AdvisoryU.S.Department ofTransportation Federal Aviation Circular Adm inistration Subje ct: Quality System for the Manufacture of Date: 07/23/2010 AC No: 21-26A Composite Structures Initiated by: AIR-200
1. Pur pose. This advisory circular (AC) provides information about the requirements of Title 14, Code of Federal Regulations (14 CFR) part 2 1, Cert ification Procedures for Products, Articles, and Parts for quality systems for the manufac ture of compos ite struct ures. These structures involve fiber-reinforced materials; for example, carbon (grap hite), boron, aramid (Kevlar) and glass reinforced polymeric materials. This AC also provides information about essential features of quality systems for composites as mentioned in AC 20-107. Composite Aircraft Structu re. This AC provides information for both an appl icant for, and a holder of, a production approva l, and refers to both as a production approval holder (P AH). This AC is not mandatory and does not const itute a regulation. This AC describes an acceptable means. but not the only means , to comply with these requirements. Howe ver, if you use the means described in the AC, you must follow it in all-important respects.
2. Audi ence. This AC affects all manufacturers or approva l holders producing under part 21, subparts F, G, K, and 0.
3. Effect ive Date. This AC is effective April 16, 2011.
4. Explanation of C hange s. This revision
a. Updates all references to 14 CFR part 21. dated October 16, 2009.
b. Updates formatting to match the current AC formatting policy.
5. Cancellati on. This AC cancels. as of its effective date, AC 21-26. Quality Control for the Manufacture of Composite Structures. dated June 26, 1989.
6. Related Publication s.
a. AC 20-1 07, Composite Aircraft Structure.
b. AC 21-3 1, Quality Cont rol for the Manufacture of Non-Metallic Compartment Interior Components.
c. 1\C 21-43, Production Under 14 CFR part 21 , subparts F, G, K, and 0.
d. AC 23-20. Acceptance Guidanc e on Material Procurem ent and Process Specifications for Polymer Matrix Composite Systems.
7. Quali ty System. A quality system established for manufacturing of composites sho uld be similar to any other qualit y system established to meet the requirements of 14 CFR § 2 1.137. For exa mple, the quality system should include procedures that ensure the quality of incoming materials, the control of in-process manufacturing methods, and testing performed to evaluate the end product for conformity to design requirements. The quality system should include standards to be used for nonde structive and destructive tests, visual inspection techniques during the manufacturing process, and product final acceptance. The standards that detennine the acceptance or rejection of manufacturing-induced defects and damage should take into account the process and inspection capability. These standards should be based on approved data developed as a result of proof-of-structure evaluation s conducted in accordance with AC 20-1 07 and applicable airworthiness standards. This AC addresses those areas of a quality system that may require further expansion to accommodate manufacturing of composites .
8. Material and Pro cess Specificati ons. Unlike parts that use metallic materials in the manufacturing process, the material properties of a composite structu re are manufactured into the structure as part of the manufacturing process. Therefore, material and process specification s used to produce compo site structures need to contain sufficient infonnation to ensure that critical parameters in the manufacturing process are identified to facilitate production and final inspection. Typical material and process specific ations, respectively, should contain the following information:
a. Basic fiber, matrix, and cured component properties (and required testing includ ing type, number. and frequency of tests); and
b. Conditions and requirements during the manufacturing process, including manufacturing method , specific tooling , environmental conditions , significant parameter s during the cure cycle, inspec tion criteria at each operation. and storage and handling conditions throughout the process.
9. Materials and Specific Quali ty System Pro cedures.
a. General. The FAA requires that all drawings submitted for type design approval contain sufficie nt information or references to material specifications or other data that clearly identify the materials and processes needed to ensure production of like articles.
(1) Proprietary process or composition. When materials for a composite structure are produced in accordance with a proprietary process, or the composition of the materials is proprietary, the P AH may exclude this proprietary data from the type de sign submitted to the FAA. ln these instances, the applicable type design drawings should refer to a specification that contains this proprietary information so that complete traceability to material compos ition is possible. These specifications will then be made 2
available to the FAA for review and approval under an arrangement between the FAA and the P AH that wi II protect the propri etary nature of the product. Under 14 CFR § 21.33 and 21.140 (Inspection and tests), the FAA is authorized to review such data upon reque st. Recommendat ions regarding quality system procedures , chemical characterization, testing techniques , and supplier control of purchased materials follow.
