APPENIDX K. ICE AND ICING CONDITIONS DETECTION K.1 ICE AND ICING CONDITIONS DETECTION.
Ice detectors may provide information to the flight crew or aircraft systems about aircraft icing.
The ice detector may be intrusive or nonintrusive to the airflow. It may be directly or indirectly sensitive to the buildup of ice or sensitive to meteorological conditions conducive to icing.
These devices may be classified as inflight icing detection system (FIDS), ground icing detection system (GIDS), or icing conditions detectors. Some possible ice detector technologies include:
latent heat of fusion, changes in electrical fields, changes in the natural frequency of vibrating components, visual cues, ultrasonic waves, optical methods such as infrared or laser devices, and friction between a rotating cylinder and scraper. Chapter 2, Section 1 of the Aircraft Icing Handbook (Reference K1) and SAE AS5498 (Reference K2) contain more information on ice detector technologies and ice detectors. GIDS are associated with detecting ice on aircraft before takeoff. Appendix V of this AC contains guidance on GIDS. The following discussion focuses on FIDS.
A FIDS that detects ice buildups will, after detecting ice, signal the flight crew or aircraft systems (such as the IPS and stall protection system), or both, about ice buildup on monitored aircraft surfaces. It may also provide information about ice thickness, ice buildup rate, LWC, cloud drop size, and where the ice is. Current FIDS will detect ice buildup resulting from the airspeed, temperature, and moisture content near the ice detector’s sensor. Since dynamic heating depends on airspeed, ice may not build up on the ice detector’s sensor or on aircraft surfaces even though those conditions may be within those defined in 14 CFR parts 25, Appendix C and 29, Appendix C. Since local airspeeds and temperatures depend on the surface curvature, AOA, aircraft speed, and other variables, ice buildup will not occur at all locations at the same time. Applicants may install a FIDS on or remote from aircraft lifting surfaces, engine inlets, or other critical, monitored surfaces. Since critical surfaces may not be the reference surfaces, you must correlate ice buildup on critical surfaces with ice on the ice detector’s reference surface (sensor). You need this correlation to ensure safe flight in icing conditions (14 CFR §§ 23.1419, 25.1419, 27.1419, and 29). The FIDS should detect aircraft icing within the icing conditions described in 14 CFR parts 25, Appendix C and 29, Appendix C.
A FIDS that detects icing conditions signals the flight crew or aircraft systems that the aircraft is in atmospheric conditions that are conducive to aircraft icing. Sensor components of the FIDS are intrusive or nonintrusive to the airflow. A FIDS that detects icing conditions by sensing a temperature threshold and the presence of moisture, is not necessarily sensitive to the presence of airframe ice buildup.
FIDS are divided into two categories:
x Advisory flight ice detector system (AFIDS) and x Primary flight ice detector system (PFIDS).
K-1
APPENDIX K. ICE AND ICING CONDITIONS DETECTION (CONTINUED) K.1.1 AFIDS.
An AFIDS provides an advisory alert to flight crews about of the presence of ice accretion or icing conditions. However, applicants should not use an AFID as the only means for ice or icing conditions detection since an AFID may not have adequate reliability. Therefore, the flight crew is responsible for monitoring icing conditions as directed in the AFM. Typical methods pilots use to monitor icing conditions include:
x Total air temperature.
x Visible moisture criteria.
x Visible ice buildups on specific areas, such as on windshield wiper blades and arms or on ice evidence probes.
Typically, the flight crew activates the aircraft IPS after the AFIDS provides an icing alert and after the flight crew confirms existence of the icing condition. Some AFIDSs activate the IPS automatically. Activation of the IPS may activate other systems (for example, activation of the IPS may adjust the stall protection schedule to provide acceptable stall margin for flight in icing conditions). The AFIDS tells the flight crew about the presence of ice accretion or icing conditions, but the crew only uses this information with other information, as described in the AFM, to determine the need for activating the IPSs.
K.1.2 PFIDS.
A PFIDS provides an alert to the flight crew about the presence of ice on the aircraft or the presence of icing conditions. The PFIDS may also give information to other aircraft systems.
