or flight diversion is necessary.

RIME ICE. A rough, milky, opaque ice formed by the instantaneous freezing of small, supercooled water droplets.

SEVERE ICING. The rate of accumulation is such that deicing/anti-icing equipment fails to reduce or control the hazard. Immediate flight diversion is necessary.


Structural Icing
Ice forms on aircraft structures and surfaces when supercooled droplets impinge on them and freeze. Small and/or narrow objects are the best collectors of droplets and ice up most rapidly. This is why a small protuberance within sight of the pilot can be used as an “ice evidence probe.” It will generally be one of the first parts of the airplane on which an appreciable amount of ice will form. An aircraft’s tailplane will be a better collector than its wings, because the tailplane presents a thinner surface to the airstream.
The type of ice that forms can be classified as clear, rime, or mixed, based on the structure and appearance of the ice. The type of ice that forms varies depending on the atmospheric and flight conditions in which it forms.

 Condições de Formação de Gelo em Aeronave

Aproximadamente todas formações de gelo em aeronave ocorre em nuvens super resfriadas. Estas são nuvens nas quais gotículas estão presentes em temperaturas abaixo de 0ºC (32ºF).
Em temperaturas perto de 0ºC, a nuvem pode consistir inteiramnete de tais gotículas, com pouca ou nenhuma partícula de gelo presente. Em temperaturas mais baixas, a probabilidade aumenta de modo que partículas de gelo serão encontradas em significantes números acompanhadas de gotículas líquidas. De fato, quando o conteúdo de água gelada aumenta, o conteúdo de água líquida tende a diminuir, desde que as partículas de gelo crescem às custas das partículas de água. Em temperaturas abaixo de -20ºC (-4ºF), muistas núvens são formadas inteiramente de partículas de gelo.

A regra geral é que quanto mais partículas de gelo e menos gotículas líquidas que estão presentes, MENOS acúmulo de gelo na estrutura.
Isto é porque as partículas de gelo tendem a quicar para fora da superfície da aeronave, enquanto as gotículas super resfriadas congelam e aderem à superfície. Como resultado, o acúmulo de gelo é frenquentemente maior em temperaturas NÃO muito distante de ZERO Celcius (0ºC), onde o conteúdo da água líquida pode ser abundante, e é usualmente desprezível em temperaturas abaixo de -20ºC.

Uma excessão à regra geral é também declarada que pode ser feita para superfícies aquecidas por um sistema te´rmico de proteção contra gelo ( ou por aquecimento aerodinâmico perto do ponto de estagnação de um componente da aeronave em velocidades em excesso de talvez 250 Knots). Para tais superfícies, partículas de gelo podem derreter sob impacto e depois retroceder para regiões traseiras mais frias e congelar.

Avião algumas vezes acumula pouco ou nenhum gelo mesmo quando voando dentro de nuvens em temperaturas NÃO muito abaixo de 0ºC. Um explicação é que embora tais nuvens sejam usualmente compostas predominantemente de gotículas de água líquida, elas algumas vezes consistem unicamente de partículas de gelo, e neste caso o acúmulo de gelo será mínimo.


Quanto maior o conteúdo de água líquida da nuvem, mais rapidamente é acumulado gelo nas superfícies da aeronave.

O tamanho das gotículas também é importante. Goticulas maiores têm maior inércia e são menos influenciadas pelo fluxo de ar em volta da aeronave do que as gotículas menores. O resultado pe que gotículas maiores colide sobre mais superfície da aeronave do que as gotículas menores.


ICING CONSIDERATIONS FOR THE PHASES OF FLIGHT

PREFLIGHT
Freezing level. Pilots should locate freezing levels on forecast maps. This will assist in developing a contingency plan in the event icing is encountered.

TAXI
While taxiing in snow or ice, leave extra space around your aircraft and taxi at a slower rate.

TAKEOFF AND CLIMBOUT
Verify that the airspeed indicator is working properly and that the pitot heat is on.

CRUISE
It is possible to exit the icing conditions by a change in altitude or a minor change in flightpath, this is certainly advisable.

DESCENT
Pilots may consider periodically disengaging the autopilot and hand flying the airplane when operating in icing conditions. If this is not desirable because of cockpit workload levels, pilots should monitor the autopilot closely for abnormal trim, trim rate, or airplane attitude.
Pilots should try to stay on top of a cloud layer as long as possible before descending into the clouds. This may not be possible for an aircraft that uses bleed air for anti-icing systems because an increase in thrust may be required to provide sufficient bleed air. If configuration changes are made during this phase of flight, they should be made with care in icing conditions.

