AC 20-146A

U.S. Department of Transportation Federal Aviation Administration Advisory Circular Subject: Methodology for Dynamic Seat Certification by Analysis for Use in Parts 23, 25, 27, and 29 Airplanes and Rotorcraft Date: 06/29/2018 Initiated By: AIR-600 AC No: 20-146A

This advisory circular (AC) sets forth an acceptable means, but not the only means, for demonstrating compliance with title 14 of the Code of Federal Regulations (14 CFR) §§ 23.562,

25.562, 27.562, and 29.562, as well as Technical Standard Order (TSO) TSO-C127a, and TSO-C127b. This AC includes guidance for certifying seats by computer modeling analysis techniques that are validated by dynamic tests. This AC defines the acceptable applications, limitations, validation processes, and minimum documentation requirements involved when substantiation by computer modeling is used to support a seat certification program.

If you have any suggestions for improvements or changes, you may use the Advisory Circular Feedback form at the end of this AC.

Dr. Michael C. Romanowski Director, Policy and Innovation Division Aircraft Certification Service

Contents Paragraph Page ii

1 Purpose1

2 Applicability1

3 Cancellation2

4 Background2

5 Related Documents3

6 Definitions5

7 Verification of Explicit Codes7

8 Computer Model Validation10

9 Application of Computer Modeling in Support of Dynamic Testing20

10 Application of Computer Modeling instead of Dynamic Testing21

11 Seat Certification and Coordination Process22

12 Documentation Requirements for Compliance24

Appendix A. Occupant Trajectory and Head Injury Criteria (HIC)A-1

Appendix B. Test-Simulation Comparison MethodologyB-1

Contents (continued) iii Figures Number Page

Figure 1. Mesh Refinement Example9

Figure 2. Schematic Diagram of Building Block Approach14

Figure A-1. Test Setup (xz-plane)A-2

Figure A-2. Numerical Model Setup (xz plane)A-3

Figure A-3. Comparison of Head Path (xz plane)A-5

Figure A-4. Comparison of Head Resultant AccelerationA-6

Figure B-1. Test and Simulated Results from Dynamic Tests: Example 1B-2

Figure B-2. Test and Simulated Results from Dynamic Tests: Example 2B-4

Figure B-3. Motion Data Offset ExampleB-5

Tables Number Page

Table 1. Example Peak Lumbar Loads17

Table 2. Example HIC Values18

1 1 PURPOSE.

This AC sets forth an acceptable means, but not the only means, for using computer modeling analysis techniques validated by dynamic tests to demonstrate compliance with §§ 23.562, 25.562, 27.562, and 29.562, Emergency landing dynamic conditions.

Hereafter, these regulations are referred to as “§ 2X.562,” as well as the applicable dynamic strength and occupant protection minimum performance standard (MPS) specified in TSO-C127, TSO-C127a, and TSO-C127b, hereafter referred to as “TSO-C127x.” This AC provides guidance on how to validate the computer model and under what conditions the model may be used in support of certification, TSO approval, or authorization.

2 APPLICABILITY.

2.1 The guidance in this AC is applicable to transport category aircraft for which an applicant is seeking approval pursuant to § 25.562 or TSO-C127x. This guidance is for aircraft manufacturers, modifiers, foreign regulatory authorities, and Federal Aviation Administration (FAA) aircraft certification engineers and designees.

2.2 The material in this AC is neither mandatory nor regulatory in nature and does not constitute a regulation. It describes acceptable means, but not the only means, for showing compliance with the applicable regulations. The FAA will consider other means of showing compliance that an applicant may elect to present. While these guidelines are not mandatory, they are derived from extensive FAA and industry experience in determining compliance with the relevant regulations. If circumstances arise whereby following this AC would not result in compliance with the applicable regulations, we will not be bound by the terms of this AC. We may require additional substantiation or design changes as a basis for finding compliance.

2.3 The material in this AC does not change or create any additional regulatory requirements, nor does it authorize changes in, or permit deviations from, existing regulatory requirements.