Applicants can use a PFIDS, along with approved AFM procedures, as the only means of detecting aircraft icing or icing conditions. PFIDSs are either primary automatic or primary manual. After detecting aircraft icing or icing conditions as a primary automatic system, the PFIDS automatically activates the airframe IPSs, may activate the propulsion system IPS, and may adjust the stall protection system attitude schedules, as necessary. Activation of the propulsion system IPS may occur earlier than the airframe IPS.
In a primary manual PFIDS, the flight crew activates the IPSs based on the PFIDS alert. Either the automatic PFIDS or manual activation of the IPS should adjust the stall protection schedules to ensure acceptable stall speed margins and stall characteristics during flight in icing conditions.
A PFIDS must be self-monitoring and should provide an indication to the flight crew when there is an ice detector fault. The flight crew can then perform AFM procedures to ensure continued safe flight and landing.
K.1.3 Ice Detector Sensor Characteristics.
To ensure the acceptability of the FIDS, the aircraft manufacturer should understand its characteristics, limits, and abilities. This understanding should include:
K-2
APPENDIX K. ICE AND ICING CONDITIONS DETECTION (CONTINUED) x Sensor ice detection technology.
x Limits.
x Atmospheric conditions and aircraft flight regimes in which the sensor will detect ice or icing conditions.
x Ice accretion rate on the sensor.
x Sensor ice collection efficiency.
x Ice or icing conditions detection response time.
x Deicing cycle of the ice detector sensor.
x Tendencies of the FIDS sensor signal to become saturated during icing conditions.
K.1.4 Freezing Fraction (n) Considerations.
The response time for an ice accretion detector to sense the presence of ice varies with the sensor’s design and technology. As the local temperature of the ice detector sensor approaches 0ºC, all the cloud water drops that impinge on the detector’s sensor may not freeze. Some of the impinging water drops may remain in the liquid phase and may shed from the detector’s sensor, resulting in a delayed detection of icing. However, local flow fields with higher velocities elsewhere on the aircraft, such as turbine engine inlets and at areas of high local velocities on lifting surfaces, may display lower local temperatures and higher rates of ice buildups.
Therefore, there are conditions where significant amounts of ice may build up on critical surfaces without detection by the ice detector.
Freezing fraction is the amount of water that freezes at contact on a surface divided by the surface’s total water catch. Applicants must consider the effects of freezing fractions less than one when installing a FIDS that detects ice buildups on an airplane. At local temperatures well below the freezing point, icing clouds water drops freeze on contact with the ice detector’s sensor, resulting in predictable ice detection response times. As the freezing fraction drops below 1, the ice detection response times for ice detectors may increase significantly. You should show that the ice detector alerts the flight crew to the presence of icing for all icing conditions that cause the airframe and engine induction system to build up unsafe quantities of ice. This requires showing that the ice accretion rate (W) of the ice detector’s sensor is higher than for any other part of the aircraft for all 14 CFR parts 25, Appendix C and 29, Appendix C icing conditions. It also requires that the airframe and propulsion system induction system ice buildups are not hazardous when the ice detection response time is delayed. Reference K3 provides one means of doing this. Current ice accretion computer codes do not provide accurate results at low freezing fractions (glaze icing conditions), therefore you should evaluate your ice detector installation using an icing wind tunnel or an icing tanker if possible, or by natural icing flight tests at freezing fractions below 1.
K-3
APPENDIX K. ICE AND ICING CONDITIONS DETECTION (CONTINUED) K.1.5 Drop Splashing, Runback, and Shedding.
Other considerations that applicants should examine for ice accretion detectors are water drop splashing, runback, and shedding. Splashing occurs when the impact of a drop on the surface causes the drop to break into smaller water particles. The fragment drop particles may bounce off the sensor surface into the airstream. The drop may remain in the airflow or reimpinge on the sensor. Runback icing occurs when impinging drops do not immediately freeze on contact with the sensor surface. The drops remain in a liquid state and may coalesce with other drops. The shear force of the local airflow may force these large drops downstream and they may shed from the sensor surface before freezing. Runback and shedding are associated with freezing fractions less than 1 and become more pronounced as the freezing fraction approaches zero. You should examine ice detector drop splashing for all conditions when the freezing fraction is less than 1.