HOLDING
During holding, an airplane may be more vulnerable to ice accumulation because of the slower speeds and lower altitudes during this phase of flight.
Caution concerning the use of the autopilot, as described above, is also applicable to holding during or after flight in icing conditions.
If the aircraft reacts adversely to a change of configuration, the pilot should return the aircraft to its original configuration.

APPROACH AND LANDING
During or after flight in icing conditions, when configuring the airplane for landing, the pilot should be alert for sudden aircraft movements.
Extension of landing gear may create excessive amounts of drag when coupled with ice. Flaps and slats should be deployed in stages, carefully noting the aircraft’s behavior at each stage.
Before beginning the approach, deicing boots should be cycled because they may increase stall speed and it is preferable not to use these systems while landing.
Once on the runway, pilots also should be prepared for possible loss of directional control caused by ice buildup on landing gear.
Windshield anti-icing and deicing systems can be overwhelmed by some icing encounters or may malfunction. Pilots have been known to look out side windows or, on small GA aircraft, attempt to remove ice accumulations with some type of tool (plotter, credit card).

Wing Stall
The wing, when contaminated with ice, will ordinarily stall at a lower AOA, and thus at a higher airspeed. Even small amounts of ice, particularly if rough, may have some effect. An increase in approach speeds may be advisable if any ice remains on the wings.
The stall characteristics of an aircraft with ice-contaminated wings may be markedly degraded, and serious roll control problems are not unusual.

Você pode notar no gráfico abaixo que há suficiente tempo para recuperar a atitude normal do voo entre o "Alerta de Stall" e o "Coeficiente Máximo de Sustentação".
Pode ser notado também o "Angulo de Ataque" [AOA] bem menor quando as asas contém gelo o que fará o STALL ocorrer em ângulos de ataque bem menores aos ângulos quando as asas não contém gelo.

Fidelidade do Simulador de Voo para Treinamento de Stall

Infrared Deicing Systems
Infrared (IR) deicing technology involves melting frost, ice, and snow from aircraft surface with infrared energy. IR energy systems are based on natural gas- or propane-fired emitters that are used to melt frost, ice, and snow. Infrared energy does not heat up the surrounding air and tests have shown that it has negligible effect on the aircraft cabin temperature [13].


[13] Environmental Protection Agency. Preliminary Data Summary Airport Deicing Operations (Revised). August 2000.


Two main manufacturers are leading the way in developing infrared-based deicing systems: Radiant Energy Corporation with the InfraTek™ system and Infra-Red Technologies with the Ice Cat™ system. The Ice Cat™ system uses IR emitters fueled by natural gas or propane mounted on booms that are fitted to specially designed trucks. Thebooms are then positioned above the aircraft surface and the IR emitters are used to remove frost, ice, and snow. Currently there is no commercial application of this system.

The InfraTek™ system consists of infrared generators, called Energy Processing Units (EPUs), located in an open-ended, hangar-type structure. The EPUs are fueled by natural gas and generate IR energy waves to melt and evaporate frost, ice, and snow. If the aircraft surface is dry, the IR waves are reflected away. A diagram of this process is shown below (figure taken from [17]):




The InfraTek™ system is designed to be operated by only one person and is mainly controlled by a computer. Before deicing can begin, the floor of the InfraTek™ system facility is heated in order to facilitate the deicing process of aircraft lower parts such as the landing gear. Depending on the type of aircraft and the severity of ice and snow build up, the energy and wavelength generated by the EPUs are adjusted. In March of 1996, InfraTek™ technology was shown to deice a Boeing 727 in six minutes, which is about the same amount of time it would take to achieve deicing results using conventional ADFs [13].

There are currently three InfraTek™ Deicing System facilities in the US (December 2009) and one in Oslo, Norway. The largest by volume is the one installed at JFK International Airport in March 2006. The facility was operating during the 2006-2007 deicing season. The system is designed to provide deicing services for up to a 747-300 size aircraft [18]. The following performance benefits have been documented:
- Approximately 90% reduction of glycol use per aircraft under snow and ice conditions.
- No glycol use for defrosting operations.