(2) Incom ing material acceptance plan. The PAH should have an incoming material acceptance plan that ensures the conform ity and uniformit y of purchased composite material s to speci fication s identified in the approved type design. Copies of supplier laboratory test reports showing actual test results should accompany each batch of material receiv ed for review and approval; however , a material supp lier's test report alone is not adequate documentation to substantiate that material s satisfy all specification requirements. Samples of these materials should be taken on a batch-to-batch basis and tested to engineering and manufacturing requirements for conformity of physical , chemical, mechanical, and processing properties to verify the accuracy of suppliers' laboratory reports. Sample testing could be accomplishe d by the PAl I at their production facilities or by an independent laboratory approved for such testing under the PAH's qualit y system . Suc h testing may be decreased, but not eliminat ed, once confidence in the quality of products from a particular source has been established as part of the PAH' s approved quality system.
b. Resin matri x sys tem. The PAH should require chemical characterization tests of the matrix material.
(1) A typical material specification sets upper and lower limits on the amount of reactive functi onal grou p per unit of material, and other design requirements such as viscos ity, color, and moisture content. The charac terization tests should identify and measure the amounts of individual constituents of the resin system such as the basic epoxide, curing agent , accelerator, hardener, and the reacted constituents caused by resin mixing, storage . fiber impregnation, and other processes.
(2) Tests such as High Performan ce Liquid Chromatography (IIPLC) , Fourier Transfo rm Infrared Spectroscopy (FTIS). and Gelation Permeation Characterization (GPC) for chemical characte rization: and Differential Scanning Calorimeter (DSC) and Dynamic Rheologica l Analysis (DRA ) for thermal analysis and viscos ity/flow/tack/ge lation characteristics: or their equivalent, should be conducted , and resulting traces compared to the standards specified in the material speci fication .
Expe rience has shown that one method of testing a particular sample may yield the same ''fingerprint" as an acceptable sample within spec ification limits , yet the sample is rejectable when tested by another method. Therefore, more than one type of characterization test may be necessary, because all chemical ingredient s affecting a given composite's performance may not be quantifiable by a single test method.
(3) The PAH' s material specification should define the combinations of acceptable test techniques and test results that adequately demonstrate the material's confom1ity and process capability. Techniques for chemical characterization are not 3
universal. but are strongly dependent upon the resin formulation. Therefore, test methods adapted to each material should be developed considering the sensitivity of the method to detect deviation s in formulation.
c. Reinforcement fibers. The PAH shou ld establish procedures in coordination with its material suppliers to control the quality of incoming fibrous materials such as roving, tow, and woven fabric. Testing for mechanical properties is the primary method used for fiber characterization. This method also qualifies or screens continuous fi laments because tensile strength and modules are sensitive indicators of change in composite material mechanical response. ln addition, a means should be developed to control the quality of fiber surface treatment , sizing, etc.
d. Preimpregnated material (Preprcg). In applications where prepreg tape, fabric , or roving are primary materials of compo site struct ures, it is necessary to test for the chemical, physical, and mechanical fiber and matrix dominated properties identified in the applicable prepreg material specification. Quality system procedures previously described in paragraph 7 of this AC for resins and fibers may be used to control the quality of incoming prepreg materials. Chemical characterization should be performed on resin extracted from the prepeg material to be used for production , where applicable.
e. Adhesive s. The quality system procedures recommended for structural adhesives are similar to those for resin matrix systems; that is, the procedure s shou ld provide assura nce that each incoming batch conforms to the chemical, physical, and mechanical properties identifi ed in the material specification.
