5.1.6
5.1.7
5.1.8 specified on an assembled module when the module is available. This inspection is an opportunity inspection based on module availability rather than piece part availability.
Forced Inspection Opportunities (by Disassembly ).
The designs should use opportunity inspections to meet the DTR levels whenever possible. In some instances, the proba bilistic analysis may indicate unacceptable risk levels when using only opportunity inspections; therefore, additional action may be required to meet the DTR. One of the many options to mitig ate this risk is to force inspection opportunities by specifying disassembly of modules or engines when a cyclic life interval has been exceeded. There are many options on how to implement forced disassembly. The options range between the following:
• Mandatory en gine removal and subsequent teardown at “not to exceed” cyclic limits (“hard-time” limits).
• Mandatory module teardown when the naturally occurring module availability exceeds the specified cyclic life inspection interval of one of the parts contained within that module (“soft-time” limits).
This AC only rec ommends the soft-time inspection option when forced disassembly of a module is needed to meet the DTR levels.
“On-Condition” Maintenance Practice Philosophy.
The soft-time inspection philosophy retains the “on-condition” maintenance practice and minimizes the impact of additional module disassembly. The inspection philosophy applies only after the engine has been removed from the aircraft for a reason other than the inspection itself, and the engine is sufficiently disassembled to afford access to the module containing the component in question. A module containing a part with cycles since last inspection (CSLI) in excess of the soft-time interval should be sufficiently disassembled to allow inspection according to the procedure specified by the engine manufacturer. The engine manufacturer should evaluate the risk associated with parts that become available for inspection before the soft-time interval to determine if the CSLI can be reset.
Consideration of Maintenance Impact of Soft-Time Intervals.
The maintenance impact of the soft-time intervals should be considered during the design phase. To develop designs that achieve the design target, but also result in acceptable soft-time intervals and procedures (should such action be required), manufacturers should use the following inputs:
•
Probabilistic ana lysis.
•
Anticipated e ngine removal rate.
•
Module availability.
• Piece part availability.
5-2
5.1.9
5.1.10
5.2
5.2.1 Interval Limits as Invoked by Soft-Time Inspection.
When invoked, the soft-time inspection approach establishes interval limits beyond which rotor components must be insp ected when the rotors are available in modular form . The soft-time inspection interval is not intended to affect or modify current forced inspec tion programs, which address the safety of flig ht concerns that arise in the course of en gine operation and maturation. Manufacturers should continue to address safety of flight concerns through aggressive inspection programs mandated through airw orthiness directives.
Communication and Implementation of Inspection Assumptions.
The inspection assumptions made in the probabilistic risk assessment must be accur ately implemented and communicated to the fiel d using the AL se ction of the ICA.
The assumptions must also be validated by engine removal rate reviews, as well as module and piece p art availability data.
For example:
• The AL section must call out an immersion ultrasonic inspection if that was an assumption used to set the original soft-time interval.
• The amount of inspected material should correspond to the analysis assumptions. •
• If field experience suggests that the opportunity inspection intervals are in excess of the assumed rates in the probabilistic risk assessment, then appropriate corrective action is required (for example, modifying the inspection plan).
Manufacturers will:
• Specify the soft-time inspection interval and reference the corresponding inspection procedures in the AL section of the ICA. This information is to be provided for all rotor parts with specified retirement life-limits that require inspection plans beyond opportunity inspections to meet DTR levels.
• Include required inspection information in the AL section of the ICA with the other rotor inspection requirements.
• Provide necessary information to focus the prescribed inspections to the highest relative risk areas.
Inspection Scenario
s.
The scenarios in the following subparagraphs clarify the action to take at a maintenance inspection opportunity. Note that the inspection plans may vary for each part, depending on the probabilistic assessment outcomes.
Maintenance Opportunity - Hardware Available for Opportunity Inspection.
Fo r hardware available in the condition to perform the specified opportunity inspection, you must inspect using the procedures specified in the AL section of the ICA. This condition would be a mandatory inspection.
5-3
AC 33.70-34/17/23
5.2.2 Maintenance Opportunity - Module Below Soft-Time Interval.
F or hardware accessible in the assembled or partially disassembled module, you may nondestructively inspect using the procedures specified in the AL section of the ICA.
You ma y reset the CSLI to zero provided the engine manufacturer has assessed the risk impact associated with this action. This condition would be a discretionary inspection.
5.2.3 Maintenance Opportunity - Module Above Soft-Time Interval.
For hardware listed in the AL section of the ICA, you must make it available for nondestructive inspection using the specified procedures. You must perform this inspection whenever the module is available, and the CSLI for any contained hardware exceeds the inspection cycle limit. This condition would be a mandatory inspection.