Furthermore, in terms of budget management, using the InfraTek™ system allows more accurate winter operations budgets for customers since the system charges a fixed fee based on the size of the aircraft. Conventional deicing methods are priced based on the volume of fluid applied, which varies based on the severity of winter weather conditions. However, cost data is currently limited by the number and scale of facilities using the InfraTek™ system.

The JFK InfraTek™ facility is reported to have cost $9.5 million [18]. With such infrastructure costs, there needs to be a commitment by airlines to make use of the facility.



Effective Training for Flight in Icing Conditions


Billy P. Barnhart
Bihrle Applied Research, Inc.
Jericho, New York 11753


Thomas P. Ratvasky
National Aeronautics and Space Administration


Glenn Research Center
Cleveland, Ohio 44135


How ice contamination affects aircraft handling so they may apply that knowledge to the operations of other aircraft undergoing testing and development. Participant feedback on the ICEFTD was very positive.
A safe way to explore the flight envelope.
University of Tennessee Space Institute
Billy P. Barnhart engineer in the development of NASA’s ICEFTD.


Thomas P. Ratvasky is an aerospace engineer at the NASA Glenn Research Center’s Icing Branch. he also served as the on-board flight test engineer during the icing flights of the Twin Otter Icing Research Aircraft.


ICEFTD Training Sessions


Four formal demonstrations of the ICEFTD took place between October 2004 and November 2005. In total, eighty-four pilots and flight test engineers from the industry, regulatory and military communities received practical lessons on an iced airplane’s reduced performance and handling qualities.


The first two demonstrations were held in October 2004 and May 2005 at the University of Tennessee Space Institute in Tullahoma, Tennessee.


Twenty-four pilots and flight test engineers from Bombardier, Cessna, Raytheon, U.S. Army, U.S.
Forestry Service, the FAA, and Canada’s Transportation Safety Board participated in the In-Flight Icing and Its Effects on Aircraft Handling Qualities short course. This short course consisted of lectures, a flight in the UTSI Navion variable stability airplane, and a simulator session in NASA’s ICEFTD.


The capability for including icing effects into flight training simulators used for initial and recurrent training will allow pilots to experience representative icing-induced aircraft handling characteristics,especially in failure case training scenarios. Presently, icing effects in even the most sophisticated flight simulators are simple models that do little more than increase aircraft weight to simulate in-flight icing.


Realistic icing simulator models however, based on aerodynamic effects of airframe icing, will enhance safety by allowing pilots to recognize important visual and tactile cues associated with an icing event.



Currently, pilots only experience the effects of an ice protection system failure for the first time in a real flight situation. As in stall and windshear training, improved icing flight simulation will better equip pilots to employ the correct procedures and techniques to effect a recovery to a safe flight condition.


The airplane chosen for this activity was a DeHavilland DHC–6 Twin Otter since NASA had extensive operational experience in icing conditions with this airplane and the Twin Otter has a known sensitivity to ice contaminated tailplane stall.

Icing Effects
These exercises demonstrated stark differences in handling characteristics and pitch control between the non-iced Twin Otter and the Twin Otter with failure case ice shapes. For example, no abrupt pitch or roll tendencies occurred during stalls with the non-iced Twin Otter. However, during stalls with failure ice, pilots experienced and commented on the “roll-off” tendency, especially with flaps set at 20°. These rolloffs resulted in steep bank angles, sometimes reaching or exceeding 90°. Figure 6 contains plots from one of the flap 20° stalls. The first 100 sec shown is a stall maneuver for the No-Ice configuration. Note that there were no strong roll off or pitch down tendencies at 70 sec when the stall angle was achieved. After 100 sec, the All-Ice switch was enabled, and the stall maneuver with ice was performed. Observe at about 170 sec the abrupt stall with large sideslip and roll the pitch angle achieved a 50° nose down attitude during the recovery.


During flap transitions with failure ice, pilots noted flap position in relation to the first signs of controllability problems. Many pilots expressed some surprise when encountering the very high control forces associated with wing flap extension, and discovered the difficulty in maintaining good pitch attitude control. All pilots experienced large pull forces to maintain 85 knots as flaps transitioned beyond 20°, and when the column force also became oscillatory, most were unable to maintain good airspeed control.


