repeated encounters should be considered part of this assessment. Any damage findings should be brought to the attention of the FAA for approval.
(6) High Vibrations. You must ensure that high vibrations are assessed for the specific aircraft installation. The acceptability of high vibrations is evaluated by the airframer. Bring high vibrations to the attention of the ACO or ECO approving the engine; it will contact the ACO approving the engine installation before the final approval, and inform them of this condition. Any high vibrations should be included in the installation manual by the engine manufacturer.
(7) Communication of Results. The installation and operating manuals required by § 33.5 should provide information describing all engine conditions observed during engine certification icing tests. Prior to these tests, the engine manufacturer should provide the FAA with a process to evaluate the acceptability of any potential damage that could occur during icing tests.
(a) Natural icing flight tests are conducted to demonstrate compliance with §§ 23.1093,
25.1093, 27.1093, or 29.1093; they require an FAA preapproved process for evaluating damage resulting from icing tests. Also, if periodic engine power run-ups are used to minimize icing damage during the ground icing operation demonstration defined in § 33.68(d), then the applicant must document the run-up procedure. Complete documentation must contain a description of the run-up requirements and the required run-up intervals, and it must be contained in the operation manual as mandatory within icing conditions. Prior to demonstration, the engine manufacturer should coordinate the proposed run-up shedding procedures with the airplane manufacturer, so as to assure that procedures are practical and appropriate for the intended operations.
(b) Any acceptable power loss anomalies from for example, ice accretion, ice shed, water runback and refreeze and shed, and the effects on performance and operation, should be documented in the installation manual. Both the engine certifying ACO and the installing ACO should carefully evaluate any high vibrations induced from ice accretions during ice testing.
These vibrations should be documented by the engine manufacturer in the engine’s installation manual.
o. Engine Systems. Applicants may use an automatic engine control system to initiate an ice protection system to meet the requirements defined in § 33.68, provided that system operation does not result in potential crew action. Examples of engine characteristics that may be noticeable to the flight crew are exhaust gas temp fluctuations, large speed fluctuations, or audible surging. Further, any engine control system required for engine ice protection certification under part 33 should not react adversely with other aircraft systems, aircraft handling qualities and performance, or human factor considerations.
(1) Any unacceptable adverse interactions may result in the engine being deemed uninstallable under parts 23 or 25 airplanes or parts 27 or 29 rotorcraft. Critical issues which you must address include: crew interface, uncommanded thrust or power changes, thrust or power setting, asymmetric engine behavior, pilot workload and appropriate flight deck indication and procedures, the effect on aircraft handling, pilot ability, and human factors. Additionally, any
engine system required to show compliance under § 33.68 should meet the following requirements:
(a) System Reliability. Demonstrate the capability of the system for reliably sensing the conditions, which enables the engine system function, throughout the approved operating envelope.
(b) Dispatch. The engine system function should be available for all dispatch configurations. The system should be configured in its most critical dispatch state for certification icing tests.
(c) Electronic Faults. Applicants should demonstrate that engine system functions will not be lost due to any probable electronic fault(s).
(d) Other Environmental Testing. Engine system functions should not be affected when the system and any associated electronic systems are exposed to the required operating environments, including high intensity radiated fields (HIRF) and lightning.
(e) Power Requirements. For those systems powered solely with a dedicated engine alternator (either directly or using another engine system such as a FADEC), applicants should demonstrate the system sensing and performance function at minimum certified rotor speeds.
Minimum certified engine speed is the minimum idle rotor speed achievable anywhere in the icing envelope.
p. Auto-Recovery Systems. Auto-recovery systems should not be needed under § 33.68 testing since these icing conditions are considered to be within the engine’s certified operational envelope and a probable encounter. The intent of § 33.68 is to certify engines to perform and operate reliably in the icing conditions depicted in appendices C and O of part 25, appendix C of part 29, and appendix D of part 33, as applicable.
