Table of Contents ECNDT '98
Design of Modern Aircraft Structure and the Role of NDIH.-J. Schmidt, B. Schmidt-Brandecker, G. Tober
Daimler-Benz Aerospace Airbus
|TABLE OF CONTENTS|
The current generation of civil transport aircraft were designed for at least 20 to 25 years and up to 90 000 flights. These design service goals are exceeded by many operators of jets and turboprops. Future aircraft types are designed for at least the same goals, but structure with higher fatigue life (endurance), higher damage tolerance capability and higher corrosion resistance are required to minimize the maintenance costs and to comply with the requirements of the operator and the enhanced airworthiness regulations.
Non destructive inspections (NDI) are still significant means to fulfill all the requirements. Further significant applications of ND1 are in the frame of another major aviation issue, the aging aircraft issue. Especially the activities regarding widespread fatigue damage (WFD) and the assessment of existing repairs require the application of newly developed and available ND1 methods.
Due to several structural damages which occurred during service and under consideration of the requirements of the US american airforce the airworthiness regulations for civil transport aircraft have been developed significantly in the past 45 years. Especially the introduction of the fatigue and damage tolerance requirements mark the major steps. Table 1 shows an overview of the regulations developed in the USA.
|Table 1: Development of airworthiness regulations in the USA|
|1953 - CAR4b:||no special regulations regarding fatigue|
|1956 - CAR4b Amendment 3:||regulations regarding 'safe life' and 'fail-safe'.|
|1962 - CAR4b Amendment 12:||regulations regarding fatigue for landing gears|
|1966 - FAR25 Amendment 10:||sonic fatigue|
|1978 - FAR25 Amendment 45:||introduction of 'damage tolerance' regulations|
|1981 - FAR25 Amendment 54:||further airworthiness regulations for aircraft certified prior to amendment 45|
To guarantee an equivalent standard of regulations in the USA and Europe harmonization meetings were held between the airworthiness authorities and the manufacturers under the umbrella of the Aviation Rulemaking Advisory Committee (ARAC). Furthermore the new aspects regarding widespread fatigue damage (WFD) were considered. The harmonized forthcoming regulation and advisory circular require: 'An evaluation of the strength, detailed design, and fabrication must show that a catastrophic failure due to fatigue, corrosion, or accidental damage, will be avoided throughout the operational life of the airplane.' and 'The ultimate purpose of the damage tolerance evaluation is the development of a recommended structural inspection program considering probable damage locations, crack initiation mechanisms, crack growth time histories and crack detectability.'
The major requirements of the damage tolerance evaluation are:
The major differences compared with the current regulations are the requirements that:
The development of the structural inspection program is shown in Fig. 1. For each structural element to be inspected the following information has to be provided which are comprised in the Maintenance Review Board (MRB) report:
|time of first inspection in flights|
|period between the repeated inspections in flights|
|detailed description of the area to be inspected including location and access|
|information of the method to be used, for ND1 methods the detailed description of the method is given in a special handbook|
Fig 1: Development of structure inspection program
In general the inspection threshold is determined by the fatigue life to crack initiation under consideration of a relevant scatter factor. For specific structure the threshold is to be based on crack growth analysis.
The inspection interval is determined from the crack growth period between the detectable crack length for the structural detail and the critical crack length under limit load divided by a scatter factor, see Fig. 2.
Fig 2: Principle of damage tolerance investigation
The damage tolerance requirements lead to three major tasks for the aircraft manufacturer:
||Nose Landing Gear A320 |
Fig 3: Design Principle 'safe life'
Design principle 'safe life'
The safe life design principle was applied in aircraft design prior to 1960. According to JAR/FAR 25.57 1 a safe life design is now allowed for the landing gear and its attachments only.
An example is given in Fig. 3. A structure designed as safe life contains a single load path only and the inspectable crack length may be in the range of the critical crack length. Consequently inspection intervals to monitor the structure cannot be defined. A failure of one of the structural elements leads to the complete failure of the safe life structure and possibly to significant consequences for the aircraft.
A fatigue resistant design of safe life structure is based on fatigue life calculations for all structural elements during the design phase and is justified by full scale fatigue test with the complete safe structure. The fatigue life calculations are performed using the linear damage accumulation according to Palmgren-Miner considering relevant load spectra and material (S-N) data. The calculated fatigue life as well as the achieved test life are divided by relevant scatter factors.