As power was added, the forces increased and when power was set to idle, the forces decreased. Likewise, as airspeed increased, the pull forces increased and were more oscillatory and when airspeed was decreased, the forces and oscillations were reduced.


When flaps were raised, all noted that the forces and oscillations went away when flaps were less than 15°.
This introduction provided a necessary practice and orientation for the more difficult approach tasks that followed.





Operational Scenarios
These exercises were developed to provide realistic pilot task with the iced aircraft. Since icing issues often arise during the approach and landing phases, it was appropriate to look at these phases for the exercise.


Performing the approach using basic instrument displays required a higher pilot workload to perform the task. From the outset it was apparent that for some pilots, the workload was high in order to fly the basic IMC approach task using the raw data glide slope and localizer presentation. This was the reason for changing the training profile during the second week, so pilots could get a feel for the basic task workload with no compensation for ice, and comparisons on the pilot performance and HQR’s could be made between the No-Ice and All-Ice configurations.


No-Ice Approach and Landing
This first IMC task allowed the pilots an opportunity to understand the workload of the basic task, while developing their instrument scan. Pilot performance and ratings indicated good flying qualities with relatively low levels of pilot compensation required to meet the desired performance of the task. This exercise was initiated in the clouds, and the pilot was given heading and altitude instructions to intercept the localizer. The flaps were transitioned from dF = 0° to dF = 30° by the final approach fix, and the pilot was to maintain 85 knots on final approach. The pilot made power changes and trim changes accordingly as flaps and speed were set to maintain descent rate along the glide slope. All pilots were able to trim the airplane and fly with one hand on the yoke, and one hand on the throttle.


Data from one of these approaches are shown in figures 7 and 8. This pilot felt the workload was fine; he could make radio frequency changes, read maps, etc. It took him some time to get established on airspeed and had some lateral overshoots on the localizer. Overall, he rated the task with an HQR = 3 (aircraft characteristics were fair with some mildly unpleasant deficiencies, but desired performance was achieved without improvement).

All-Iced Approach and Landing



Pilot performance and HQR’s for the approach and landing in the All-Ice configurations generally indicated a much more challenging airplane to fly.


The second and third approaches were made with All-Ice and dF = 20° and dF = 30°, respectively. When the flaps were lowered, the control anomaliesexperienced during the flap transition were revisited.


With dF = 20°, a slight pull force was required even with full-nose up trim in order to maintain the 85 knot target approach speed. With dF = 30°, a significant pull force (15 lb) was required to maintain 85 knots. Force oscillations occurred on top of these steady pull forces, making speed and attitude control difficult. Most pilots used two hands on the yoke to control the airplane.


Many pilots found that the workload associated with flying an approach to CAT I minimums with the failure case icing condition was in many cases at the limits of their abilities. Handling problems were caused by horizontal tail ice, but the task was further complicated the failure ice on the wing, which resulted in lateral handling problems made it difficult to control airspeed.
In some cases, wing stall recoveries were difficult due to the high induced drag at high angle of attack and low altitude. Increased power was not enough to break the stall, and there was not much altitude to exchange for airspeed. These cases would sometimes result in a crash. All experienced pilots found that most of their attention had to be devoted to controlling pitch to achieve vertical path performance. This element of the task consumed most of their attention because of icing related instability, high control forces, inability to trim, and related pitch control anomalies. Although not sensed by the pilot, during the pitch excursions, the Gmeter would often cycle between 0 to 2.5 G. Because of the intense amount of attention required for pitch control using a basic attitude indicator, lateral path performance and airspeed control usually suffered.


This pilot had difficulty throughout the approach. Airspeed was consistently high, which made the pitch control more difficult. Power changes were large and abrupt, which degraded handling further.


Concentrating on the longitudinal problems, he failed to intercept the localizer initially and had multiple lateral overshoots throughout the approach. Towards the middle of the approach, the large pitch oscillations caused multiple stall warnings as G was increased and airspeed decreased. Since it was clear that the pilot was task saturated, the instructor suggested raising the flaps to 10° to finish the approach. The workload was considerably deceased with this configuration, and the pilot was able to successfully land the airplane. Note in figure when flaps reached 10°, the oscillations in angle of attack, pitch angle, and longitudinal input were greatly reduced. This pilot rated the iced airplane with dF = 30° as an HQR = 10 (aircraft characteristics have major deficiencies, and that control would be lost during some point of the operation).

