5 characteristics tests, the applicant must show that its presence would have no effect on measured airplane performance and flight characteristics.
(e) Section 21.35(b)(2) requires the applicant to conduct sufficient flight testing the FAA finds necessary to determine whether there is reasonable assurance that the airplane, its components, and its equipment are reliable and function properly. Appendix 2 to this AC provides guidance for showing compliance with this requirement.
(f) Acceptable use of simulation in lieu of flight testing. It is difficult to establish guidance for using simulation in lieu of flight testing that applies in all situations. However, the following general principles can be used as guidance for determining the acceptability for using simulation in lieu of flight testing:
1 In general, flight test demonstrations are the preferred method to show compliance.
2 Simulation may be an acceptable alternative to flight demonstrations in certain situations, such as the following:
(aa) A flight demonstration would be too risky even after attempts are taken to mitigate these risks (e.g., by mock takeoffs/landings in the air at a safe altitude);
(bb) The required environmental or airplane conditions are too difficult to attain, such as in the examples listed below:
(i) validation of system safety analyses failure cases involving high crosswinds,
(ii) development of crosswind guidance for slippery runway operations; and
(iii) conditions involving minimum allowable weight where the minimum allowable weight cannot be achieved because of the weight of required test equipment.
In this case, simulation data can be used to supplement flight test data obtained at the minimum practicable test weight.
(cc) The simulation is used to augment a reasonably broad flight test program; or
(dd) The simulation is used to demonstrate repeatability, or to demonstrate performance of a specific scenario for a range of pilots.
3 Simulation Criteria. If it is agreed that a simulation will be used to establish compliance, to be acceptable for use in showing compliance with the performance and handling qualities requirements, the simulation should:
(aa) Be of a type and fidelity that is appropriate for the task. For example, is motion or an exterior view needed, or is the fidelity or customizability of an engineering simulator needed?
(bb) Be suitably validated by flight test data for the conditions of interest.
(i) This does not mean that there must be flight test data at the exact conditions of interest. The reason simulation is being used may be that it is too difficult or risky to obtain flight test data at the conditions of interest.
(ii) The level of substantiation of the simulator to flight correlation should be commensurate with the level of compliance (i.e., the closer the case is to being noncompliant, the higher the required fidelity of the simulation).
(cc) Be conducted in a manner appropriate to the case and conditions of interest.
(i) If closed-loop responses are important, the simulation should be piloted by a human pilot.
(ii) For piloted simulations, the controls/displays and cues should be substantially equivalent to what would be available in the real airplane (unless it is determined that not doing so would provide added conservatism).
(2) Section 25.21(c) - Proof of Compliance (Altitude Effect on Flight Characteristics).
(a) Any of the flying qualities affected by altitude, including controllability, stability, trim, and stall characteristics, must be investigated at the most adverse altitude conditions approved for operations.
(b) Consideration should be given in the test program to any aerodynamic control system changes that occur with changes in altitude (e.g., maximum control surface displacement or auto slats that may be inhibited by Mach number above a specific altitude).
(3) Section 25.21(d) - Proof of Compliance (Flight Test Tolerances).
(a) To allow for variations from precise test values, acceptable tolerances during flight testing must be maintained. The purpose of these tolerances is to allow for small variations in flight test values of certain variables from the targeted value. They are not intended for compliance tests to be planned for other than the critical condition, nor are they to be considered as an allowable measurement error.
(b) Where variation in the parameter for which a tolerance is allowed will have an effect on the results of the test, the results should be corrected to the most critical value of that parameter within the approved operating envelope. If such a correction is impossible or
7 impractical, the average test conditions should assure that the measured characteristics represent the actual critical value.
1 Weight limits. Figure 3-1, below, presents weight tolerances that have been found acceptable for the specified flight tests. Many flight tests need to be conducted at or very near the maximum operating weight for the airplane configuration, particularly those tests used to establish airplane flight manual (AFM) performance information. As noted in paragraph
(a) above, the purpose of the test tolerances is to allow for variations in flight test values, not to routinely schedule tests at less than critical weight conditions or to allow for compliance to be shown at less than the critical weight condition. In addition, the tolerances can be used to help determine when to interrupt a series of test conditions in order to refuel the airplane if necessary to remain within the acceptable weight tolerance.