(1) Auto-recovery systems are considered back-up devices that are only needed following rare ice ingestion events resulting from icing conditions outside appendices C and O of part 25, appendix C of part 29, and appendix D of part 33. If auto-recovery systems activate during the icing test, then the applicant should alert the installing ACO of the activation. Auto-recovery systems are not the primary protection for continued safe engine operation during normal ice sheds or accretion while operating in icing conditions depicted in appendices C and O of part 25, appendix C of part 29, and appendix D of part 33.
(2) Therefore, applicants may perform the § 33.68 compliance testing with auto-recovery systems enabled, but the systems should not activate during the § 33.68 test sequence.
Additionally, continuous ignition should not be selected during the § 33.68 compliance testing.
To assure non-activation of an enabled auto-recovery system, the applicant should have displayed an instrumented signal that monitors the auto-recovery system activation. If activation monitoring cannot be accomplished, then disabling the auto-recovery system will be necessary.
q. Operation Instructions. Any operating procedure (for example, ground run-up procedures) required to ensure continued operational compliance with ground icing conditions or falling and blowing snow evaluated under § 33.68(d), and the installation requirements of §§ 23.1093(b)(1)(ii), 23.1093(b)(2), 25.1093, 27.1093(b)(1)(ii) or 29.1093(b)(1)(ii), should be communicated to the installer in the operating instructions as a requirement. The requirement for each applicable operating procedure should further be included in the limitation section of the airplane flight manual. It may be necessary to coordinate with the installer on these procedures to ensure that they can be effectively implemented in-service. The installer may translate the operating instructions into a procedure tailored at the airplane level if it results in an operation that is equivalent to the demonstrated operating procedure. An equivalent operation should be conservatively representative of the demonstrated operating procedure.
r. Special Considerations for Mixed Phase and Glaciated Atmospheric Conditions (part 33 turbofan and turboprop only). Part 33 requirements for the type certification approval of engines operating in an ice crystal environment are located in § 33.68, and appendix D of part 33. These requirements have been developed in response to service events. The root cause of these events can be traced to ice buildup within the core flow path of the affected engines. Engine inlet probe blockage and the resulting signal corruption should also be analyzed. Adverse effects created by this type of icing on the engine include un-commanded rollback of power or flameout, as well as compressor stall and core hardware damage.
(1) Ice crystals have only been recognized as a threat to turbine engines in recent years. In response, the FAA has worked with industry to develop standards that address this threat.
During this process, we found that although capabilities are developing, the present ice crystal tools and test techniques have not been fully developed and validated sufficiently. Therefore, we developed a phased-in approach to address the ice crystal threat during an engine type certification program. The following paragraphs describe this phased-in approach.
(2) New engines must address the known in-service experience in ice crystal environments (for example, core damage and engine flameout events). Until ice crystal tools and test techniques have been developed and validated, the engine manufacturer should use a comparative analysis to specific field events. This analysis approach should show that new engine cycle or design features, or both, will result in acceptable engine operation.
(3) Acceptable operation includes the absence of rollback, rundown, stall, flameout, and unacceptable compressor blade damage as described earlier in paragraph 8.a of this AC.
Additionally, we recommend the engine manufacturer incorporate developed and validated technology as it becomes available, and work toward a full engine test substantiation of operation in ice crystal environments.
(4) Long-term, we anticipate an acceptable demonstration will eventually include a CPA of an ice crystal environment as depicted in appendix D of part 33. All engine power levels, including in-flight idle operation, will be evaluated in these conditions. The critical conditions will be demonstrated to the FAA through a combination of testing and validated analysis using the latest tools and technology when proposing the compliance methodology. Computational
tools to be used in this analysis would be calibrated by either rig calibration test data or engine test measurements.
s. Comparative Analysis Guidance for Ice Crystal Icing (part 33 turbofan and turboprop only). As stated earlier, until ice crystal tools and test techniques have been developed and validated, the engine manufacturer should use a comparative analysis to specific field events.
This analysis should show the new engine cycle or design feature, or both, would result in an acceptable engine operation when subjected to the ice crystal environment depicted in appendix D of part 33. This comparative analysis should consider both suspected susceptible design features, as well as mitigating design features.