2.4 This AC applies to the following groups:

2.4.1 Applicants who include the seat as part of the aircraft design. In this case, the applicant is not using a TSO-C127x seat eligible for installation in the aircraft. The applicant would satisfy the requirements of § 2X.562 for the baseline seat using the specific interior configuration of the target aircraft (including attachment hardware and fittings). The applicant can substantiate modifications to this seat design and installation using the guidance in this AC.

2.4.2 Seat manufacturers building seats to the requirements of TSO-C127x. In this case, the applicant may hold either a letter of TSO design approval or TSO authorization for those seats. In this instance, the TSO manufacturer would demonstrate compliance to the TSO standard by test for the baseline seat design. The TSO manufacturer can show

2 compliance to the TSO for subsequent changes to this seat by using the analytical techniques and limitations described in this AC.

2.4.3 Aircraft manufacturers or modifiers who wish to install a TSO-C127x seat as part of the aircraft type design. In accordance with AC 21-50, Installation of TSOA Articles and LODA Appliances, dated February 11, 2011, the data approved under the TSO-C127x seat approval and installation limitations may be used to support certification of the seat installation. Modifications by the installation approval holder to the TSO seat system are eligible for certification via the analytical techniques and limitations described in this AC.

2.5 If all pass/fail criteria identified in § 2X.562 or TSO-C127x are satisfied, applicants can use computer modeling analytical techniques to complete the following:

2.5.1 Establish the critical seat installation or configuration in preparation for dynamic testing (refer to section 9 of this AC). However, applicants can still use any other FAA-approved methods to determine the critical test configuration.

2.5.2 Demonstrate compliance with § 2X.562 or TSO-C127x for changes to a baseline seat design, where the applicant has shown with dynamic tests that the baseline seat design meets these requirements. Changes can include geometric or material changes to primary and non-primary structure (refer to section 10 of this AC). Some changes to fittings and joints are acceptable to substantiate with modeling; however, significant changes to the material or mechanism of load transfer are considered major changes.

Major changes to the TSO-C127x seat design require demonstration by dynamic testing, unless a deviation is granted to substantiate via analysis. Additional guidance on the classification of seat design change is contained in FAA Policy Memorandum PS-AIR100-9/8/2003, Classification of Design Changes to TSO-C39b, TSO-C127, and TSO-C127a Articles, dated September 8, 2003.

3 CANCELLATION.

This AC cancels AC 20-146, Methodology for Dynamic Seat Certification by Analysis for Use in Parts 23, 25, 27, and 29 Airplanes and Rotorcraft, dated May 19, 2003.

4 BACKGROUND.

4.1 Attendees at the Small Airplane Airworthiness Review Conference, held in October 1984, developed a series of proposals that focused on cabin safety and design requirements for occupant protection. The proposals culminated with a notice of proposed rulemaking (NPRM) 86-19 (51 FR 44878, December 12, 1986), which included proposed rule § 23.562. The preamble to that proposed rule stated that no sufficient database existed to permit the use of numerical analysis instead of dynamic testing to show compliance with this requirement. However, the preamble also states that the language of § 23.562 is intended to provide flexibility when the state of

3 analytical techniques evolve sufficiently to permit these techniques instead of dynamic tests.

4.2 This AC provides guidance for demonstrating compliance pursuant to § 23.562 by means of computer modeling techniques. Recognizing that this guidance is equally applicable to aircraft other than small airplanes, this AC also applies to §§ 25.562, 27.562, and 29.562.

4.3 This AC defines the acceptable applications, limitations, validation processes, and minimum documentation requirements involved when an applicant uses substantiation by computer modeling to support a seat certification program. The original release of this AC was the culmination of the efforts of the Advanced General Aviation Transportation Experiments (AGATE) Program’s Advanced Crashworthiness Work-package Team. The AGATE project was a consortium consisting of representatives from private industry, research institutions, academia, and the Federal Government. As part of the AGATE effort, the team developed a methodology for seat certification and design by analysis. This AC is a direct result of that methodology.

4.4 In addition, TSO-C127x established a standard for the dynamic testing of seats as specified by § 2X.562. Although installation approval under the airworthiness standards is required for every TSO item, applicants or holders of TSO-C127x may use the methodology presented in this AC in accordance with the applicability defined in paragraph 2 of this AC.