Fig. 4 shows a single load path design where the justification is based on the following analyses. Fatigue life calculations are performed to justify the reliability during service and to determine the inspection threshold. For future projects the inspection threshold has to be based on crack growth analysis according to the forthcoming regulations. The inspection interval is determined from the crack growth period between the detectable and the critical crack length divided by a scatter factor. The calculation of the crack growth is based on the Forman equation or equivalent.
Fig 4: Design principle 'damage tolerant - single load path'
The 'multiple load path' category is sub-divided into three groups:
Only the latter group is described here, see Fig. 5. For structures 'damage tolerant - multiple load path - inspectable for less than one complete load path failure' again fatigue life calculations are performed to show sufficient reliability during service and to determine the inspection threshold, which is derived from the structural element with the lowest fatigue life. The inspection interval is based on the crack growth behavior of both load paths were in the primary load path an initial flaw of 1.27 mm is assumed and in the secondary load path an initial flaw of 0.127 mm. The interval is determined by the crack growth period between the detectable crack length in the primary load path and the critical crack length in the secondary load path divided by an appropriate factor. For the crack growth calculations the same method as for single load path structure is applied.
Fig 5: Design principle 'damage tolerant - multiple load path -
inspectable for less than one complete load path failure'
The current, and forthcoming, regulations allow both damage tolerance categories, i.e. single load path and multiple load path. The multiple load path design, however, is highly recommended in the interpretation of the regulations (advisory circular AC/ACJ 25.571). The recommended multiple load path design leads to additional safety, but causes, in exceptional cases, significant costs during design and production.
Fig 6: Application of ND1 in structural inspection program of A320-100
Fig 7: External inspections of upper and side shells of A320-100 center fuselage section
Fig. 6 shows the distribution of the inspection levels for the structural significant items (SSI's) of the major aircraft components using the standard body Airbus A320400 as an example. Several SSI's comprise more than one inspection task. Except for the safe life landing gears the 5.percentages of the ND1 tasks are 6 percent for the stabilizer (mainly composite), 11 percent for fuselage and doors, 18 percent for wing and 19 percent for the pylons. The percentage of ND1 tasks may be higher for widebody aircraft which have in general higher stress levels in most of the structural details leading to faster crack propagation and lower critical crack length. Therefore sometimes an ND1 method is chosen to reach a sufficient inspection interval.
The external inspections of the upper and side shells of the A320-100 are given in Fig. 7. Besides a general visual inspection of the complete shells, special tasks of general visual inspections, also covered by the zonal program, are described for the upper panel of the longitudinal lap joints. Detailed inspections are to be performed of the skin at the circumferential joints in the upper area, the surrounding of cut-outs in the upper shell, the skin and the window frames and the cut-out comers of the emergency exits. ND1 methods are used for the strap at the circumferential joints (upper area) and, offered as an alternative to a detailed inspection of externally visible cracks, for the lower panel of the longitudinal lap joint in the upper shell. In principle these external inspections are typical examples for the fuselage upper and side shells at standard body and wide-body Airbus aircraft. The only exception are the cut-out comers of the doors where on widebody aircraft mostly ND1 are applied due to the higher stress level.
Fig 8: Design of aircraft structures
Fig 9: Planned Airbus megaliner A3XX
Fig 10: Two-bay-crack criterion
Several aspects of the design of modern aircraft structure are described here using the fuselage of the planned Airbus megaliner A3XX as an example, see Fig. 9. This aircraft is to be designed for the following goals:
|24 000 flights|
12 000 flights
6 000 flights
The design criteria to be met are static strength, residual strength, durability, crack growth, sonic fatigue strength and the so-called two-bay-crack criterion. This requires the consideration of corresponding loads as static loads, residual strength loads, discrete source damage loads, operational loads and sonic fatigue loads. Furthermore the corrosion resistance, the repairability and the inspectability have to be taken into account.
One of the major criteria which an aircraft has to fulfill to reach the safety standard of the competitors is the two-bay-crack criterion, see Fig. 10. It has to be shown, that a longitudinal crack in the skin of the pressurized fuselage with a length of two frame bays above a broken center frame does not lead to a complete failure of the structure. The load case to be considered is 1.15 of the onerational cabin differential nressure at cruise altitude without consideration of external loads.