(1) Susceptible Design Features. These features could include:
(a) Stagnation points which could provide an increased accretion potential.
(b) Exposed core entrance (as opposed to hidden core).
(c) High turning rates in the inlet, in the booster and core flowpath (particularly compound turning elements).
(d) Protrusions into the core flowpath (for example, bleed door edges and measurement probes).
(e) Unheated surfaces on booster and front core stages.
(f) Narrow vane-to-vane circumferential stator spacing leading to a small stator passage hydraulic diameter.
(2) Mitigation Features. These susceptible design features can be significantly mitigated by one or more of the following design features. Mitigating design features could include:
(a) Heated surfaces in the fan, booster, and forward core compressor stages.
(b) Elevated rotor speeds.
(c) Hidden core entrance.
(d) Low frontal cross-sectional area on flowpath probes.
(e) Inter-compressor bleed scheduling to remove both the ice crystal media and any upstream shedding, from the flowpath.
(f) Circumferential spacing of stators set to enhance tolerance to ice blockage (generally denoted by the hydraulic diameter of the stator passage).
(g) Increasing core compressor airfoil tolerance to soft-body damage. Soft-body damage is typically airfoil bending, whereas hard-body damage results in tearing and fracture.
(3) Similarity to Engines Proven Safe to Operate in Mixed Phase or Glaciated Atmospheric Conditions (part 33). Although ice crystal icing conditions are hazardous to turbine engine operation, severe incidents involving this type of meteorological condition are not common. Many currently certified engine designs have been proven by their field service experience to be safe to operate in these conditions. New engine designs that are similar to those of proven engine designs are allowed to show compliance by comparative analysis.
(a) Several steps are required to demonstrate compliance by similarity analysis. First, you should identify the baseline (certified) engine, and supply evidence that this engine is safe to operate in mixed phase or glaciated icing conditions. This evidence can be field service experience and/or a certification report.
(b) Second, identify the icing-pertinent engine features that may influence mixed phase or glaciated ice accretion within the target (certifying) engine; then compare them to the baseline engine. This comparison should establish that the new engine model is less than or equally susceptible to icing as the baseline model.
(4) Comparative Analysis Versus CPA. A comparative severity analysis should be performed to show the operational envelope of the target engine does not make it more susceptible to ice crystal icing than the baseline engine. As the icing community’s knowledge improves through research, in the long term applicants should perform a CPA to fully assess ice crystal icing. This severity analysis should consider both environmental and engine operational effects on accumulation, accretion locations, as well as ice shedding.
(5) Innovative or Novel Designs. If the new engine cycle or design feature, or both, contains innovative design ideas such that a comparative analysis with current engines and specific engine events is not possible, then you should demonstrate that the innovative feature(s) will not be susceptible to any adverse effects of operating in an ice crystal environment. Such a demonstration entails a two part process. In part one you should document the physical basis regarding why this design would result in acceptable operation in ice crystals. In part two you should generate physical evidence to substantiate the claims of part one. For purposes of substantiation, water particles may be used instead of ice crystals. However, data collected with water particles should be corrected to account for ice crystal characteristics, including liquid to solid phase thermodynamic effects. The use of water to simulate crystals will be re-evaluated by the FAA as ice crystal test techniques improve through research and development.
Table A. Example Steps That May Be Used
Part 1 Part 2 Discuss consequences of ice accretion. Identify test article (engine or component).
Identify design features that promote tolerance Establish test conditions promoting ice to critical ice accretion. accretion (with allowance for use of liquid water if testing with ice crystals is not feasible).
Discuss consequences of ice crystals on Evaluate tolerance of new design features over promotion of ice accretion. a range of test points.
t. Compliance Considerations (part 33). Engine icing events in an ice crystal environment appear to be associated with accretion of ice deeper in the compressor, as compared to supercooled droplet environments where accretion occurs farther forward.
u. Ice Crystal Accretion Explanation. The accretion process is believed to be a result of ice crystals passing through the engine inlet and into the front compressor stages. This process continues until the local conditions within the engine are conducive to forming a liquid layer on vane and blade surfaces. Once the local conditions are conducive to the formation of a liquid layer, the vane and blade surface condition allows impinging crystals to melt and stick to the liquid layer and accrete on the surface. This is a combination of conductive and convective heat transfer from ice melting, evaporation, and surface contact.