5 RELATED DOCUMENTS.

5.1 Regulations.

The following regulations are related to this AC. You can download the full text of these regulations at the U.S. Government Printing Office e-CFR website. You can order a paper copy by sending a request to the U.S. Superintendent of Documents, U.S.

Government Printing Office, Washington, DC 20402-0001; by calling telephone number 202-512-1800; or by sending a request by facsimile to 202-512-2250.

 Title 14 of the Code of Federal Regulations (14 CFR) part 21, Certification Procedures for Products and Parts.

 14 CFR part 23, subpart C—Structure.

 14 CFR part 25, subpart C—Structure.

 14 CFR part 27, subpart C—Strength Requirements.

 14 CFR part 29, subpart C—Strength Requirements.

 Title 49 of the Code of Federal Regulations (49 CFR) part 572, Anthropomorphic Test Devices.

5.2 FAA Orders.

If any FAA order is revised after publication of this AC, you should refer to the latest version at http://www.faa.gov/regulations_policies/orders_notices/.

 FAA Order 8110.4C, Type Certification, dated October 12, 2005.

 FAA Order 8150.1D, Technical Standard Order Program, dated March 17, 2017.

5.3 Advisory Circulars.

The following ACs are related to the guidance in this AC. The latest version of each AC at the time of publication of this AC is identified below. If any AC is revised after publication of this AC, you should refer to the latest version for guidance, which can be downloaded from the Internet at http://www.faa.gov/regulations_policies/advisory_circulars/.

 AC 20-107B, Composite Aircraft Structure, dated September 8, 2009.

 AC 21-25B, Approval of Modified Seating Systems Initially Approved Under a Technical Standard Order, dated January 11, 2016.

 AC 21-46A, Technical Standard Order Program, dated February 23, 2017.

 AC 21-50, Installation of TSOA Articles and LODA Appliances, dated February 11, 2011.

 AC 23.562-1, Dynamic Testing of Part 23 Airplane Seat/Restraint Systems and Occupant Protection, dated June 22, 1989.

 AC 25.562-1B, Dynamic Evaluation of Seat Restraint Systems and Occupant Protection on Transport Airplanes, dated January 19, 1996.

 AC 27-1B, Certification of Normal Category Rotorcraft, dated September 30, 2008.

 AC 29-2C, Certification of Transport Category Rotorcraft, dated September 30, 2008.

5.4 Technical Sources.

 TSO-C127b, Rotorcraft, Transport Airplane, and Normal and Utility Airplane Seating Systems, dated June 6, 2014.

5.5 Industry Documents.

 Methodology for Seat Design & Certification by Analysis, NASA Advanced General Aviation Transportation Experiments (AGATE) Report WP3.4-034012- 077, NASA, Washington, DC, August 31, 2001.

 Performance Standards for Seats in Civil Rotorcraft, Transport Aircraft, and General Aviation Aircraft, SAE International Aerospace Standard, SAE AS8049 Revision C, SAE International, August 14, 2015.

5  Analytical Methods for Aircraft Seat Design and Evaluation, SAE Aerospace Recommended Practice (ARP), SAE ARP 5765, Revision A, SAE International, December 2015.

 Instrumentation for Impact Test - Part 1 - Electronic Instrumentation, SAE Surface Vehicle Recommended Practice, SAE J211-1, SAE International, July 2007.

 Instrumentation for Impact Test - Part 2 - Photographic Instrumentation, SAE Surface Vehicle Recommended Practice, SAE J211-2, SAE International, November 2008.

 Gowdy, V., DeWeese, R., Beebe, M., Wade, B., Duncan, J., Kelly, R., and Blaker, J., A Lumbar Spine Modification to the Hybrid III ATD For Aircraft Seat Tests, SAE Technical Paper 1999-01-1609, General, Corporate, and Regional Aviation Meeting and Exposition, Wichita, KS, April 20-22, 1999.

 Metallic Materials Properties Development and Standardization (MMPDS) Handbook, MMPDS-07, Battelle Memorial Institute, April 2012.

 Composite Materials Handbook (CMH-17), SAE International, July 11, 2012.