10. Manufacturing Controls. The manufacture of acceptable and reliable compos ite structure s depends on the type of process contro ls employe d during the manufacturing cycle . If all pertinent process variables are adequately controlled. there is added assurance that the parts and structures produced will be acceptable. The quality system should , therefor e, establish and implement a plan which verifies that (I) the parameters affecti ng material integrity and proces s capabi lity are operating under controlled cond itions; and (2) individual items, batches. or lots conform to specified quality standards. To ensure that quality system objectives have been met, process procedures should clearly define specific materials. tooling, equipment, cure cycle parameters.
quality standards, operator qualification s. storage and handling requirements. traceability records , and any other special requirement s.
a. General. The fo llowing are manufacturing process controls that should be included in a manufacturer's quality system:
(1) The PAl I should develop integrated quality and production cont rol procedures for operat ions that define product configurat ion, selection of materials, tooling and facility equipment. calibration. sequence of manufacturin g operat ions. critical in-proces s paramete rs and processing tolerances, and conformity to quality standards.
(2) Environmental parameters (temperature, humidity, chemical contamination) should be defined and controlled, as required, where composite par1s and structures will be produced, in particular, cutting, layup, and bonding areas.
(a) Unless otherwise validated for the material system in use, the area should be temperature- and humidity-controlled such that the minimum temperature is 65 degrees F with a corresponding relative humidity not greater than 63 percent and the maximum temperature is 75 degrees F with a corresponding relative humidity not greater than 46 percent. The temperature and relative humidity values between the minimum and maximum acceptab le values listed above should form a straight-line relationship.
Procedures for the quality system to verify and record temperature and humidity conditions should be establ ished to ensure environmental stability.
(b) Contamination restrictions in environmentally controlled areas should prohibit the use of uncontrolled sprays, exposure to dust, handling contamination, fumes.
oily vapors. and the presence of other particulate or chemical matter that may adversely affect the manufacturing process. For examp le, release agents or material containing uncured silicon should not be permitted. Also, conditions under which operators may handle materials shou ld be defined.
(c) Layup and clean room air filtration and pressurization systems should be capable of providing a slight positive overpressure.
(3) The PAH should establish a program to train and qualify operators, as appropriate. This program should measure operator performance to production standards and provide for qualification as necessar).
(4) Before production. manufacturing processes should be qualified by demonstrating that the combination of materials, tooling, equipment, procedures, and other controls making up the process will produce parts having consistent material properties that conform to design requir ements. As part of the process qualification, appropriate destructive and nondestructive inspect ion (NO I) of appropriate tool proofing specimens should be conducted to determine conformity to specified design requirements. Destructive tests of specimens verify conformity to the specified physical and mechanical properties. NDI of specimens verifies that discrepancies caused by manufacturing procedures remain within allowab le limits.
(5) Once the manufacturing process has been established, it should not be changed w1less a comparability study and necessary testing of differences has been completed. In addition, processes should be reviewed and requalified, if necessary, whenever any significant changes are made to the process such as sources of material, cure cycle, equipment controls, or autoclave loading patterns and tool design changes.
Proce ss capability should be demonstrated by inspection and testing. as necessary. to determine conformity to design requirements.
(6) To control adhesives and matrix fom1ulation when resin is mixed during part manufacture, documented procedures should be employed to cover the identity of required constituents, respective quantities, and mixing methods and techniques. Tests should be performed on eac h resin mix to ensure conformity to specificatio n requirements and should include sampling methods for chemical testing. The mixing process performed under these procedures should ensure that the sequence of additions and mixing techniques are followed. proper chemical reactions are achieved, aeration of the mix is prevented, and the mixing equipment is properly cleaned.
(7) After initial process qualification, testing (for example. process control panels) for conformity to design requirements should continue regularly to ensure that the manufacturing process, materials, and associated too ling continue to operate in a state of control and produce confo rming parts.
(8) For accepting or rejecting cured structures, NDI equipment and procedures sho uld evaluate specified material defects resulting from manufacturing and assembly operations. The NDI technique used for inspection should have the sensitivity to detect maximum allowable discrepancy type and size in the part.
(9) Spec ifications used by the quality system sho uld define allowable limits for each discrepancy such as adhesive voids, porosity, delaminations, damaged core, core node bond separations, potting cracks, short core, and lack of adhesive.
b. Manufacture of parts. The following are manufacturing controls unique to laminate and wet layup, filament winding. and pultrusion that a quality system should include:
(1) Laminate layup.