(1) The combination of both the liquid water and solid ice crystal draw significant heat away from the impinged surface. An impingement surface that was initially as high as 1200 Fahrenheit (490 Celsius) can be cooled down to 320 Fahrenheit (00 Celsius) by this dual phase ice and water media on the surface.
v. Ice Crystal Icing Assessment (part 33, turbofan and turboprop only). You should assess engine operation across the whole icing envelope. Cruise, hold, and descent power settings should be evaluated in mixed phase and glaciated atmospheric conditions as depicted in appendix D of part 33. Note that no established standardized CPA for ice crystal icing currently exists. However, to aid you, we are recommending some possible CPA point selections described in Table B below. We also recommend applicants use the associated general criteria when selecting proposed conditions to be evaluated.
Table B. General Criteria for Selecting Conditions to be Evaluated
Service History This suggests that a selection of conditions at high ambient temperatures depicted in appendix D of part 33 icing envelope, are consistent with higher TWC levels. However, lower temperatures might push the accretion aft, where there may be a more critical ice accretion site with lower TWC. Therefore, both high and low temperatures within the envelope should be evaluated.
Power Levels To promote ice crystal melting, power levels with internal total air temperature within the core flowpath should be between 320 Fahrenheit (00 Celsius) and approximately 1200 Fahrenheit (490 Celsius). The power level adjusts the accretion site forward or aft within the engine.
High Altitude High altitude (low air density) allows greater ice accretion mass before shedding from static surfaces.
(1) The operating conditions for this evaluation should be chosen from appendix D of part 33, covering the flight phases to include climb, cruise, idle-descent, and holding. The water contents depicted in appendix D represent the level for a standard exposure distance of 17.4 nautical miles. To adjust the water content level as required, so that it reflects the expected icing exposure period, use the distance scale factor depicted in appendix D. Service experience suggests that straight line exposure distances of 20 to 80 nautical miles may be encountered, as well as holding in cloud. As our understanding of water content in relation to distance flown becomes more precise, this encounter distance may be adjusted. Currently we hold that applicants who test to this greater distance will assure their engines are robust and meet the minimum standards of the regulations.
(2) As noted earlier, ice crystals promote icing at engine sites rearward and at higher local air temperatures than would exist with only supercooled liquid. These rearward internal flowpath accretion sites will also have higher air loads, which can limit ice accretion due to shedding. Therefore, the CPA should consider not only ice accretion, but also the likelihood of ice shedding.
w. Appendix O of Part 25, Water Impingement.
(1) The water impingement region may become greater when encountering the large water droplets depicted in appendix O of part 25. This occurs because larger droplets have greater inertia and follow a more ballistic trajectory less prone to influence by local airflow streamlines that bend to deflect air around an object. A way to illustrate this droplet inertia effect is through the use of the modified inertia parameter variable, Ko , as shown in Figure 1 of this AC (see reference g, in section 5 of this AC. This parameter is a relationship based on the momentum of the droplet, air viscosity, and size scale of the impingement region (with a correction for Reynolds number). It is used to correlate the streamline effect of the airflow in deflecting a droplet and preventing it from impinging on an object. The equation for Ko is provided below 31
for a droplet with diameter (d) with an upstream velocity (Vo) traveling in an airflow with a viscosity (µ).
(2) The impingement efficiency Em is simply a decimal fraction of the water droplets that are aligned with an object far upstream that actually do impinge on the surface. For very large droplets, impingement efficiency approaches unity and is progressively less for smaller droplets.
Published correlations exist for various object shapes of interest for ice impingement. Such a correlation between Ko and impingement efficiency Em, is shown below in Figure 1 for several object shapes.