Splashing or shedding may occur on the ice detector sensor, but not on larger airplane surfaces.
This may result in delayed ice detection by the ice detector, even though ice is building up on aircraft surfaces. Failure or delay in the ice detector’s detection of ice could result in hazardous ice buildup on surfaces (for example, on the wing, horizontal stabilizer, or the engine induction system inlets). You should ensure safe flight in all icing conditions.
Some impingement computer codes calculate the quantity of runback icing. They do not address the runback ice characteristics nor water mass loss because of drop splashing or shedding. Also, the runback results of these codes have not been validated using wind tunnel or flight tests.
Therefore, you should use icing wind tunnel tests to examine the splashing, shedding, and runback characteristics of the ice detector.
K.2 INSTALLATION LOCATION.
Applicants should install the ice detector where it can detect aircraft icing reliably before there is hazardous ice on the airframe or engine inlets. After mounting the ice detector, you must ensure that it announces icing before there is unsafe ice accretion on monitored surfaces. Proper location is critical to ensure adequate detection of icing. Locate the ice detector on the airframe surface where the sensor is adequately exposed to the icing environment. Proper location of the ice detector requires flow field and boundary layer thickness analyses of candidate sensor locations. Perform the analysis at AOAs, airspeeds, sideslip angles, and with aircraft configurations approved for flight in icing. Local flow conditions and boundary-layer conditions may shield the sensor from acceptable sampling of the icing environment. If this is true, finding a better location for the ice detector is important. Understand that local flow conditions may:
x Alter the freestream icing characteristics by preventing drops of certain sizes from impinging on the ice detector sensor (shielding or shadowing the sensor location by carrying off drops not having enough momentum to penetrate local flow conditions);
x Result in unacceptable airspeeds when compared with the freestream airspeed; and K-4
x APPENDIX K. ICE AND ICING CONDITIONS DETECTION (CONTINUED) Result in unacceptable variation in orientation of the sensor to local flow angularity and airspeed with changes of the airplane’s configuration and flight regime.
You should show acceptable performance of the ice detector when installed on the aircraft. To ensure locating the ice detector properly on the aircraft, you may model the airplane threedimensionally in a viscous-flow CFD code analysis. Examine effects of the aircraft boundary layer on the ability of the ice detector to detect icing. (Reference K3 provides useful information on the acceptable installation of a magnetostrictive technology PFIDS). If the wing or other aircraft components do not influence the airflow at the candidate ice detector location, you may confine the CFD analysis to the aircraft area of interest (for example, the forward fuselage).
Consider the effects of the propulsion system, such as the propeller slipstream and flow field variations that may occur with thrust changes if the detector is located in and near turbine engines. You may need to perform wind tunnel, tanker, and natural icing tests to verify these analyses.
Pilots have occasionally detected ice before receiving the PFIDS alert. This can occur because of poor ice detector placement (as well as other reasons like other components on the airplane having higher ice collection efficiency or low freezing fraction conditions). Poor sensor placement can cause a delay or failure of an ice detector to detect ice buildup or icing conditions.
This may result in a hazardous condition. Reference K3 documents that improper placement of an ice detector sensor can result in the detection of large drops and small drops, but that mediumsized drops can be aerodynamically shadowed and prevent ice detection in a timely manner.
This result is counterintuitive and shows that you must perform thorough analyses and testing to ensure proper placement of PFIDS sensors on an airframe and selection of PFID sensors that will detect all drop sizes. Helicopters may require a system to induce airflow across PFIDS sensors during low speed flight. You should conduct similar analyses and testing to ensure detection of ice buildups or icing conditions before an unsafe condition occurs on rotorcraft. CFD codes are a practical way to perform the placement analysis. Validation of the analyses is typically done during natural icing flight tests. Verify the ability of the ice detector to detect icing before there is ice accretion on the monitored surface during measured natural icing flight tests.
Besides the proper placement of the FIDS’ sensors for ice detection, you should show that ice buildups forward of the sensor (for example, on the radome) do not interfere with the ice detector’s performance. You should show that shed-ice will not damage the detectors. Also, other probes installed forward of the ice detector should not adversely interfere with the ice detector’s performance.