Figure 3-1. Weight Tolerance Limits
Flight Test Conditions Weight Tolerance Limit ±5% ±10% Stall Speeds X Stall Characteristics X All Other Flight Characteristics X Climb Performance X Takeoff Flight Paths X Landing Braking Distance X Landing Air Distance X Takeoff Distance & Speed X Accelerate-Stop Distance X Maximum Energy RTOs X Minimum Unstick Speed X Minimum Control Speed X
NOTE: A -5 percent tolerance limit means that the weight for the particular test may be up to 5 percent less than the test target value. A +5 percent tolerance limit means that the weight for the particular test may be up to 5 percent higher than the test target value.
(aa) It can be difficult or impossible to conduct testing at the airplane’s minimum allowable weight with an airplane configured for conducting a flight test program. If the minimum weight cannot be obtained (within the specified tolerance limit) and compliance at the minimum weight cannot be clearly deduced from the results at the tested weight, the testing
8 should be conducted on a production airplane (or other airplane on which the minimum weight can be obtained). If the instrumentation or equipment needed to conduct safe testing cannot be installed on the production airplane configuration, or the weight of such instrumentation still prevents the minimum weight from being obtained, consider the use of simulation to extend the results obtained at the minimum practical test weight. (See paragraph 3a(1)(f).)
(bb) For follow-on airplane certification programs involving an increase in the maximum allowable gross weight, the test weight limits of Figure 3-1 have been applied as extrapolation limits on the original test data in order to minimize additional testing. For the test weight tolerance limits to be applied in this manner, the original test data must be from an existing certificated database for an aerodynamically similar model of the same airplane type.
The tolerance limit should be applied to the maximum weight at which the original testing was conducted, not to the maximum certified weight.
(cc) Equivalent weight extrapolation limits. For follow-on airplane certification programs where it is desired to increase a maximum operating weight based on existing certified performance parameters that have weight as one of their independent terms, those parameters should be examined for equivalent compliance with the weight tolerance limits of Figure 3-1. An example would be the reduction of an airplane’s landing flap position, to one approved on a similar model of the same airplane type, which would incur an increase in landing speeds and brake energy, relative to the original certificated landing flap, at any given weight.
The brake energy, at the maximum certificated landing weight, should be calculated for the reduced landing flap. This brake energy should account for the increased landing speeds and reduced aerodynamic drag associated with the reduced flap setting. It should then be determined what equivalent gross weight would have rendered that brake energy with the original landing flap. (See Figure 3-2 for an example of how this can be done.) If the resulting equivalent gross weight does not exceed the certificated maximum landing weight by more than the five percent weight extrapolation limit specified in Figure 3-1, the reduced flap certification may be eligible for a reduced flight test program (e.g., limited to stall speed verification, handling characteristics, and a qualitative landing demonstration). Further limitations may be imposed by the criteria of technical standard order (TSO) C135a (“Transport Airplane Wheels and Wheel and Brake Assemblies,” dated July 1, 2009).
9 Figure 3-2. Equivalent Weight Extrapolation
∆ G.W. Not greater than 5% Gross Weight - Pounds ∆ KE Max. Landing Wt.
“Equivalent” Gross Wt.
Flap Brake Energy Ori ginal Reduced Flap
2 Wind Limits. A wind velocity limit of 10 knots (from any direction) or
0.11 VSR1 (whichever is lower) is considered the maximum acceptable for obtaining valid takeoff and landing flight test data. Takeoff and landing performance data obtained under runway wind conditions greater than 5 knots may be inconsistent and unreliable because winds of that magnitude are likely to be unsteady. However, performance data obtained with winds between 5 and 10 knots should not necessarily be discarded. Their validity should be checked against data obtained in conditions with lesser winds. Wind velocity should be measured at the height of the wing mean aerodynamic chord (MAC), as determined with the airplane in a static ground attitude. When measuring test wind velocity at the wing MAC height, a height of six feet above the ground should be considered as a minimum measurement height to avoid possible measurement inaccuracies due to surface interference.
3 C.G. Limits. A test tolerance of +7 percent of the total c.g. range is intended to allow for inflight c.g. movement. This tolerance is only acceptable when the test data scatter is on both sides of the limiting c.g. or when adjusting the data from the test c.g. to the limit c.g. is acceptable. If compliance with a requirement is marginal at a test condition that is inside of the c.g. limits, the test should be repeated at the c.g. limits.