The structure of a pressurized fuselage which fulfills this criterion has to guarantee that neither the crack in the skin becomes unstable nor that the stiffeners perpendicular to the crack (i.e. the frames) fail statically. The two-bay-crack criterion is the designing criterion for large areas in the upper and side shells of the pressurized fuselage of medium and long range aircraft. These aircraft types have lower design service goals in flights compared with short range aircraft with the result that the fatigue and damage tolerance criteria have less influence on the design. To limit the implications on the weight due to the compliance with the two-bay-crack requirement following precautions are possible:
During the initial design phase of the Airbus A3XX the application of new materials and production methods is considered to reduce the production costs and the weight and to comply with the forthcoming regulations. To substitute the fuselage material of the current Airbus types, i.e. the 8.aluminium alloy 2024, three different materials are under consideration; these are 2524,60 13 and GLARE, see table 2.
|Table 2: Materials for fuselage skin|
|material data||2024T3 clad||2524T3 clad||6013T4/T unclad||GLARE4 (LT/TL) unclad|
|Rm||(in %)||100||100||~75||190 / 120|
|Rp0.2||(in %)||100||100||-94||ll0 / 80|
|blunt notch||(in %)||100||100||not tested||l43 / 100|
|young's modulus(tension)||(in %)||100||100||~95||79 / 70|
|KC||(in %)||100||-120||~115||~120 / -110|
|corrosion resistance||basis||equal||equal / less||higher|
The materials 2524 and GLARE4 show significantly higher fracture toughness compared with 2024 which results in significant weight reductions in those areas which are designed by the two-bay-crack criterion. The disadvantage of both materials is the higher price. For the GLARE4 material this may be (partly) compensated by a simplified design and production, GLARE4 has additionally advantages with respect to the static strength, the yield strength and the corrosion resistance. Furthermore GLARE4 shows a very good bum through behavior which should be taken into account besides the structural aspects. The material 6013 leads to similar structural weights as 2524 considering the slightly lower yield strength which is approximately compensated by the lower density. 60 13 can be welded which allows to substitute the bonding or riveting of the stringers to the skin by welding. This new production method is very promising with respect to the reduction of the production costs.
The different material data allow an increase of the allowable circumferential stresses in the fuselage of the A3XX for all of the three new materials. An increase of the allowable longitudinal stress in the fuselage is possible when using 2524T3. Table 3 contains the allowable skin stresses for a the frame pitch of 656 mm. The allowable stresses in circumferential direction result from the two-bay-crack criterion, the criterion for the longitudinal stresses is either the crack growth,i.e. the inspection interval, or the two-bay-crack criterion depending on the ratio of static and fatigue loads.
|Table 3: Allowable stresses for fuselage skin|
|skin material||allowable stress in allowable stress in circumferential direction longitudinal direction (residual strength)||allowable stress in longitudinal direction (crack growth / residual strength)|
|2024T3 clad||100 %||100 % / 100 %|
|2524T3 clad||120 %||113 % / 110 %|
|6013T4/T6 unclad (integral stringers)||115 %||104 % / 70 %|
|CLARE4 clad||120 %||120 % / 100 %|
|Fig 11: Design criteria for A3XX fuselage sections|
The improvements given in table 3 lead to weight reductions in those areas where the damage tolerance aspects are the dimensioning criteria. Further design cases to be considered are e.g. the
static tension and compression strength and the engine rotor failure.
Fig. 11 shows the design criteria in the different fuselage areas for an A3XX depending on the skin material.
Finite element analyses were carried out for two fuselage sections of a length of 5.3 m and 2.7 m (forward and aft of the center section) considering the different design cases and the allowable stresses. The resulting structural weights for the skin and the stringers were determined, see table 4. If the weight of the frame is taken into account in addition the total weight reductions are less, e.g. for GLARE4 the weight reduction of the fuselage shell (skin plus stringers plus frames) is 12 percent instead of 16 percent for the skin and stringers only.
|Table 4: Weights of two fuselage sections|
|skin material||cabin differential pressure||weight of two fuselage sections skin and stringer only (frame pitch 656 mm)|
|2024T3 clad||605 hPa||100%|
|2524T3 clad||605 hPa||94%|
|6013T4/T6 unclad||605 hPa||103%|
|GLARE4 clad||605 hPa||84%|
Special ND1 application
The development of a new production technique such as the laser beam welding (LBW) requires a comprehensive use of sophisticated inspection methods, especially the ND1 techniques. During the development of the LBW technique for connection of the stringers to the fuselage skin the following standard ND1 methods are used:
The overall target is to provide an online ND1 method for valuation of the welding beam quality, i.e. methods should be available in the field of production for:
The well known Aloha accident near Hawaii in April 1988 which led to the loss of an upper forward fuselage segment, resulted in worldwide activities to increase the safety of the aging aircraft fleet. Further events showed that the damage mechanism which led to the Aloha accident was not a single case and that the issue of widespread fatigue damage (WFD) was not sufficiently covered by the current regulations.
