Appendix 1
1.E-05
1.E-06
1.E-07
1.E-08 Probability of Exceedance per Unit Area (1 / in2) Figure A1-3 2004 Default Distribution for Manufacturing Induced Anomalies in Circular Holes
0.001 0.010 0.100 Damage Depth (in) Rem aining Life Data ai Cycles σ1 σ2 σ3 σ4 σ5 Risk Calculation Service Life # Events per Service Lifetime 18
0.010 0.100 Depth (in) Appendix 1 Inspection POD 2004 Default Distribution for Manufacturing Induced Anomalies in Circular Holes
1.E-08
1.E-07
1.E-06
1.E-05
0.001 Damage Probability of Exceedance per Unit Area (1 / in2 ) Risk Calculation Service Life Events/Service Life Remaining Life Data ai Cycles Figure A1-4 19
Appendix 1
4. Results. A number of manufacturers ha ve performed this test case. A statistical analysis of the results, given in events per service life, was performed and demonstrated the statistical values shown in the table below (m = mean value; s = sample standard deviation):
Table A1-1 Failure Risk Mean m – 1.65 s m + 1.65 s Events/Service Life value: m Without in-service 2.88E-04 2.58E-04 3.22E-04 inspection With an in-service 1.28E-04 7.29E-05 2.25E-04 inspection All results in the range of (m-1.65s, m+1.65s) for both conditions are considered acceptable.
NOTE: This range is defined as the inte rval, centered on the mean value, covering 90-percent of the result population assuming a log-normal distribution.
Appendix 2 APPENDIX 2. DEFAULT ANOMALY DISTRIBUTION CURVES
1. Manufacturing Anomaly Distribution Curves for Circular Holes.
a. This appendix shows the anomaly distribution curve associated with the manufacturinginduced anomalies in circular holes in titanium, steel, nickel, or powder nickel engine rotors.
(1) Figure A2-1 illustrates a probability of exceedance curve for manufacturinginduced anomalies for circular holes. The horizontal axis represents the anomaly depth in inches. The vertical axis represents probability of exceedance per unit surface area (1/in2) of the single hole in a feature (see Appendix 5 for a feature definition).
(2) Table A2-1 shows the tabular data corresponding to the plot in figure A2-1. The distribution reflected in figure A2-1 is defined by the equation (Eq. A2-1).
(3) The exceedance curve in figure A2-1 defines the distribution corresponding to the hole geometry with relatively high L/D ratio (L/D > 1.3). For lower values of L/D, use a frequency reduction factor as a multiplication factor on equation (Eq. A2-1).
(4) The above frequency reduction factor is defined below in a form of a plot (figure A2-2), a table (Table A2-2) and an equation (Eq. A2-2). Use the frequency reduction, ν, as a multiplication coefficient for the right-hand-side of Eq. A2-1 for the cases where 1.0 < L/D <
1.3. Outside of this range, the frequency factor is defined as follows:
ν = 1.0 for any L/D > 1.3 ν = 0.04 for any L/D < 1.0 Figure A2-1 2004 Default Distribution for Manufacturing Induced Anomalies in Circular Holes Probability of Exceedance per Unit Area (1 / in
2)
1.E-05
1.E-06
1.E-07
1.E-08
0.001 0.010 0.100 Damage Depth (in) 21
Appendix 2 Table A2-1. Tabular Data for Figure A2-1 Depth, in
0.001 5.10E-06
0.002 4.80E-06
0.003 4.51E-06
0.004 4.24E-06
0.005 3.99E-06
0.006 3.75E-06
0.007 3.53E-06
0.008 3.31E-06
0.009 3.12E-06
0.010 2.93E-06
0.011 2.76E-06
0.012 2.59E-06
0.013 2.44E-06
0.014 2.29E-06
0.015 2.15E-06
0.016 2.03E-06
0.017 1.91E-06
0.018 1.79E-06
0.019 1.68E-06
0.020 1.58E-06
0.021 1.49E-06
0.022 1.40E-06
0.023 1.32E-06
0.024 1.24E-06
0.025 1.16E-06
0.026 1.09E-06
0.027 1.03E-06
0.028 9.68E-07
0.029 9.10E-07
0.030 8.56E-07
0.035 6.29E-07
0.040 4.63E-07
0.050 2.50E-07
0.060 1.35E-07
0.070 7.30E-08
0.080 3.94E-08
0.090 2.13E-08
0.100 1.15E-08 Exceedance, 1/in2 22
Appendix 2 Equation for the default hole anomaly exceedance curve:
Equation A2-1 F(x) = ν * 5.42E-06 * EXP(- 61.546*(x)) where x is anomaly depth in inches, and F(x) is exceedance probability per unit area (1/in2).