4 Airspeed Limits. Normally, tests conducted within 3 percent or 3 knots (whichever is the higher) of the desired test speed are considered acceptable.
5 Thrust/Power Limits. Thrust critical tests, such as minimum control speeds, should be conducted at the highest thrust (or power) allowable on the engine given the constraints of temperature and altitude. It is then permitted to calculate further corrections to allow extrapolation of data to cover the entire operating envelope. These thrust (or power) corrections should be limited to 5 percent of test day thrust (or power), unless a detailed analysis is performed.
(c) It is not the purpose of these tolerances to allow flights at values in excess of those authorized in the type design. If such flights are to be conducted, adequate structural
10 substantiation for the flight conditions should be available. These flights should always be conducted under controlled conditions and with the flight test crew’s full knowledge of the situation. Examples of such flights are:
1 Takeoff at greater than maximum takeoff weight to reach a test area at the maximum takeoff weight.
2 Landing at greater than maximum landing weights during the course of conducting takeoff tests.
3 Flights to obtain data for future approvals beyond that substantiated for the initial type design.
(d) The table in Figure 3-3 indicates the cases for which corrections are normally allowed. Any corrections to flight test data should be made by methods that are agreed to by the FAA.
Figure 3-3. Test Parameters That Normally Can Be Corrected
Flight Test Condition Correctable Parameters Wt. C.G. Airspeed Altitude Power/ Thrust Wind Airspeed calibration X --- --- --- --- --- Stall speeds X X --- --- X --- Climb performance X X X X X --- Landing performance X --- X X --- X Takeoff performance X X --- X X X Accelerate-stop perf. X X --- X X X Minimum control speed --- --- --- --- X --- Minimum unstick speed X X X --- X --- Buffet boundary X X --- X --- ---
(e) All instrumentation used in the flight test program should be appropriately calibrated and acceptable to the FAA test team.
(4) Section 25.21(f) - Proof of Compliance (Wind Measurement and Corrections).
The relationship between the wind measured at one height and the corresponding wind at another height may be obtained by the following equation:
11 VW2 = VW1(H2/H1)1/7
Where: H = Height above the runway surface VW2 = Wind velocity at H2 VW1 = Wind velocity at H1
This equation is presented graphically below. Values of H less than 5 feet should not be used in this relationship.
0 10 20 30 40 50 60 70 0 10 20 30 40 50 60
WIND VELOCITY - KNOTS
HEIGHT ABOVE FIELD ELEVATION - FEET 10 meters
Figure 3-4. Wind Profile Variation
(5) Wind Profile Variation for Test Data. The performance data of airplanes should be obtained in such a manner that the effect of wind on the test data may be determined. The test wind velocity should be corrected from the recorded height above the test surface to the height of the airplane wing mean aerodynamic chord. If the wind profile variation is not measured, the variation may be calculated using the equation in paragraph (4) above. The following examples are methods of handling wind profile variation data. Other methods have also been found acceptable.
Example: Test Data
Given:
12 o Height of mean-aerodynamic-chord with airplane on surface 8.0 ft.
o Height of wind measurement 6.0 ft.
o Measured wind velocity 4.8 kts.
Results:
o Test wind velocity with airplane 50 ft. above landing surface
4.8((50 + 8)/6)1/7 = 6.6 kts.
o Test wind velocity with airplane 35 ft. above takeoff surface
4.8((35 + 8)/6)1/7 = 6.4 kts.
o Test wind velocity with airplane on surface
4.8(8/6)1/7 = 5.0 kts.
(6) Wind Profile Variation for AFM Data. When expanding the data to the AFM conditions, the result should include the effective velocity, at the airplane’s wing mean aerodynamic chord, which corresponds to the wind condition as measured at 10 meters (32.81 ft.) above the takeoff surface, and corrected for the wind factors of § 25.105(d)(1).
Example: AFM Data
Given:
o Height of mean aerodynamic chord with airplane on surface 8.0 ft.
o Reported headwind at 10 meters 40.0 kts.
o Section 25.105(d)(1) wind factor 0.5
Results:
o Factored wind velocity with airplane 50 ft. above landing surface
(0.5)(40)((50 + 8)/32.81)1/7 = 21.7 kts.
o Factored wind velocity with airplane 35 ft. above takeoff surface
(0.5)(40)((35 + 8)/32.81)1/7 = 20.8 kts.
o Factored wind velocity with airplane on surface
(0.5)(40)(8/32.81)1/7 = 16.3 kts.