Handling Quality Ratings
After each approach and landing, pilots familiar with HQR’s were asked to use the Cooper-Harper rating system to rate their ability to complete the approach and landing tasks within specific adequate and desired performance metrics. The HQR’s from sixteen pilots who performed the approaches in the No-Ice/ dF = 30°,


All-Ice/dF = 20°, and All-Ice/dF = 30° configurations were compiled into figure.


For the No-Ice, dF = 30° configuration, pilots rated the approach and landing between an HQR = 2 to HQR = 4. Many pilots commented on the lack of practice as contributing to the higher rating, and this point should be considered in using the actual HQR number. Even without much practice, the general rating was that desired performance was achieved and aircraft characteristics were good to fair with some deficiencies.


For the All-Ice, dF = 20° configuration, pilot ratings ranged from an HQR = 4 to HQR = 9. The increase in the ratings was certainly due to the reduced stabilityand controllability as well as the handling anomalies associated with icing. The increased spread in the numbers reflected to some degree the skill level and experience of the individual pilot receiving the training. Additionally, the spread could have been influenced by the “learning curve” or proficiencygained as the training progressed and pilots became more familiar with the displays and aircraft characteristics. This was an expected result as pilot task performance generally improved with practice and it did affect pilot ratings. In a more rigorous pilot evaluation setting, a certain amount of time would be given to pilots to develop a baseline level of proficiency and familiarization before rating a task.


For the All-Ice, dF = 30° configuration, pilot ratings ranged from an HQR = 7 to HQR = 10. This increase in the ratings reflects the increased workload associated with the greater amount of pilot compensation at the higher flap setting. In this configuration, none of the pilots were able to meet the adequate performance metrics of the approach due to the inability to trim, the large and oscillating column forces, large pitch excursions when making power changes (one-handed operation), and loss of situational awareness on the lateral position. As stated above, some pilots encountered wing stalls and were unable to recover in this configuration due to increased drag and limited altitude. Although most pilots rated the dF = 30° approach the worst case condition, a few rated this task slightly better (lower compensation) than the dF = 20° approach. These pilots felt that the proficiency acquired on the dF = 20° approach, which was flown before they flew the technically more difficult dF = 30° approach, was the reason. Overall, pilot ratings accurately reflected the inability of meeting performance requirements with flaps extended. Most pilots exited the training device perspiring freely and commenting that no additional workout was needed for the rest of the day. These comments confirmed the physical effort and intense concentration required to perform the training tasks.


Pilot Comments on ICEFTD Training Sessions
All pilots commented very favorably on their training experience and the applicability it had to their present occupations. Pilots who had participated in development programs for aircraft with reversible controls identified strongly with the characteristics shown by the ICEFTD. Others who had not had this experience were in general surprised by the amount of effort it took to perform the given tasks. None had ever encountered the levels of control forces or feedback activity as was demonstrated during this simulation.


One pilot commented that he recently completed an extensive icing development and certification program on a Part 25 business jet with a major aerospace company. During the development phase with 22.5 min ice shapes on the horizontal tail leading edge, he observed large, uncommanded stick pumping and pitch transients when flaps were moved to the landing configuration. He also noted that the aircraft could not be trimmed at the approach speed. Stick pumping developed and strengthened as the aircraft accelerated with increase power setting. The simple procedure of raising the flaps to the last setting completely eliminated the problem. This pilot was pleasantly surprised to observe that the ICEFTD accurately simulated the same phenomenon that he had experienced, even though the two aircraft were radically different in design, tailplane configuration, and powerplant. The lectures and the ICEFTD demonstration gave him a clear understanding of what he had experienced during their icing program. He strongly recommend the ICEFTD to any flight test crew (test pilots and flight test engineers) who are in the process of preparing for an icing development or certification program. He considered this training a“must have”.


REFERENCE


Barnhart, B., Dickes, E., Gingras, D., Ratvasky, T., Simulation Model Development for Icing Effects Flight Training, SAE 2002–01–1527, April 2002.


Gingras, D.R., Dickes, E.G., Ratvasky, T.P., and Barnhart, B.P., Modeling of In-Flight Icing Effects for Pilot Training, AIAA Modeling and Simulation Technologies Conference and Exhibit, Aug. 5–8, 2002, Monterey, CA, AIAA Paper 2002–4605.