K.3 SYSTEM CONSIDERATIONS.
During previous certifications, the FAA considered an extremely improbable unannunciated failure to detect ice or icing conditions to be the ice detector’s top-level system requirement.
This failure is not expected to occur during the life of the fleet. The assumed hazard level is catastrophic. Applicants may consider reduced hazard levels for a specific aircraft when showing compliance with 14 CFR §§ 23.1309, 25.1309, 27.1309, and 29.1309 for new PFIDS K-5
APPENDIX K. ICE AND ICING CONDITIONS DETECTION (CONTINUED) certifications, if you have enough evidence of the reduced hazard. You should get concurrence from the FAA before using reduced hazard levels in designing a PFIDS.
There should be at least dual ice detector systems. You should separate the detector systems to ensure that a common fault or condition will not interfere with ice detection. Each detector should have an independent failure monitoring system. You must show the effectiveness of the PFIDS during the icing flight tests for 14 CFR parts 23, 25, 27, and §§ 29.1093 and 29.1419.
The PFIDS should also have the following minimum abilities:
x The threshold ice buildup, or icing condition, chosen to activate the detector’s alert of icing or icing conditions and the ice buildup that occurs before the IPS becomes effective must allow safe flight of the aircraft. Do this by ensuring that preactivation ice is not hazardous and that when the ice buildup sheds there will be no hazardous or unsafe damage to the engines and airframe.
x The ice detector should not be excessively sensitive and produce nuisance alerts or IPS activation. Nuisance alerts (frequent changes from “on” to “off”) may encourage pilots to ignore ice detector information. However, the ice detector must be sensitive enough to detect sudden exposure to icing conditions reliably throughout the icing conditions of 14 CFR parts 25, Appendix C and 29, Appendix C.
x If overheating of structural material (such as the engine inlet cowl) can result from automatic activation of the IPS by the PFIDS (including nuisance and false alerts), provide the flight crew an alert so they perform IPS shutdown procedures as provided by the AFM.
x The AFM must address the use of the PFIDS and its limits. The AFM must also provide procedures that should be performed following failure alerts.
The PFIDS must be reliable and meet the requirements of 14 CFR §§ 25.901(c), 23.1309,
25.1309, 27.1309, 29.901(c), and 29.1309. This means that combinations of system failures that result in undetected ice or icing conditions must be extremely improbable. The requirement exists because flight crews must be informed when the PFIDS has failed. This allows the crew to revert to other acceptable methods of ice detection, such as monitoring temperature, atmospheric moisture, and other icing cues. You may consider multiple and independent ice detectors, automatic fault monitoring, built-in test equipment (BITE), and preflight status tests, for example, when showing the required ice detector reliability. More detailed guidance is in ACs 23.1309-1B, 25.1309-1A, 27-1B, 29-2C, and 33-2B.
K.4 CERTIFICATION OF ICE DETECTORS.
K.4.1 Certification of an AFIDS.
When used to get approval for flight in icing conditions, ice detector equipment become a part of the aircraft’s IPS. The ice detector performance must allow the ice protection of the airframe, engine, engine induction system, propeller, windshields, and air data system to comply with the K-6
APPENDIX K. ICE AND ICING CONDITIONS DETECTION (CONTINUED) requirements discussed in section 6 of this AC. An AFIDS should detect ice within the icing environment defined by 14 CFR parts 25, Appendix C and 29, Appendix C or as limited by the rotorcraft’s flight envelope. Applicants should show this for the requested aircraft configurations and airspeeds. The AFIDS and the primary methods of ice detection (for example, temperature and moisture, visible ice accretion) should advise the pilot to begin operation of the IPS using AFM procedures before there are hazardous buildups of ice or unsafe flight.
Following ice detection, you should define the ice buildups on the aircraft when the IPS becomes effective. These ice buildups should account for reasonable time delays in detecting ice and in activating the IPS. With these ice buildups, you should evaluate aircraft performance and handling qualities to ensure safe flight, as discussed in section 6.3 of this AC. The AFIDS should perform its intended functions for all aircraft configurations and flight regimes, and in the required icing environment.