Figure A2-2. Frequency Reduction Factor Frequency Factor vs L/D
0.0
0.2
0.4
0.6
0.8
1.0
0.7 1.0 1.3 L/D Freq 23
Appendix 2 Table A2-2. Frequency Reduction Factor as a Function of Hole L/D
L/D L/D freq.
0.000 0.040
1.000 0.040
1.015 0.047
1.030 0.055
1.045 0.065
1.060 0.076
1.075 0.089
1.090 0.105
1.105 0.123
1.120 0.145
1.135 0.170
1.150 0.200
1.165 0.235
1.180 0.276
1.195 0.324
1.210 0.381
1.225 0.447
1.240 0.525
1.255 0.617
1.270 0.725
1.285 0.851
1.300 1.000
2.000 1.000 Equation for the L/D frequency reduction factor as a function of L/D is:
Equation A2-2 ν = 0.04 * EXP[10.7296 * (L/D - 1)] where L/D is a ratio of the hole length L over diameter D, and the dimensionless frequency factor ν should be used as a multiplication coefficient for the right-hand-side of the equation Eq. A2-1 for cases where 1 < L/D < 1.3.
Appendix 3 APPENDIX 3. DEFAULT POD APPLICABILITY
1. Use of Default POD Data.
a. This appendix defines conditions relevant to the use of accepted estimates of the probability of detection for specific types of anomalies and specific nondestructive evaluation/inspection techniques, which may be considered as default values when applied under appropriately similar conditions. The conditions defined in this appendix do not necessarily guarantee the validity of these POD values. For example, if inspection parameters, such as penetrant concentration or temperature, are inadequately controlled, the penetrant capability shown in the accompanying figures in Appendix 4 will not be attained, even if the correct penetrant is selected. We recommend use of a written plan for controlling and monitoring inspection processes (see paragraph 3 of this appendix).
b. If the conditions described for each inspection are not satisfied, the resultant inspection capability and reliability will be reduced. Use of the default POD values in Appendix 4 would then be inappropriate and would result in an overly optimistic damage tolerance assessment.
2. Demonstrations of Inspection Capability.
a. For eddy current inspection techniques, a default POD curve is provided in Appendix 4.
Use of the default POD curve must be based on demonstration that the stated calibration and reject signal levels are attainable on the component being inspected, and, for example, are not prevented by noise or geometrical features. The demonstration conditions should be appropriate to the properties of the part inspected that may affect the inspectability, such as surface conditions, depth to be inspected, or proximity to edges. No other demonstration of this default capability is necessary as long as the requirements for the specific inspection technique are satisfied (see paragraphs 3-5 of this appendix).
b. For specific inspection techniques applied under well-controlled conditions, it might be possible to achieve POD values that are significantly better than the default POD values.
Manufacturers may take advantage of any such improved capability and reliability only if it is supported by a well-documented demonstration program. Similarly, a well-documented demonstration program may be necessary to measure POD values appropriate to specific conditions excluded from those default values supported by explicit, reasonable assumptions.
3. General Restrictions and Applicability.
a. The inspection process must be well controlled and performed in accordance with acceptable procedures, such as those defined by the engine standard practices manual, and consistent with good industrial inspection practices, such as those defined by MIL-STD’s or equivalent industry standards.
b. Pertinent inspection process parameters such as coverage, probe indexing, and scanning speeds (see paragraph 5 of chapter 2, and paragraphs 4-5 of this appendix), should be governed by written procedures. Inspection plans and any inspection fixturing should be designed to minimize human and other sources of variability.