5-4
AC 33.70-34/17/23 Appendix A APPENDIX A. CALIBRATION TEST CASE A.1 Overview of Self-contained Package.
This appendix provides a self-contained package for the calibration of a probabilistic risk assessment methodology. The package includes all required input data for the test case, analysis guidelines, and a test case analysis section. The test case analysis section permits manufacturers to estimate the level of acceptability of their risk calculations and gain insights on intermediate results.
A.2 Test Case Input Data.
A.2.1 Anomaly Distribution Curve.
The FAA is providing the anomaly distributions in figure A-11 for this test case. Billet and forging manufacturing inspections are fully accounted for in this curve; no additional modifications are necessary.
1. It is assumed that anomalies are spherically shaped and uniformly distributed throughout the part.
2. The anomaly distribution should be linearly extrapolated when anomaly sizes are required outside the range of data provided.
A.2.2 Probability of Detection.
The POD curve used to determine the effect of an in-service inspection is contained in appendix E of this AC. The default curve to use is the mean POD for ultrasonic inspection of field components with reject indications equal to or greater than those from a 3/64 inch (1.19 mm) diameter FBH. For the test case, it is assumed that this curve applies to the whole volume, including the near-surface volume of the component.
A.2.3 Maintenance Exposure Interval.
Assume that 100 percent of the fleet is ultrasonically inspected at 10,000 cycles, which represents 50 percent of the certified part life (20,000 cycles).
A.2.4 Incubation.
No anomaly incubation life should be assumed.
A.2.5 Stress.
The hoop stress is the limiting operational principal stress.
A.2.6 Material Data.
Two sets of material data are provided:
1. Physical properties. Data required:
Density: 4,450 kg/m3 or 0.161 lb/in 3 Young modulus: 120,000 MPa or 17.4E3 ksi A-1
Appendix A Poisson' s ratio: 0.361
2. Crack Gro wth. Assume the following data represents both air and vacuum crack propagation. Crack propagation rate:
da/dN = 9.25 E-13 (ΔK)3.87 (da/dN in m/cycle and ΔK in MPa√m) ─ or ─ da/dN = 5.248 E-11 (ΔK)3.87 (da/dN in in/cycle and ΔK in ksi √in-) K threshold = 0.0 MPa√m or 0 ksi√in Fracture toughness = 64.5 MPa√m or 58.7 ksi√in Yield = 834 MPa or 121.0 ksi UTS = 910 MPa or 132.0 ksi Note 1: The above data applies at the test case component temperature.
Note 2: Crack propagation data are for a stress ratio of zero; therefore, no stress ratio correction is requir ed.
Note 3: The FAA obtained this data from MCIC-HB-01R, Damage Tolerant Design Handbook - A Compilation o f Fracture and Crack-Growth Data for High- Strength Alloys, vol. 1, dated December 1983, (page 411.257, figure 4.113.104). It represents generic Ti 6-4 Paris fit data. The FAA is providing this data for example pur poses only. It does not constitute a recommendation for analyzing actual components.
A.3 Test Case Analysis Guidelines.
The FAA is providing the analytical guidelines for the probabilistic assessments with the intent to minimize the variations of the applicant’s results due to analytical assumptions.
The practice presented is based on a typical, embedded anomaly probabilistic fracturemechanics approach. The component is subdivided into zones, the relative risk or probability of failure (POF) is calculated for each zone, and results for each zone are summed statistically to arrive at the total component POF or relative risk.
This analytical approach can be divided into the following five basic steps:
1. Stress analysis.
2. Zone definition and volume calculation.
3. Crack growth model definition.
4. Crack growth calculation.
5. Zone and total part POF calculation.
Note: Para graph A.4 provides a systematic example for the calibration test case.
A-2
AC 33.70-34/17/23 Appendix A A.3.1 General Analytical Guidelines.
A.3.1.1 A.3.1.2 A.3.1.3 A.3.1.4 Stress Analysis.
The applicant determines the level of mesh refinement of the part model.
However, applicants should take steps to ensure that the final answer does not change by a significant amount (5 percent on relative risk or POF) if a finer mesh is chosen.
Zone Definition.
Zones are defined as regions of the component (typically made up of a number of finite elements) where life is approximately constant for a given initial crack size. Grouping elements into zones based on stress intervals of 5 ksi (34.5 MPa) is a suggested practice for initial zone definition. Figure A-1 provides a general description of the typical types of zones.
Crack Growth Calculation.