To install an AFIDS properly on an aircraft, you should perform analyses similar to those performed for a PFIDS. You should understand the characteristics and limits of the AFIDS.
Pilots may place the same reliance on an AFIDS as they would place on a PFIDS—even though the AFIDS may not have the same reliability. This underscores the importance of performing these necessary analyses and providing suitable AFM information to ensure proper functioning and use of the AFIDS.
An AFIDS approval should include flight tests in measured natural icing conditions to verify analyses and laboratory test results. The natural icing flight tests should show that the ice detector performs its intended function within the required icing environment and for the requested aircraft configurations and flight regimes. You may use laboratory tests, such as icing wind tunnel testing or icing tanker testing, to show the ice detector’s ability to perform properly in 14 CFR parts 25, Appendix C and 29, Appendix C icing conditions not found during natural icing flight tests. An AFIDS should meet reliability requirements if the AFIDS is used with other icing cues to show the reliability of the IPS (14 CFR §§ 23.1301, 23.1309, 25.901(c),
25.1301, 25.1309, 27.901(b)(1), 27.1301, 27.1309, 29.901(c), 29.1301, and 29.1309). Consider the probability of flying in icing conditions as 1. ACs 23.1309-1C, 25.1309-1A, 27-1B, 29-2C, and 33-2B provide detailed guidance on equipment reliability. You should provide procedures for using the ice detector as part of the IPS in the AFM (14 CFR §§ 23.1585, 25.1585(a)(6),
27.1585, and 29.1585).
K.4.2 Certification of a PFIDS.
When used to get approval for flight in icing conditions, ice detector equipment becomes a part of the aircraft’s IPS. The ice detection equipment must meet the requirements of 14 CFR §§ 23.1419, 23.1301, 23.1309, 25.901(c), 25.1419, 25.1301, 25.1309, 27.901(b), 27.1419,
27.1301, 27.1309, 29.901(c), 29.1419, 29.1301, and 29.1309. A PFIDS must detect aircraft icing in the icing conditions of 14 CFR parts 25, Appendix C and 29, Appendix C, or the icing environment that is limited by the rotorcraft’s flight envelope. The PFIDS must also detect icing for the icing conditions required for aircraft components (for example the engines), unless there are alternate AFM procedures for these components. The PFIDS must detect aircraft icing for all aircraft configurations and airspeeds requested. An icing condition PFIDS must detect icing K-7
APPENDIX K. ICE AND ICING CONDITIONS DETECTION (CONTINUED) conditions of 14 CFR parts 25, Appendix C and 29, Appendix C or as limited by the rotorcraft’s flight envelope. The PFIDS must e nsure safe flight by either alerting the pilot to use the IPS according to AFM procedures or by automatically activating the IPS before hazardous or unsafe buildups of ice on the airframe, engine, or engine inlets, unless the applicant shows that the undetected icing is not hazardous to the aircraft. Follow the safe flight requirements discussed in section 6.3 of this AC. The PFIDS must be able to perform its intended functions for all aircraft configurations, flight regimes, and in the required icing environment. The PFIDS performance should account for reasonable time delays in the activation and effectiveness of the IPS. The PFIDS performance should also account for the effects of residual and intercycle ice if the PFIDS is part of an automated deicing system.
You should describe the aircraft’s ice buildups when the IPS becomes fully effective after detection of icing. You should show that the aircraft’s performance and handling qualities, as well as the engine’s performance and operability, comply with the requirements of safe flight with these ice buildups. As part of the ice protection system, approval of the PFIDS must include natural icing flight tests to verify analyses and laboratory test results (14 CFR §§ 23.1419, 25.1419, 27.1419, and 29.1419). PFIDS approval must also verify that the ice detector performs its intended function. Laboratory tests should show the ice detector’s ability to perform properly with the required icing environment and requested flight regimes (for example, airspeed, angle of attack, altitude, and Mach number). You may not experience these conditions during flight tests.
To show the performance and reliability of a PFIDS, you will need to perform detailed analysis (14 CFR §§ 23.1301, 23.1309, 25.901(c), 25.1301, 25.1309, 27.901(b)(1), 27.1301, 27.1309,
29.901(c), 29.1301, and 29.1309). The analyses should include a thorough understanding of the characteristics of the ice detector’s sensors and their influences on airframe and engine icing.