Icing for General Aviation, NASA Glenn, 2001, http://iceboxesn.grc.nasa.gov/ext/education/video/video.html


Papadakis, M., Gile Laflin, B.E., Youssef, G.M., and Ratvasky, T.P., Aerodynamic Scaling Experiments with Simulated Ice Accretions, AIAA 39th Aerospace Sciences Meeting and Exhibits, Jan. 8–11, 2001, Reno, NV, AIAA Paper 2001–0833.


Ratvasky, T., Blankenship, K., Rieke, W., Brinker, D., Iced Aircraft Flight Data for Flight Simulator Validation, SAE–2002–01–1528, April 2002.


Ratvasky, T.P., Ranaudo, R.J., Barnhart, B.P., Dickes, E.G., and Gingras, D.R., Development and Utility of a Piloted Flight Simulator for Icing Effects Training, AIAA 41st Aerospace Sciences Meeting and Exhibit, Jan. 6–9, 2003, Reno, NV, AIAA Paper 2003–0022.


Russell, P., Pardee, J., Joint Safety Analysis Team-CAST Approved Final Report Loss of Control JSAT Results and Analysis, December 2000. Tailplane Icing, NASA Glenn, 1998, http://iceboxesn.grc.nasa.gov/ext/education/video/video.html


REFERENCES


[1] Aviation Safety Network. Retrieved from http://aviation-safety.net/database/record.php?id=19820113-0
[2] Aviation Safety Network. Retrieved from http://aviation-safety.net/database/record.php?id=20041121-0
[3] Federal Aviation Administration. Code of Federal Regulations. Section 121.629.
[4] Federal Aviation Administration. Ground Deicing and Anti-icing Program. Circular No. 120-60B.
[5] Federal Aviation Administration. Ground Deicing and Anti-Icing Training and Checking. Circular No. 135-16.
[6] Federal Aviation Administration. Pilot Guide Large Aircraft Ground Deicing. Circular No. 120-58.
[7] Federal Aviation Administration. Pilot Guide – Small Aircraft Ground Deicing. Circular No. 135-17.
[8] Federal Aviation Administration. Design of Aircraft Deicing Facilities. Circular No. 150/5300-14B.
[9] International Civil Aviation Organization Requirements and Guidance. Retrieved from http://www.icao.int/anb/FLS/flsicing.html
[10] Airport Cooperative Research Program. Deicing Planning Guidelines and Practices for Stormwater Management Systems. Report 14.
[11] Environmental Protection Agency. Environmental Impact and Benefit Assessment for Proposed Effluent Limitation Guidelines and Standards for the Airport Deicing Category. July 2009.
[12] Airport Cooperative Research Program. Deicing Practices. Fact Sheet 4.
[13] Environmental Protection Agency. Preliminary Data Summary Airport Deicing Operations (Revised). August 2000.
[14] Dejak, J., Monaghan, M. M., & Salenieks, E. Central de-icing facility at Toronto’s airport. Retrieved from http://www.esemag.com/0100/deicing.html
[15] Environmental Protection Agency. Technical Development Document for Proposed Effluent Limitation Guidelines and Standards for the Airport Deicing Category. July 2009.
[16] Welte, M. (2009). EPA Targets Plane De-Icing Chemical. CNSNews.com. Retrieved from http://www.cnsnews.com/news/article/54705
[17] Hessing, H., Knoesel, E., & Sharkey, I. Infrared Aircraft Deicing Facility at John F. Kennedy International Airport. Retrieved from http://www.mmtmagazine.org/spring_06_jfk_deicing.html
[18] Airport Cooperative Research Program. Deicing Practices. Fact Sheet 6.
[19] Federal Aviation Administration. Ground Deicing Using Infrared Energy. Circular No. 120-89.
[20] Airport Cooperative Research Program. Deicing Practices. Fact Sheet 6.
[21] The FAA and NCAR are working to solve the Aircraft Ground Icing Problem. Weather Support to Deicing Decision Making. Retrieved from http://www.rap.ucar.edu/projects/wsddm/
[22] Federal Aviation Administration. Airport Winter Safety and Operations. Circular No. 150/5200-30C.
[23] Richardson, D. (2008). Aircraft Deicing Fluid: The Recycling Solution. Stormwater. Retrieved from http://www.stormh20.com/march-april-2008/aircraft-deicing-pollutant.aspx