(a) Standards and methods should be established to ensure the proper orientation, stacking, and nesting ofthe plies during layup operations. The programs that control tape head and table motion s, and tape feed and tape stacking, should be approved in the quality system. Standards should be established for such in-process variables as tape orientation. gaps, and overlap.
(b) For automatic and hand layup methods, controls should be established for in-process variables that affect the cured laminate quality, for example, resin content, ply compaction, laminate density, porosity that may result from debulking, pre-bleeding, and bagging operations.
(2) Wet layup.
(a) Procedures should be established to ensure the correct material selection.
orientat ion and stacking , or nesting of the plies during wet layup operations.
(b) Procedures should be established to control in-proce ss variables such as resin content, aeration, and air pockets.
(3) Filament winding.
(a) Preimpregnated filament s for dry winding should be stored and handled according to procedures similar to those for other prepreg material s to ensure that the material's original propert ies have been maintained, especially material flow and tack to ensure proper bonding for subsequent winding operations .
(b) Procedures should be establi shed to ensure that the working life ofthe resin system exceeds the anticipated winding time and to ensure gelat ion does not take place before comp letion of windi ng.
(c) Standards and method s shou ld be established to control machine-dominant parameters such as feed rate, feeder arm and mandrel motions .
mandrel dwell angle, number of circuit s per pattern. total num ber of circuit s for complete coverage, numb er of plies per layer, winding angle. fiber tension and alignment, bandwidth, and fiber/resin ratio.
(d) Procedures should be estab lished to control process variables during the winding operation such as resin viscosity , fiber wetting, fiber tension, fiber bandwidth and alignment , air entrapment, degree of compaction, and fiber dama ge.
(4) Pultru sion.
(a) Procedures should be established to control pultru sion start-up, steady state, and shutdown operations including dispos ition of material produced during start-up and shutdown operations.
(b) Process procedures should define limits for important process parameters that determine product quality such as line speed, die temperature profile for the particular operating conditions and resin syste m, clamping pull-through lead, resin temperature , die input temperature. material orien tation for preform operation, and material tension.
c. Assembly of components. The following are manufacturin g controls unique to the assemb ly of manufactured composite parts and structures that a manufacturer's qua~ity system should include:
(1) Sandwich construction.
(a) Core material should be inspected for cell damage. crush. correct length and thickness, and shou ld be properl y cleaned and protected against con tamination.
07/23/20 10 AC 21- 26A
(b) Cured laminates should be inspected for dimensional conformity. Faying surfaces should be properly cleaned and protected from contamination before subsequent assembly and bonding operations.
(c) Uncured laminates used for subsequent co-curing operations should be properly protected and stored to prevent contamination and to retard chemical advancement.
(d) Procedures should be established to control those critical parameters of the assembly process that affect the integrity and uniformity of the end item such as bondline thickness, bondline pressure and temperature distribution, and part ··floating'· within the tooling when the resin or adhesive is in the liquid state during cure cycle (baggin g and tooling requirements).
(e) The performance of the sandwich construction process should be evaluated for each autoclave or oven load by testing specimens of the sandwich construction. Test specimens should be made from specially prepared panels that have been manufactured from the same materials, and using the same production methods.
environment al working conditions, and cure cycle as the production parts they represent.
When co-curing of laminates is required to manufacture an assembly, a solid laminate process control panel shou ld also be prepared for specimen testing. Cutouts or cutoffs from production parts may also be used, providing they are representative of the production part configuration.
(2) Curin g process.
(a) Procedures should be developed to control the critical parameters of the process, namely chemical reactions of the resin and consolidation of the plies to achieve manufacturing consistency and quality parts that are of uniform consolidation, within void tolerances, and of correct fiber volume.
(b) Procedures should identify the equipme nt and materials that will be used to bag the uncured laminate or sandwich construction. This will ensure the proper control of resin flow. volatile content , and ply consolidation during the cure cycle because such materials (for example, peel plies) may alter properties of the structure.
(c) Procedures should define the relationship of the variables (time, temperature, pressure) in the cure cycle (and post-cure cycles) that control compaction and consol idation and cure reaction. These controls should include the acceptab le limits of variables during the cure cycle and action to be taken when such limits are exceeded;
for example, submit cured parts through the quality system for evaluation and disposition.