(7) Airplane Airspeed Variation Due to Wind Profile Variation Combined With Speed Changes Due to Airplane Dynamic Performance. In the reduction of test data and in the expansion of such data to AFM conditions, the increase or decrease of speed due to the dynamic effect of the forces on the airplane are shown only by the change in ground speed. These
13 changes in ground speed may be generalized either as speed increments or speed ratios. The changes in airspeed due to wind profile variation are superimposed on these speed changes.
Example: Determination of True Airspeed from Ground Speed -- Takeoff Test Data
Given:
o Ground speed at liftoff, VLOF 139.0 kts.
o Ground speed at 35 ft. above takeoff surface 140.6 kts.
o Speed change due to airplane dynamic performance 1.6 kts.
o Test headwind at liftoff 5.0 kts.
o Test headwind with airplane 35 ft. above takeoff surface 6.4 kts.
Results:
o True airspeed at liftoff, VLOF 139.0 + 5.0 = 144.0 kts.
o True airspeed at 35 ft. above takeoff surface 140.6 + 6.4 147.0 kts.
Example: Determination of Rotation Speed from True Airspeed at 35 ft. Height -- AFM Data
Given:
o Factored headwind at liftoff 16.3 kts.
o Factored headwind with airplane 35 ft. above takeoff surface 20.8 kts.
o Ground speed change, (V35-VLOF) 1.6 kts.
o Ground speed change, (VLOF-VR) 0.5 kts.
o True airspeed required at 35 ft. 150.0 kts.
Results:
o Ground speed required at 35 ft. 150 - 20.8 = 129.2 kts.
o Ground speed at liftoff 129.2 - 1.6 = 127.6 kts.
o True airspeed at liftoff 127.6 + 16.3 = 143.9 kts.
o Ground speed at rotation 127.6 - 0.5 = 127.1 kts.
o True airspeed at rotation (for AFM speed and distances) 127.1+16.3 = 143.4 kts.
14 Example: Landing — AFM Data
Given:
o Factored headwind with airplane 50 ft. above landing surface 21.7 kts.
o Factored headwind with airplane on landing surface 16.3 kts.
o Ground speed change for 50 ft. to touchdown (V50 – VTD) 4.0 kts.
o True airspeed required at 50 ft. 130 kts.
Results:
o Ground speed at 50 ft. 130 - 21.7 = 108.3 kts.
o Ground speed at touchdown 108.3 - 4.0 = 104.3 kts.
o True airspeed at touchdown 104.3 + 16.3 = 120.6 kts.
(8) Expansion of Takeoff and Landing Data for a Range of Airport Elevations.
(a) These guidelines apply to expanding AFM takeoff and landing data above and below the altitude at which the airplane takeoff and landing performance tests are conducted.
(b) Historically, limits were placed on the extrapolation of takeoff data. In the past, takeoff data could generally be extrapolated 6,000 feet above and 3,000 feet below the test field elevation when proven testing and data reduction methods were used. For extrapolations beyond these limits, a 2 percent takeoff distance penalty was to be applied for every additional 1,000 feet extrapolation. Such limitations were generally not applied to extrapolation of landing data, provided the effect of the higher true airspeed on landing distance was taken into account.
(c) Since then, considerably more experience has since been gained both in terms of modeling airplane and propulsion system (i.e., turbine engines and propellers, where appropriate) performance and in verifying the accuracy of these models for determining high (and low) altitude takeoff and landing performance. This experience has shown that the soundness of the extrapolation is primarily a function of the accuracy of the propulsion system performance model and its integration with the airplane drag model. The basic aerodynamic characteristics of the airplane do not change significantly with altitude or ambient temperature, and any such effects are readily taken into account by standard airplane performance modeling practices.