Appendix 3
c. Inspectors must be fully qualified and trained for at least one of the following: NAS 410/prEN4179, MIL-STD-410, ASNT-TC-1A, ATA-105, or equivalent, and provided with adequate training instructions in the specific inspection method.
d. The specified default ECI and FPI POD data apply to all materials. Geometric conditions, such as radii and edges, can create areas where inspections cannot be accomplished.
Limitations also exist relative to depth of penetration and near surface resolution. Conditions under which the default POD data were acquired are outlined in paragraphs 4-5 of this appendix.
Seek advice about the equivalence of alternative conditions from those with expertise in NDE.
Areas of high compressive residual stress can have negative effects on the capability of various NDE techniques, most notably penetrant inspection.
e. Applicability of the default POD data is limited to components that exhibit no abnormal surface condition and that have been properly cleaned for each shop manual requirements. No other special pre-inspection cleaning or polishing is required.
4. Restrictions and Applicability: ECI.
a. Eddy current inspection is an inspection technique suitable for the detection of surface or near-surface anomalies. For the purposes of this AC, ECI is intended primarily for application to engine-run components. The default POD data were acquired under the following conditions:
• Probes containing absolute coils, with inspection frequency in the range 2-6 MHz.
• Probe fixturing is capable of following surface contours of the component being inspected, with adequate control of attitude, lift-off and scan indexing. Scan direction was parallel with any uniform feature changes.
• Provision was made for automatic recording of the inspection process signals or automated alarm, or both, when the inspection threshold is exceeded.
b. The default POD data apply to surface-connected, low-cycle fatigue cracks. Cracks are assumed to have a 2-to-1 aspect (length:depth) ratio. Crack sizes are expressed in terms of the length at the surface. Cracks must not be obscured by oxide, contaminants, etc. Inspected surfaces should be flat or only moderately curved. Choice of the appropriate POD curve from those provided must be based on component demonstration of the attainable inspection sensitivity (see paragraph 2 of this appendix).
5. Restrictions and Applicability : Penetrant Inspection (PT/FPI).
a. PT is an inspection technique suitable for detection of anomalies that are open to the inspected surface. For the purposes of this AC, PT is intended primarily for application to engine-run components. The default POD data were acquired under the following conditions:
(1) Fluorescent penetrants were qualified as level 4 by MIL-I-25135, or equivalent, and used with dry powder developer (as a minimum).
(2) Application of penetrant and developer was automated, or for each, standard practices were employed. See, for example, MIL-STD-6866 and AMS-2847.
Appendix 3
(3) Manipulation of the part was possible to present to the inspector an unimpeded view of the surface to be inspected.
b. The default POD data apply to surface-connected low-cycle fatigue cracks. Cracks are assumed to have a 2-to-1 aspect (length:depth) ratio. Crack sizes are expressed in terms of the length at the surface. Cracks must not be obscured by oxide, contaminants, etc. Inspected surfaces should be readily visible. Choice of the appropriate POD curve from those provided must be based on whether focused or full field inspection conditions apply, as those terms are defined in paragraph 4 (“Definitions”) in the body of this AC.
Appendix 4 APPENDIX 4. DEFAULT POD CURVES Figure A4-1 MEAN (50% CONFIDENCE) POD FOR FLUORESCENT PENETRANT INSPECTION OF
FINISH-MACHINED SURFACES 0 10 20 30 40 50 60 70 80 90 100 PERCENT POD Full-field inspection Directed inspection Applicability:
MIL-I-25135 Level 4 penetrants with dry powder developer (as minimum) 10 100 1000 ANOMALY LENGTH (mils) 28
Appendix 4 Figure A4-2 MEAN (50% CONFIDENCE) POD FOR EDDY CURRENT OF FINISH
MACHINED SURFACES 0 10 20 30 40 50 60 70 80 90 100
PERCENT POD Reject at the calibration notch threshold response and above Reject at one half the calibration notch threshold response and above Applicabilty:
Absolute coils, 2-6 MHz inspection frequency Calibration on 0.030" x 0.015" notch Applicable only if the indicated sensitivity is attainable 10 100 ANOMALY LENGTH (mils) 29