(d) Procedures should be established to ensure that. when the autoclave or oven is loaded with uncured parts with different configurations. materials. bagging configurat ions, and tooling heat-up characteristics, each part is properly cured.
(e) Procedures should describe actions to be taken when heating, cooling, vacuum, or pressure interruptions are experienced during autoclave /oven cure; for example, submitted through the quality system for disposition.
(f) The performance of the curing process should be evaluated for each autoclave or oven load by testing specimens that are representative of the composite structure. Test specimens should be manufactured from the same materials using the same production methods. environmental working conditions, and cure cycle as the production parts they represent. Cutouts or cutoffs from production parts may also be used provided that they are representative of the production part configuration. The use of validated direct-cure monitoring techniques may also be used , if found to be directly applicable.
(3) Secondary Bonding.
(a) Cleaning procedures for surfaces to be bonded together should specify the chemicals and abrasives to be used on faying surfaces. Standards for determining properly cleaned surfaces and methods for protecting clean surfaces from contamination (including peel plies) shou ld be available at the place of manufacture.
(b) Manufacturing standards should establish adhesive bondline thickness and pressure to be applied along the bondline during the cure cycle.
(c) Tool proofing procedures should demonstrate the capability of the tooling to maintain correct bondline thickness. equal pressure distribution along the bondline.
and properly cured adhesive.
(d) Before the application of adhesives, the production parts should be preassembled and inspected in the tooling to determine that the expected bondline thickness and pressure distribution on the bondline will be within design limits. Shims required to fill bonding gaps should only be used in accordance with an approved quality system procedure.
(e) Methods for applying adhesive films should preclude the formation of defects such as voids caused by air entrapment, bridging, and gaps between adjacent films.
(f) Standards for the application of adhesives (for example. pastes) should require a complete and even coating of adhesive with wetting of surfaces and exc lusion of air in the adhesive.
(g) Controls for high temperature curing adhesives should be recorded and should include heat-up and cool-down rates, cure time and temperature. vacuum. and pressure parameters.
(h) Controls for room temperature curing of adhesives should specify:
07/23/20 10 AC 21- 26A !.: Minimum and maximum time between mixing of adhesive and mating, clamping, or tooling of parts into final configuration; and
2. Minimum time at particular temperatur es to achieve adequate handling strength for subsequent processing, and the conditions (time and temperature) for achieving full cure.
(i) The conditions under which bonded structure s may be subjected to a subsequent curing operation should specify:
l. Max imum allowable temperature ;
~· Maximum number of subsequent curing cycles; and J. Procedu res to prevent bond line separation, blistering , and de lamination s of the structure when subsequent curing temperature s approach that of the previous cure temperature.
(4) Oth er Moldin g Method s. For technique s such as resin injection, compress ion, and transfer moldi ng processe s, specifications should define the limits for all critical process parameters that determine product quality. Such parameters are, for example, resin mix, feed-rate and temperature , mold temperature, and back pressure or vacuum. In addition, molding process specifications should identify the timing and sequence of automatic operations. These critical parameters should be defined in the process specification and identified in the associated operation sheets.
11. Final Acceptanc e.
a. Final acceptance require ments and quality system procedures should prov ide added assurance that the completed struct ure meets its functiona l and design requirements.
b. Final acceptance records should provide evidence that the following significant production and quality system activities, specifically designed to assure the quality of composite structures, have been completed:
(I) Incoming material acceptance:
(2) In-process manufacturing and assembly controls:
(3) Maintenance of tooling and facility equipment;
(4) Calibration of inspection and laboratory test equipment:
07/23/20 10 AC 21- 26A
(5) Inspection acceptance of functional characteri stics at detail and assembly levels;
(6) Nondestructive inspection acceptance~
(7) Configuration control; and
(8) Any other requirements as defined under part 21, subparts F, G, K, and 0.
c. Nondestructi ve Inspection. Several NDI techniques detect discrepancies in composite structures; however, the most commonly used techniques are visual, audio sonic (coin tapping), radiograp hy, ultrasonics, and mechanical impedanc e testing.