Aging aircraft initiatives
The Aloha accident prompted considerable aviation community activity related to aging air frames. Manufacturers, operators and authorities got together to initiate changes to the system for safety improvement. A number of industry committees were formed and the first was the Air worthiness Assurance Task Force (AATF) later renamed as the Airworthiness Assurance Working Group (AAWG) which works under the umbrella of the Aviation Regulatory Advisory Committee (ARAC). Two other committees were formed which were the Industry Committee on WFD to study this phenomenon, and the Structural Audit Evaluation Task Group (SAETG) which was charged to develop guidelines to establish the beginning of WFD.
The FAA organized a number of conferences on aging aircraft and structural integrity which were supported by NASA. They created centers of excellence by providing funding; two examples are the Georgia Institute of Technology tasked with the issue of computational mechanics and the Iowa State University tasked with non destructive evaluation. Furthermore, rule changes were initiated to require full scale fatigue testing and inspection threshold determination for new aircraft as described in chapter 2.
Early in all these activities an interim solution was defined for eleven aircraft types which were defined prior to the introduction of FAR 25.57 1 Amendment 45. These models are:
Boeing B707, B727, B737, B747, Douglas DCS, DC9, DClO, Lockheed LlOll, BAe BAC 111, Fokker F28 and Airbus A300.
For these aircraft types the following activities were defined:
The aging aircraft issue 'Widespread Fatigue Damage'
The main issue of the aging aircraft fleet is the occurrence of multiple damages at adjacent locations which influence each other. Two types of multiple damages are known. The sketch on the upper righthand side of Fig. 12 shows an example of multiple site damage (MSD), which is characterized by the simultaneous presence of fatigue cracks in the same structural element. The second type is the multiple element damage (MED), which is characterized by the simultaneous presence of fatigue cracks in similar adjacent structural elements. Both, MSD and MED, are a source of WFD which is reached when the MSD or MED cracks are of sufficient size and density that the structure will not longer meet its damage tolerance requirement.
Fig 12: Effect of multiple site damage
Fig 13: Effect of MSD on residual strength of a lead crack
The effect of MSD is shown in Fig. 12. The lefthand diagram describes the effect of MSD on a single lead crack used to establish the inspection program. In the presence of MSD adjacent to the lead crack the critical crack or the residual strength, respectively, are reduced drastically. The righthand diagram shows the reduction of the crack growth period due to the reduction of the critical crack length.
Boeing has made investigations about the effect of MSD on the residual strength of a lead crack which are published in /l/, see Fig. 13. The residual strength load of a 14 inch (356 mm) long lead crack is reduced in the presence of adjacent MSD cracks of 0.05 inch (1.27 mm) by 30 percent. This demonstrates the dramatic effect even of small MSD cracks which are uninspectable by state of the art techniques.