(d) As a result, with installed propulsion system performance characteristics that have been adequately defined and verified, airplane takeoff and landing performance data obtained at one field elevation may be extrapolated to higher and lower altitudes within the limits of the operating envelope without applying additional performance conservatisms. It should be noted, however, that extrapolation of the propulsion system data used in the determination and validation of propulsion system performance characteristics is typically limited to 3,000 feet above the highest altitude at which propulsion system parameters were evaluated for the pertinent power/thrust setting. (See paragraph 9 of this AC for more information on an
15 acceptable means of establishing and verifying installed propulsion system performance characteristics.)
(e) Note that certification testing for operation at airports that are above 8,000 feet should also include functional tests of the cabin pressurization system in accordance with paragraph 87b(3) of this AC. Consideration should be given to any sensitivity to, or dependency upon airport altitude, such as: engine and auxiliary power unit (APU) starting, passenger oxygen, autopilot, autoland, autothrottle system power/thrust set/operation.
(9) Tailwind Takeoff and Landing.
(a) Wind Velocities of 10 Knots or Less - Approval may be given for performance, controllability, and engine operating characteristics for operations in reported tailwind velocities up to 10 knots without conducting additional flight tests at specific wind speeds.
(b) Wind Velocities Greater than 10 Knots.
1 Performance. It is considered that takeoff, rejected takeoff, and landing distances, measured in tailwind conditions greater than 10 knots, are unreliable for use in determining airplane performance. Wind conditions of such magnitude are generally not sufficiently consistent over the length of the runway or over the time period required to perform the test maneuver. The 150 percent operational tailwind factor, required by §§ 25.105(d)(1) and
25.125(f), provides a satisfactory level of safety for operation in tailwinds up to 15 knots when using AFM data based on flight tests in nominally calm wind conditions.
NOTE: The design requirements of § 25.479 (Level landing conditions) also require the effects of increased contact speeds to be investigated if approval for landings with tailwinds greater than 10 knots is desired.
2 Control Characteristics. The test tailwind velocity for demonstrating handling qualities should be equal to the proposed limit tailwind factored by 150 percent. The intent of the 150 percent factor is to provide adequate margin for wind variability in operations, including currency of the wind data, averaging of the data by the measuring and reporting method, and the highly variable nature of higher wind conditions. Therefore, the test wind condition of 150 percent of the proposed tailwind limit should be an averaged or smoothed wind speed, not a peak wind speed. Airplane control characteristics should be evaluated under the following conditions with the c.g. at the aft limit and the test mean tailwind velocity equal to the proposed limit tailwind factored by 150 percent:
(aa) Takeoff. Both all-engines-operating and one-engine-inoperative
(i.e., with a simulated failure of the critical engine at the engine failure speed, VEF) takeoffs should be evaluated at a light weight with maximum approved takeoff flap deflection.
(bb) Landing. Approach and landing at light weight with maximum approved landing flap deflection.
(cc) Determination of the increased ground speed effect on gear vibration or shimmy, and flight director, or autopilot instrument landing system (ILS) approaches, terrain awareness warning system (TAWS) sink rate modes, etc.
(dd) If engine idle power or thrust is increased to account for the increased tailwind velocity, ensure that deviations above the glideslope are recoverable.
3 Weight Limits. Consistent with the requirements of §§ 25.105(d)(1) and
25.125(f), the maximum takeoff and maximum quick turnaround weights should be determined using brake energies and tire speeds, as appropriate, calculated with the limit tailwind velocity factored by 150 percent.
4 Engine Operating Characteristics. Satisfactory engine operation should be demonstrated at the limit tailwind velocity factored by 150 percent. The demonstrations should include:
(aa) Zero groundspeed operation.
(bb) Takeoff power or thrust setting procedure used for AFM performance (typically completed by approximately 80 knots), both manually and automatically (autothrottle).
(cc) Reverse thrust operations.
5 Airplane Flight Manual. The AFM should contain a statement that the limitation for tailwinds greater than 10 knots reflects the capability of the airplane as evaluated in terms of airworthiness but does not constitute approval for operation in tailwinds exceeding 10 knots.
b. Procedures.
(1) The performance-related flight test procedures are discussed in each of the following paragraphs of this AC:
10. Takeoff and Takeoff Speeds
11. Accelerate-Stop Distance
12. Takeoff Path
13. Takeoff Distance and Takeoff Run
14. Takeoff Flight Path
15. Climb: General
16. Landing Climb
17. Climb: One Engine Inoperative
18. En Route Flight Path
19. Landing