(1) Visual. Visual inspection is the most widely used NDI method.
Discrepancies that can usually be observed include: discoloration (due to overheating).
foreign matter, crazing, cracks, scratches, blisters. dents, orange peeling, pitting, air bubbles , porosity, resin rich and resin poor areas, and surface wrinkles. Reflected light is used for observing surface irregularities and other defects, while transmitted light (assuming both surfaces are accessible and the material is translucent) revea ls discrepancies within the speci men.
(2) Audio Sonic. Sonic testing uses frequencies in the audible range. ··Coin tapping" is a common technique used for detecting delaminations. When tappin g any area, a coin or other suitab le object may be used. When this technique is used, a clear, sharp, ringing sound is indicative of a well-bonded solid structure, while a dull sound or thud indicates a delamination. Automated sonic devices that produce a consistent tapping rate and force are available and can be used for this test.
(3) Radio gra phy. Radiography is generally used in composite manufacturing for the detection of bondl ine defects. In add ition, radiography may also be used to inspect for the presenc e of foreign material, adhesive voids, location of internal parts, honeycomb core defects, mislocated or misdrilled holes, poor fit-up, thick bonds, fiber discontinuities, poor tape layup. or lack of adhesive fillets. In the case of carbon/epoxy, glass/epoxy, and aramid (kevlar) /epoxy, the resolution differences are low and defect detection becomes difficult due to the low contrast in the film. Radiography is usually performed through the thickness of the product for detection of anomalies.
(4) Ultrasonic. In this technique, the attenuation of sound wave energy is used for flaw detection. Two methods of ultrasonic techniques are available: Through transmission, and pulse echo. Three methods of recording and display generally used are: A-scan, 8-scan, and C-scan. An A-scan is an amplitude versus time display for a specific point on the part inspected. It generally lends itself to the contact pulse-echo methods. The B-scan displays a long cross-sectional view of the part under test and shows discovered discrepancies. C-scans display the effect in a plane view usually on a paper that is rolled across the printer bar. C-scans provide no defect depth or orientation information; however, they are capable of detecting defects in the magnitude of 0.01 11
square inch. Ultrasonic techniques can generally be used to detect porosity , laminar inclusions, delaminations , and fastener hole flaws. The disadvantage of the ultrasonic method is the inability of the method to distinguish between voids and small delaminations since the attenuation characteristics ofthe two flaws are identical.
(5) Mechanical Impedance. These are methods in which structural response to strain excitation is measured to detect delaminated areas or disbonds . The sensitivity of this method decreases with increasing structural flexibility or discrepancy depth.
d. Controlling NDJ Techniques . In order for NDI techniques to be effective, repeatable, and reliable , certain controls are necessary. These controls should be approved by the FAA. The following controls are recommended:
(1) A quality system approved NDI specification and procedure to be used.
(2) Periodic qualification of personnel conducting the inspection technique. This would nonnally include regularly scheduled vision examinations and inspection of a standard with a known defect.
(3) Establishment of realistic acceptance standards for use by in-process and final inspection personnel.
(4) Calibration of equipment used in the inspection technique including any quality system standards with known defects that may be used. The calibration system should provide for periodic requalification of any such equipment at specific time intervals.
(5) An internal audit program for validating the effectiveness of the NDI program.
12. Storage and Handling. Composite materials and adhesives may be subject to deterioration if not stored under proper environmental conditions. Storing these materials at low temperatures (for examp le 0 degrees) in freezers retards partial curing of polymer materials and extends their shelf life. Therefore, it is essentia l that composite material handling and storage procedures are established , followed, and subjected to periodic independent auditing to ensure continued confonnity to the materials ' chemical , physical , and specified mechanical properties.
a. Uncured Material Storage. Because composite structures are made from polymeric materials (resins , prepregs, adhesives) , procedures that ensure these materials are acceptable at the time of use in the manufacturing area should be established.
(1) Material suppliers should provide documents with each shipment that identify the environmental conditions under whic h the purchased material was previously stored and shipped , temperature and humidity conditions , and accumulated time in and out of 12