The Industry Committee on WFD has evaluated the experience of the participating manufacturers based on the results of large component and full scale fatigue tests as well as in service experience in order to identify the locations potentially susceptible to WED. From this compilation of data each area was assessed for its susceptible to WFD and was then characterized as either multiple element and/or multiple site damage. Fourteen areas were identified as potentially susceptible to WFD:
Wing and empennage:
|Fig 14: Example of area potentially susceptible to WFD, circumferential joints and stringers|
For each of these fourteen areas a typical design was given and the type and possible location of MSD/MED was defined. An example is given in Fig. 14 showing circumferential joints and stringers. In detail the following damage types were defined:
without outer doubler:
- splice plate - between and/or at the inner two rivet rows
- skin - forward and aft rivet row of splice plate
- skin - at first fastener of stringer coupling
with outer doubler:
- skin - outer rivet rows
- splice plate/outer doubler - inner rivet rows
In August 1997 the FAA has tasked the ARAC to continue the activities on the WFD assessment and to extend them to all transport category jets and turboprops with maximum gross weights greater than 75000 lbs. The ARAC then chartered a new group in frame of the AAWG called Task Planning Group (TPG) with the following activities:
The AAWG-TPG started their work in autumn 1997 in order to complete it within 18 months. The TPG has defined eight tasks to fulfill their charter:
|Task 1 - Background:||Review actions done|
|Task 2 - Technology issues:||Technology readiness and validation|
|Task 3 - Model specific issues:||Establishment of time frame|
|Task 4 - Regulatory issues:||FAA recourses if OEM fails to voluntary complete WFD audit|
|Task 5 - Management of MSD/MED in fleet:||Inspection programs, replacement|
|Task 6 - Aircraft to be considered in recommendation:||Define aircraft|
|Task 7 - March ARAC report issues and items:||Issues to be presented to ARAC and AAWG response|
|Task 8 - Final report:||Results of tasks 1 to 5|
One major item of task 2 deals with the readiness of the ND1 technology. In frame of this subtask four actions were defined to push the development of the methods needed:
Repair assessment for aging aircraft
Continuous airworthiness assessment of existiong repairs was identified as one of the five significant concerns by the AAWG which formed a Repair Assessment Task Group (RATG) with participation of operators, manufacturers and authorities. The final draft report of this task group which was issued in December 1996 has recommended a one time structural repair assessment task for the external fuselage pressure boundary (skin and bulkhead webs) to assure the continued airworthiness. This recommendation is again applicable to the eleven aircraft models certified prior to introduction of FAR 25.571 amendment 45. Consequently guidelines were developed to assess the damage tolerance of existing structural repairs which may have been designed without using damage tolerance criteria.
Fig 15: Airbus repair assessment process
Fig 16: External skin repair
Based on the general three stage program, which was developed in a common effort by the major manufacturers and operators for categorization of the repairs, the Airbus repair assessment process was defined, see Fig. 15. Stage 1 (Data Collection) specifies what should be assessed for repairs. If a repair is on structure in an area of concern the analysis continues, otherwise the repair does not require classification as per this program. Stage 2 (Repair Categorization) categorizes the repairs regarding maintenance actions to be applied. The repair categorization contains several steps which consider the general conditions of the repair, the quality of the static design, the proximity to other repairs. Stage 3 (Determination of supplementary maintenance requirements) contains the definition of the necessary maintenance program for the repair.
For the Airbus A300 aircraft Repair Assessment Guidelines(RAG) were developed which allow the operators to determine the inspection threshold and interval for the category B repairs. Fig. 16 shows a principle sketch of an external skin repair. In principle four fatigue sensitive locations exist which have to be assessed:
|Fig 17: Determination of repair parameters Fig 18: Inspection of skin and external repair doubler|
The determination of the inspection threshold and interval requires the exact knowledge about the geometry, materials and fastener data to calculate the correct values for threshold and interval. For dat not known conservative assumptions are to be made which would lead to a worse threshold and / or interval. If the data are not available in a repair documentation, they may be taken directly from the aircraft. Some of the data may not easil be measured, but NDI methods have to applied. Fig. 17 shows the application of NDI methods to determine the cut-out size hidden by the repair doubler, the thickness of skin and doubler and the rivet material.
The inspection interval for the repair is based on the crack size detectable by NDI means. Fig. 18 contains the NDI procedures for inspection of the skin and the external repair doubler. All procedures have been qualified and comply with the defined inspection requirements that the defect size to be detected is determined with a probability of detection (POD) of 90 percent at a confidence level of 95 percent.
The next aircraft generation has to comply with the forthcoming more stringent regulations, e.g. regarding widespread fatigue damage and initial flaw concept for threshold determination. Furthermore the general aviation standard with respect to the two-bay-crack criterion should be reached without special design precautions, such as crack stoppers, and without disadvantages in weight. Additionally the requirements of the airlines regarding reduction of the maintenance costs have to be considered, i.e. among others the inspection intervals have to be increased by decreasing the crack growth. These goals may be reached for fuselage structures by application of new materials. The development and application of new material is still under investigation to reach the optimum of material and production costs, weight and maintenance costs. During the development and certification of an aircraft the NDI plays a major role as shown in this paper. Further significant applications of NDI are within the frame of the aging aircraft activities where the detection of MSD and MED is an important item during the assessment of the structure susceptible to widespread fatigue damage.
The Repair Assessment Guidelines which were developed by Airbus also rely on NDI for determination of the repair parameters and the inspections of the repair.