Detailed equipment reliability guidance information is in ACs 23.1309-1C, 25.1309-1A, 27-1B, 29-2C, and 33-2B. Consider the probability of flying in icing conditions as 1. Provide procedures for using the ice detector as part of the IPS in the AFM (14 CFR §§ 23.1585,
25.1585(a)(6), 27.1585, and 29.1585). Analyses that may be required include analyses of the ice detector’s: sensor characteristic; installation location; freezing fraction; the effects of impinging drop splashing, shedding, and runback on the ice detectors’ response time; and reliability.
K.5 REFERENCES.
K1. “Aircraft Icing Handbook,” FAA Technical Report DOT/FAA/CT-88/8-1 (1991), 3 vols., FAA William J. Hughes Technical Center, Atlantic City International Airport, NJ 08405.
K2. “ Minimum Operational Performance Specification for Inflight Icing Detection Systems,” SAE AS5498, October 2001.
K3. “Certification and Integration Aspects of a Primary Ice Detection System,” D. Jackson, D.
Owens, D. Cronin, and J. Severson, paper no. AIAA-2001-0398, AIAA 39th Aerospace Sciences Meeting, Jan. 8-11, 2001. Reprints available from American Institute of Aeronautics and Astronautics, Reston, VA 20191-4344.
K-8
APPENIDX L. WINDSHIELD ICE PROTECTION Applicants should show the acceptability of the windshield IPS by testing. You should perform inner and outer windshield surface temperature surveys of the protected areas to verify design thermal analyses. The thermal analyses should show that surface temperatures are warm enough to provide acceptable anti-icing protection without causing damage to the windshield. Evaluate the transparency of the protected windshield area, including optical distortion effects, for day and night operations. Also review the size and location of the protected area for adequate visibility, especially for approach and landing conditions. A probable single failure of a transparency heating system should not adversely affect the integrity of the airplane cabin or create a potential fire hazard. You should develop safety analysis using guidance provided in AC 23.1309-1C and AC 25.1309-1. This shows the windshield IPS meets reliability and safety requirements.
L-1 (and L-2)
APPENIDX M. FINDING NATURAL ICING CONDITIONS FOR TEST PURPOSES M.1 BACKGROUND.
Icing conditions for flight-testing may be elusive, especially when aircraft program schedules require testing during seasons with weather that is not conducive to inflight icing conditions.
Even when the weather is conducive to inflight icing conditions, the prevailing icing intensity may not be enough to achieve test objectives. Figure M-1 shows statistical information on the occurrence of atmospheric icing conditions. Information in the following section describes how to get icing conditions forecasts that may ease planning for natural icing flight tests.
M.2 STATISTICAL GUIDANCE.
The LWC values graphed in figures E-1 and E-4 of appendix E of this AC are about the 99th percentile values expected during random flights in icing conditions. Applicants will seldom find these large values in practice. Test flights usually experience much smaller values. The FAA William J. Hughes Technical Center compiled about 28,000 nm of inflight measurements of cloud water content, drop sizes, temperatures, and other variables in supercooled clouds over portions of North America, Europe, and the northern oceans. This information provides more practical statistical data on icing conditions. The following paragraphs describe some of the results.
M.2.1 Natural Probabilities for LWC Averages in Stratiform Icing Conditions.
Figure M-1 depicts the statistical probability of finding various average values of LWC versus averaging distance in continuous stratiform clouds. The 50 percent curve shows that half of all random icing events will yield less than 0.2 g/m3, depending on the event distance. The 90 percent curve shows that 90 percent of the events will contain less than 0.45 g/m3.
g ( ) 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9 1
1.1
1.2
1.3 1 10 100 1000 LWC (g/m3)
99.9% 99% 95% 90% 50%
17.4 nm i Averaging Distance (nm i) Figure M-1. Natural Proba bilities for LWC Averages in Stratiform Cloud Icing Conditions at 32oF to 14oF, With an Average MVD of 15 Pm.
M-1