·Table of Contents
·Workshop - Reliability
NDI Reliability Rules used by Transport Aircraft - European View Point -G. Tober and W. B. Klemmt
Measurement and Test Technology
The present generation of civil transport aircraft is designed according to the "damage tolerance concept". This concept meets the requirements of the airworthiness regulations as well as the demands for high operational safety and economic efficiency.
The design concept of a damage-tolerant structure demands that a defective structure sustains loads until the defects are detected by a planned inspection or a non-critical function failure. All defects in safety relevant components which are above the tolerance level according to this concept must be reliably detected by means of non-destructive inspection methods (NDI).
This report explains the quantitative definition of the term "reliable detectable defect size" which is currently in use. Procedures are developed on the basis of past experience, of which observation is mandatory if the qualified test method shall supply reproducible and reliable results in field applications.
The current generation of civil transport aircraft is designed according to the damage tolerance concept for at least 20 to 25 years of operations and up to 60.000 flights. This concept meets the requirements of the airworthiness regulations as well as the demands for high operational safety and economic efficiency.
The design concept of a damage-tolerant structure demands that a defective structure sustains loads until the defects are detected by a planned inspection or a non-critical function failure. All defects in safety relevant components which are above the tolerance level according to this concept must be reliably detected by means of non-destructive inspection methods.
Non-destructive inspection methods have been in use for many years in production as well as in-service applications. As a general rule, inspection equipments are calibrated on standards using artificially induced substitute defects. With the sensitivity set to the respective level, the part is then inspected and evaluated as required by the inspection procedure.
In the middle of 1970, various full scale fatigue tests were performed for Airbus aircraft. These tests were accompanied by the application of non-destructive inspection methods. For the development of the inspection procedures, lab investigations were carried out during which calibration standards with artificial defects were used that were similar to the structure. With the determined inspection parameters, the test cells were then inspected at intervals. In the final phase of the tests after approx. 2 to 2.5 service lives, it became evident that the defect sizes that occurred on the actual structure were far longer than the detectable defect sizes that were determined in the lab, and that the test structure ruptured unexpectedly.
As a consequence of these events, comprehensive investigations were carried out in order to determine the causes. The magnitude of the differences is illustrated by an example for an ultrasonic test performed on a riveted longitudinal lap joint with cracks in the hidden layer. A principle representation of this test is given in figure 1.
|Fig 1: Ultrasonic inspection for cracks in the first and second layer||Fig 2: Influence of the defect nature on the POD curve for ultrasonic inspection|
In the lab, saw cuts of lengths of 0.5 mm starting at the rivet holes were clearly detected on calibration standards that were designed similarly to the actual structure. With 1 mm long saw cuts, we got a large signal/noise ratio (approx. 3 to 1), so that we were certain we would find such crack lengths. The subsequently determined POD curves for the artificial defects and natural cracks, respectively, is shown in figure 2. (POD = Probability of Detection)
The curves were prepared on the basis of measured results of several certified inspectors. The inspection was carried out in accordance with a documented procedure. The defect locations were not known to the inspectors, though blind test.
The 90 % value for artificial defects was achieved for 2 mm and that for natural cracks for 5.5 mm. The cause for the significant difference lies are the crack path and the crack starting points which are far less favorable for ultrasonic tests. In addition, varying paint layer thicknesses and press fits of rivets in the holes contribute to the effect in practice.
Past experiences have shown that tear down investigations using full scale fatigue specimens result in indispensable knowledge for structural mechanics as well as for metrology and inspection technology. It would be either not possible or possible only on a very limited scale to gain such knowledge with the desired amount of signifycant information using calculation models. Therefore, all full scale test specimens are subjected to a tear down investigation at Airbus Industrie.
The main goals of the investigations are as follows:
For the implementation of the tear down investigation, we have defined a set of rules. These investigations are based on the prepared structure inspection program (SIP). The air-worthiness regulations require an SIP as per which the aircraft structure is inspected during the utilization phase. The quality and intensity with which this is carried out depend upon the safety relevance of the inspected areas and the knowledge gained from the utilization of the aircraft. For each structural element, a detailed analysis shall be carried out. The facts to be taken into account are shown in figure 3, where experiences gained during the utilization phase, e.g. corrosion, are included additionally. Depending on the design principle and the result of the analysis, the inspection level is then defined.
|Fig 3: Development of a Structure Inspection Program (SIP)|
Our rule is that the tear down inspection is performed by means of an inspection method which is more sensitive and safer than the specified SIP requirement.
The knowledge gained from the tear down investigation on the actual damage is a key element for the development of reliable NDI procedures.
Design in accordance with the -damage tolerance concept", however, requires reliable, reproducible inspection procedures not only for production but also and especially during the utilization phase (field) /3/. With this concept, the point is not
with the specified inspection method and procedure. The NDI engineer must be able to provide reasonable answers to this question and meet the requirements posed by this integral safety concept.
First, we will need objective evaluation criteria for the performance demonstration of an inspection procedure and a definition of the term
The two main criteria for the performance demonstration of an inspection procedure are /1/:
At Airbus Industry, we have defined the following:
With a defined inspection procedure containing sufficient details, a defect size will be reliably detected if
The definition of the findings rate is sufficient to meet the safety requirements /2/, but economic reasons have made us define a minimum requirement for the false alarm rate, as well. Usually, the false alarm rate is below 1% with the procedures in use. The false alarm rate determines directly the refuse rate or the scope of the verification tests with more sensitive procedures.
Fulfillment of such high requirements must be demonstrated only for inspection procedures for safety relevant structural elements.
For the selection of the inspection method and the preparation of the inspection procedure, the analysis of the inspection task is of great importance. Apart from the material and the construction, the analysis should take at least the following parameters into account/4/
Knowledge of the real defect behavior is of great importance, even minor wrong assumptions can influence the findings rate significantly. Apart form the mathematical requirements of structural mechanics, knowledge gained from the tear down investigations is very helpful. The continuous damage analysis of the Airbus fleet is another good source for correlating the defect behavior assumed during the development of the procedure with the actual conditions.
For the qualification of an inspection procedure, i.e. the determination of the POD curve including the false alarm rate, certain rules must be obeyed, as well, if the occurrence of severely false results despite the great expenditure are to be avoided. The basic requirements shall be as follows:
The application of the damage tolerance concept for structural design in the aircraft construction branch implements an integral safety philosophy and requires reliable inspection results to be able to ensure structural integrity. In context with the issues of an aging aircraft fleet and Widespread Fatigue Damage (WFD), the requirement for reliability will gain yet more importance.
As the preceding discussion has shown, a reliable inspection result is the product of a complex process that is influenced by many individual findings from different technical fields.
In this context, the executing inspector must not be disregarded as a factor. His training, experience and motivation are significant factors for the reliability of an inspection. Only qualified inspectors are able to compensate shortcomings of the inspection procedure or equipment
|© AIPnD , created by NDT.net|||Home| |Top||