5.6 Other References.

 American Society of Mechanical Engineers (ASME) V&V 10-2006, Guide for Verification and Validation in Computational Solid Mechanics, 2006.

 Belytschko, T., Liu, W.K., and Moran, B., Nonlinear Finite Elements for Continua and Structures, John Wiley and Sons, Chichester, West Sussex, England, 2000.

6 DEFINITIONS.

6.1 Baseline Seat.

The seat that encompasses the common design methodology from which other seats within a family of seats can be derived. This includes items such as geometry, material, manufacturing method, and attachment methods. The baseline seat is typically the seat selected for testing based on its criticality. Other seats in the family will be similar to the baseline seat but might vary based on select details. Some of these details include, but are not limited to, leg spacing, placement of reinforcements (to support attachments), and rivet locations.

6.2 Baseline Testing.

The initial series of tests performed on the baseline seat as part of the original certification to substantiate the seat family.

6.3 Computer Modeling.

The use of computer-based finite element or multi-body transient analysis to simulate the emergency landing dynamic condition of the applicable airworthiness standard.

These codes typically follow an explicit formulation for calculation of the structural

6 response, but can follow a centered mass formulation for calculation of the occupant response.

6.4 Family of Seats.

A group of seat assemblies, regardless of the number of seat places, built from equivalent components in the primary load path.

6.5 Mass Scaling.

In finite element modeling, the process of adding nonphysical mass to the structure to increase the time step, thereby reducing the run time.

6.6 Occupant.

In this AC, occupant is not synonymous with the word passenger; occupant refers to the anthropomorphic test device (ATD). The occupant model is used to correlate the behavior of the ATD, as opposed to human biodynamic behavior.

6.7 Seating Configuration.

The interior floor plan of the aircraft, which defines the seating positions available to crew and passengers during taxi, takeoff, landing, and in-flight conditions.

6.8 Seat Primary Load Path.

The components within the seat that carry the load from the point of load application to the structure that reacts the load from the seat system or sub-system. The seat primary load path varies depending on the parameter being evaluated, as follows:

 Structural—from seat belt/harness to fittings attaching seat system to aircraft structure.

 Lumbar—from bottom cushion to fittings attaching seat system to aircraft structure.

 Row-to-row head injury criterion (HIC)—from point of ATD head contact to the attachment of seat primary structure.

 Head path (for example, front row or large pitch seats)—from seat belt/harness to fittings attaching seat system to aircraft structure.

6.9 Seating/Restraint System.

A system includes the seat structure, cushions, upholstery, safety belt, shoulder harness, and attachment devices.

7 7 VERIFICATION OF EXPLICIT CODES.

The verification process determines the computational model accurately represents the underlying mathematical model and its solution (reference paragraph 5.6. of this AC).

Verification of explicit codes precedes validation, as it is important to minimize errors before progressing. Verification consists of two components, code verification, and calculation verification. Calculation verification is further divided into temporal and spatial convergence.

7.1 Code Verification.

7.2 Code verification is the process of determining that the numerical algorithms are correctly implemented in the computer code and identifying errors in the software.

(Refer to ASME V&V 10-2006 listed in paragraph 5.6 of this AC). Verification helps ensure the implementation of the mathematical model and solution algorithms are working correctly in the code, and that the code solution predicts the analytical solution to a certain extent.

Computer analyses used for certification purposes are typically conducted on a commercially available computer code. It is the applicant’s responsibility to select and assess that computer code produces valid results. The applicant or their software provider performs code verification on the hardware and software environment of their design. In addition to performing code verification, the applicant or their software provider should also include copies of their quality assurance policies and procedures in the verification data submitted for certification.

7.3 Calculation Verification.

Calculation verification, also called solution verification, is the process of determining the solution accuracy of a particular calculation, (refer to ASME V&V 10-2006). The goal of calculation verification is to show that discretization errors due to insufficient spatial or temporal discretization is small, and that the model converges to a unique solution with spatial and temporal refinement, respectively. The convergence errors in the numerical model outputs for the system response quantities of interest should be minor compared to the differences allowed in validation comparisons.