Figure 1. Droplet Impingement Efficiency for Different Shapes
Figure 1a. Definition of Inertia Parameter, Equation 3.10 from Reference 5.g of this AC
Inertia Parameter K = 2(d/2)2 ρH20 V0 (Dimensionless) EQU 3.10 9 LC µ Where Characteristic length LC of object is:
Airfoils LC = Chord Cylinders LC = Radius Spheres LC = Radius Cones L C = Base Radius
(3) Lower impingement efficiencies (less than a value of one) are indicative of droplet deflection from the object. When droplets are deflected, they don’t accrete on the object. This deflection is caused by streamlines moving around the object. For most shapes, a modified inertial parameter of 10 or greater implies a ballistic path for the droplets. A ballistic path means the droplet does not follow stream lines and can impinge the vane farther back than a nonballistic droplet. This results in ice accretion farther back on the vane.
(4) Impingement of a supercooled water droplet on a surface is dependent on the droplet size, and the size of the geometric feature. The smaller the drop size and the bigger the feature, the more likely the drop will follow the aerodynamic streamlines around the body and fail to impinge. Larger drops, with more inertia are less likely to follow streamlines, and therefore they will impinge on larger features. In the engine, such features include the inlet and spinner. The applicant must account for the potential impingement of larger drops on engine geometric features of this scale. Figure 2 of this AC illustrates the consequences of encountering the larger water droplets depicted in appendix O of part 25. Figure 2 also shows what happens with target objects of typical size ranges for larger transport engines. The shaded area in Figure 2 covers appendix C of part 25 and appendix C of part 29, as applicable. For a given modified inertia parameter value (Ko) of 10, the droplet size range is seen to behave ballistically for small “target” objects, such as engine airfoils, airfoil cascades, and probes. Consequently, large droplets as depicted in appendix O of part 25 will not significantly affect surface water impingement and resulting ice accretions on small airfoils within engines. For larger targets, however, applicants should anticipate greater impingement on surfaces, for example: the nacelle and engine spinner. Other surfaces unprotected by surface heating systems are also at risk for water impingement by the sized droplets depicted in appendix C of parts 25 and 29, as applicable.
Figure 2. Ballistic Boundary as a Function of Target and Droplet Size
10 100 1000 10000 Droplet Diameter - microns Airfoils Probes Nacelle Lip Spinner Inlet Diameter Cascades .01 .1 1 10 “Target” Dimension - ft Ballistic (Minimal Deflection by Streamlines) Locus of Modified Inertia Parameter = 10 Droplet Size Range of Appdx C Ko > 10 (example) Ko = 10 Ko < 10 (example) 10 100 1000 10000 Droplet Diameter - microns Airfoils Probes Nacelle Lip Spinner Inlet Diameter Cascades .01 .1 1 10 “Target” Dimension - ft Ballistic (Minimal Deflection by Streamlines) Locus of Modified Inertia Parameter = 10 Droplet Size Range of Appdx C Ko > 10 (example)Ko > 10 (example) Ko = 10 Ko < 10 (example) 34
10. Section 33.77, Ice Slab Ingestion.
a. Why the Ice Slab Ingestion Test is Conducted. The intent of the ice slab ingestion test is to demonstrate tolerance to ice ingestion due to shedding from inlet (nacelle) surfaces in the front of the engine. This test also establishes limits for ice released from other aircraft or rotorcraft surfaces during parts 23, 25, 27, and 29 certifications. The dimensions of the test slab are related to engine size (defined by inlet highlight area), based on service experience. Manufacturers of turboshaft engines with complex inlet designs should propose an equivalent inlet highlight area once they have defined the ice slab size. Engine manufacturers should also include in their compliance plan, an analysis of the potential installation effects of the engine induction system.
