Another important aspect of reliability is consistency. There
is a measure of reliability called Probability of Detection (POD) which takes into account all aspects of the inspection including the inspection, instrumentation, technique, inspector and human factors, and is referred to as having a POD of 90/95. This basically means that there is a 95 percent confidence that a flaw of a particular size will be found 90 percent of the time. Figure 2 shows POD curves for five typical inspectors. The x-axis represents flaw length and the y-axis represents probability of detection. You can see that while the average POD was an acceptable 0.7 mm, inspector variation went from approximately 0.3mm to a high of approximately 0.9 mm. The variability from inspector to inspector is just as important as the average POD. Considerably different results might be obtained by two inspectors, one on night shift
properly all the time as it involves a great deal of chemistry and physics that is not obvious from the apparently simple steps of the process. Examples of this are shown below in Figure 3. The picture at left shows the failure of a fan hub on an MD-881 which killed two people. The photographs on the right show the remains of an DC-102 that experienced a fan hub failure that caused the loss of all hydraulic systems and the death of 112 people and another 171 injuries. In both of these cases, the failure cracks were present and detectable at overhauls prior to the incidents, but were not found. These inspections were performed by competent engine overhaul shops. Nothing was found wrong with the inspection processes performed and the NTSB and FAA cited “human factors” as the reason for missing them. While the root cause of these failures were titanium inclusions in the disks that acted as stress concentrations and the source of the cracks, it is possible that the failures could have been prevented if the cracks had been found. The fact that human factors were cited as the reason for missing the cracks, speaks to the need for refresher training3. One facility that offers the opportunity to work on actual retired
aircraft is the FAA’s Airworthiness Assurance NDI Validation Center (AANC). It is located in a 24,000 sq ft hangar facility at the west end of Albuquerque International Airport. Established in 1991 as the FAA Aging Aircraft NDI Validation Center, the center was renamed in 1995 the Airworthiness Assurance NDI Validation Center. The AANC is operated by Sandia National Laboratories in Albuquerque, N.M. for the FAA William J. Hughes Technical Center, Atlantic City, N.J. The AANC offers the type of hands-on proficiency opportunities
Figure 2. Variability goes here with the title: Typical POD curves
and one on days, doing the same job. Another scenario would be trying to assess the seriousness of a fleet-wide problem when inspectors of significantly different performance were reporting results from several scattered repair stations. In this case, in order to make good engineering and program decisions, it is necessary to get consistent information from the field. To assess the need for refresher training an informal reliability
study was conducted. It studied a number of qualified, conscientious Level II and III technicians and revealed the following results. They were tasked to inspect 420 sites that contained 68 defects. No technician found all the defects. The proportion of defects detected ranged from 29 percent to 71 percent. The cracks that were missed ranged in size from 0.058” to 0.286” (3 were over 0.100”). This is a potential safety issue. The false call rate (defect- free areas rejected) was as high as 19.4 percent. This is also an economic issue. Every false call results in unnecessary maintenance dollars spent.
This exercise clearly demonstrated the range in performance that can be encountered every day. The refresher class students that participated in this test unanimously agreed that it improved their performance by making them aware of things they had forgotten and bad habits they had fallen into. Based on the debriefings of the POD study, all left with a better awareness of the human factors that can effect their performance. Complacency is one of the major human factors contributing to NDT performance issues. This can come from doing the same job over and over, or from doing the same job for a long time and not finding any defects. The fluorescent penetrant method is very susceptible to
complacency. Penetrant is often thought of as the simplest NDT method whereas in fact—there are many subtle things that can go wrong—it is actually one of the most difficult to perform
32 Aviation Maintenance |
avmain-mag.com | October / November 2011
that would be very limited at a general aviation facility. The AANC was specifically designed with this type of effort in mind and is therefore unique. AANC offers a wide variety of retired aircraft with known structural problems such as fatigue cracks, corrosion, heat damage, etc. The aircraft available on which to train include:
Boeing 737 MD-88 Boeing 747 McDonnell Douglas DC-9
U.S. Coast Guard HU-25
Dassault Falcon 20
Fairchild Metro II Bell 205 (UH-1) and 206 (TH-57) helicopters
Two large McDonnell Douglas DC-9 structural sections
Figure 3. Fan hub failures
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63