7.3.1 Temporal Convergence.

7.3.1.1 The dividing of the total time of a simulation into smaller segments is called temporal discretization. Each segment is typically referred to as a time step, denoted as Δt below. The stability of explicit time integration methods depends on the time step. If the time step is too large for a given element size and material, the method fails, either because of stability issues or poor accuracy. If the time step size is getting too small, the solution time becomes impractical, thus diminishing the effectiveness of the method. The critical time step for a given finite element according to the Courant stability condition is Δtcr = 2/ω = min(L/c). In this equation, ω is the natural frequency of the finite element, L is the characteristic

8 length of the finite element, and c is the sound wave speed (a function of material stiffness and density) as described by Belytschko, et al. (reference paragraph 5.6 of this AC). The time step used for the calculation is the smallest Δtcr of all finite elements in the model. The time step selected for the analysis should be smaller than the time for the sound wave speed to cross the smallest element in the finite element mesh. Otherwise, numerical instability might develop and cause the solution to diverge.

Belytschko addresses the inherent numerical instabilities encountered with explicit dynamic analysis codes.

7.3.1.2 In theory, the most numerically efficient solution is obtained when an integrating time step equivalent to the stability limit is chosen.

Commercial codes, such as MADYMO, LS-DYNA, Virtual Performance Solution (which contains PAM-CRASH), and Altair RADIOSS, attempt to offset the problems of numerical instability by regulating and constantly updating the time interval used throughout the analysis. The simulation uses a time step based on the mesh, material, and element type. The time step is recomputed at each cycle based on the mesh size and stiffness of the materials.

7.3.1.3 Occasionally, the simulation time step is controlled by only a few small or stiff elements in the model. When this happens, it is typically useful to remesh the controlling elements. In the case of extreme compression, such as with foams consolidating, some codes can automatically remove elements when the length becomes a small fraction of the initial length to avoid extreme time steps or instabilities.

7.3.1.4 When updating the mesh is not practical, mass scaling can be employed.

Mass scaling is the process of adding nonphysical mass to the structure to increase the time step, thereby reducing the run time. Choose a minimum time step (Δt) that is greater than the time step required to satisfy the Courant criteria. The software adds mass in all the elements with a critical integration time step smaller than this given value. A scale factor is applied on the critical time step to take into account that the time step for the current configuration is computed based on the previous configuration (t−Δt). A scale factor applied to the time step can be used to prevent the instability by enforcing the time integration algorithm to use a smaller time step. Depending on the quality of the mesh, and strain rate properties a different scale factor may be used. Typically, the scale factor varies from

0.6 to 0.9.

7.3.1.5 Mass scaling should only be used on the smallest elements contained in the model, so as not to affect the overall mass of the system. We recommend that the overall mass scaling should not exceed 5 percent for critical seat components. For non-critical seat components, an increase in mass of up to 10 percent is acceptable. For quasi-static simulations, it is

9 acceptable to increase the mass scaling up to 10 percent since the kinetic energy is small. Rigid body techniques do not use mass scaling.

7.3.2 Spatial Convergence.

7.3.2.1 The finite element analysis technique divides a continuum into finite elements, volumes, surfaces, and line segments that are, interconnected at a discrete number of points, called nodes, and solve the boundary-value problem. The number of elements and the types of elements used greatly affect the accuracy of the result. For example, a coarse mesh can produce erroneous results. Construction of a model includes a trade-off between the accuracy of the solution resulting from the mesh, and the amount of time it takes to run the simulation. Typically, one uses the coarsest mesh that produces a sufficient level of accuracy. As such, evaluation of spatial convergence is necessary in components that are in the critical load path.

7.3.2.2 An estimate of the spatial convergence can be generated based on two or more mesh refinements. If the results of the numerical solution do not change significantly from the refinement, one can expect to be close to the asymptotic region. For example, the compression of a square cushion via a weight can be computed for differing mesh sizes. (See figure 1 below.)

The results have converged at the fourth mesh, but the third mesh may be acceptable. Exact calculation of the spatial convergence error of an explicit structural analysis is a non-trivial pursuit that is an ongoing research activity. Adaptive meshing techniques and extreme compression of elements or removal of elements will have unknown effects on the spatial convergence.

Figure 1. Mesh Refinement Example