(1) The engine manufacturer and the installer should closely coordinate the ice slab sizing and density. This coordination ensures that ice accumulation sites on the airframe which can potentially be ingested by the engine are addressed under § 33.77. Airframe locations that should be considered include for example, the inboard section of the wing for an aft fuselage mounted engine, the radome, and antenna. Ice slab ingestion is demonstrated under § 33.77, but must be addressed and shown as acceptable for the installation under §§ 23.901(d)(2), 23.1093,
25.1091(e), 25.1093, 27.1093, or 29.1093 (see paragraph 11 of this AC).
(2) The airframer responsible for the induction system should assess the inlet cowl ice accumulations in accordance with §§ 23.901(d)(2), 23.1093, 25.1093, 27.1093, or 29.1093.
They should also provide pertinent test variables to the engine manufacturer for incorporation into a § 33.77 test demonstration. If an application or product inlet has not been selected at the time of engine certification, the engine manufacturer should include in the engine installation manual all pertinent inlet icing assumptions, test data, and results. See paragraph 11 of this AC for installer assessment guidance.
b. Compliance Considerations. Compliance may be demonstrated by the standard engine ice slab test or by using a validated analysis procedure that uses equivalent soft body testing. The test demonstration should use ice slab trajectories aimed at critical engine locations. Applicants should pick locations based on the ice accretion and shed characteristics of the induction system likely to be installed on the engine. The most critical impact location should be tested.
(1) Slab Dimension Considerations. If the engine manufacturer type certificate applicant lacks specific knowledge of the icing characteristics of an aircraft inlet system, it may select test conditions typical of an installation already in-service, in combination with conservative assumptions on the future installation. The ice slab size, thickness, and density defined in the § 33.77 compliance demonstration should be evaluated against parts 23, 25, 27, and 29 engine installation and inlet system icing requirements. Applicants should use a minimum ice slab density equivalent to a 0.9 specific gravity, unless a different value is considered more appropriate (see AC 20-73A, AC 23.1419-2D, AC 25.1419-1, AC 27-1B, or AC 29-2C (or latest revisions) for more guidance on ice shedding).
(2) Engine Test Considerations. Applicants should determine if the ice slab size, thickness, and density are appropriate for the specific engine installation. If appropriate, the test
results of § 33.77 might be used by the airframe manufacturer to comply with the natural icing flight test requirements of §§ 23.1093, 25.1093, 27.1093, and 29.1093. Execution of the ice slab ingestion test typically involves targeting the slab to enter the air stream ahead of the fan, intact, at the outer diameter of the inlet duct. This is intended to mimic the ice release from the inlet and results in impact on the outer diameter of the fan.
c. Validated Analysis with Equivalent Soft Body Tolerance Testing. Compliance may also be shown by a validated analysis procedure that uses appropriate soft body damage testing. If the applicant elects this alternate compliance approach, the engine will be certified for the minimum standard ice slab consistent with the engine inlet area as defined in § 33.77. We recognize that alternate soft body damage testing may, in some circumstances, include objects that are larger than the standard ice slab based on inlet area. However, certification of an ice slab larger than the standard size by this validated analysis method is not currently allowed.
d. Elements of a Validated Analysis. This analytical model may be used alone or in conjunction with the results of a certification medium bird or other soft body ingestion test. A validated analysis must contain sufficient elements to show compliance. These elements may include:
• Full fan (fan engines) or first stage compressor (non-fan engines) blade modeling using the latest techniques such as finite element analysis;
• Blade material properties for yield or failure, or both, as appropriate;
• Dynamic and time variant capability;
• Thrust or power variance prediction if required to account for blade or other damage;
and
• Appropriate engine or component testing, or both, with impact at the outer 1/3 of the first stage blade span location. The fan is the first stage blade row for turbofan engines.
(1) The analysis of the ice slab impact on the fan must properly account for critical controlling parameters:
• Relative kinetic energy normal to the leading edge chord;
• Incidence angle – relative slab speed and blade speed;
• Slab dimensions; and
• Slab orientation.
(2) Any predicted power loss or blade damage (distortion, cracking, tearing) must be assessed against the criteria of this AC. Figures 3 and 4 of this AC help describe the contribution of these parameters when establishing the threat to the blade.