Applicants should base the crack growth-life assumed for each zone on the minimum life location in the zone. This conservative assumption may require that regions of the component that make a significant contribution to the total part POF be broken down into multiple zones. Carry out this subdivision process until you reach convergence of the risk calculation.
Note: The method described in this AC has been calibrated against industry experience without considering the surface enhancement effects, such as shot peening, on predicted crack propagation lives. Therefore, it is inappropriate to include the beneficial effects of such enhancements.
Probability of Failure (POF) and/or Risk Calculation.
The POF of the part is calculated by statistically combining the POF of each zone (surface and subsurface). The POF of each zone can be calculated in either of the following two ways:
1. An integrated probabilistic method.
2. The “Monte Carlo” method. The number of simulations required is related to the computed risk. The general rule is that the number of simulations should be at least two orders of magnitude higher than the computed risk. For example, if risk is one failure in 104 parts, the number of samples required is 106. This ensures that about 100 “failed” parts are involved in the assessment.
A-3
Appendix A A.3.2 Specific Guidelines fo r Fracture-Mechanics Modeling, Zone Definition, and Volume Calculations.
A.3.2.1 Subsurface Zones.
Applicants should:
1. Not model surface enhancements for the test case. The method described in this AC has been calibrated against industry experience without considering the surface enhancement effects, such as shot peening, on predicted crack propagation lives. Therefore, it is inappropriate to include the beneficial effects of such enhancements.
2. Use the maximum principal stress in each zone in the crack growth calculations.
3. Consider the impact of stress gradients. To reach a converged solution, high stress, near-surface regions of the part may require additional refinement beyond the 5 ksi bands suggested in the general guidelines (for example, disk bores and bore sides). Subdividing the regions into sub-surface layers will likely capture the rapid change in life from surface to subsurface and reduce conservatism in the prediction.
Engineering judgment and experimentation will be required to determine the optimum near-surface zone geometry (for example, width and thickness).
4. Consider a surface crack growth correction factor in the stress intensity (K) solution for cracks transitioning to surface cracks.
5. Position the crack at the life-limiting location in each zone.
6. Assume a circular crack geometry (a = c).
7. Consider the defect area equal to the circular crack area.
8. Assume the zone volume to be equal to the volume of the finite elements (or fractions of elements) used to construct the zone.
9. Take crack depth (a) as the diameter of the subsurface crack (2a) when transitioning to a surface crack, just as it touches the surface (see figure A-2).
10. Use average air crack growth data.
A.3.2.2 Surface Zones.
Applicants should:
1. Not model surface enhancement for the test case. The method described in this AC has been calibrated against industry experience without considering the surface enhancement effects, such as shot peening, on predicted crack propagation lives. Therefore, it is inappropriate to include the beneficial effects of such enhancements.
2. Use the maximum principal stress in each zone in the crack growth calculations.
A-4
Appendix A A.3.2.3
3. Consider the impact of stress gradients.
4. Consider a surfa ce crack growth correction factor in the stress intensity (K) solution.
5. Position the crack at the life-limiting location in each zone.
6. Assum e a 2:1 crack aspect ratio with surface length (2c) equal to twice the depth (a).
7. Assume the def ect area equal to 1/2 the area of a circle with a radius of crack depth (a).
8. B ase the volume on the zone surface face length and on an onion skin thickness of 0.020 in (0.5 mm).
9. Use ave rage air crack growth data.
S urface Corner Zones.
Applicants should:
1. Not model surface enhancements for the test case. The method described in this AC has been calibrated against industry experience without considering the surface enhancement effects, such as shot peening, on predicted crack propagation lives. Therefore, it is inappropriate to include the beneficial effects of such enhancements.
2. Use the maximum principal stress in each zone in the crack growth calculations.
3. Consider the impact of stress gradients.
4. Consider a surface crack growth correction factor in the stress intensity (K) solution.
5. Position the crack at the life-limiting location in each zone.
6. Assume a 1:1 crack aspect ratio with surface length (c) equal to depth
(a).
7. Assume the defect area equal to 1/4 the area of a circle with the radius of crack depth (a).
8. Base the volume on the zone surface face lengths and on an onion skin thickness of 0.020 in (0.5 mm).
9. Use average air crack growth data.
A-5
@ Volume subsurface @ surface R1 = 0.425m (1673 in) Volume surface ~ Subsurface crack Surface crack g cb_ a Comer crack R2 = 0.3m (11.81 in) Csubsmface @ @ @ @ @ @ 0 ® Volumseurface come r Csurface Nole . Eleme nt volume grouped due lo similar FM mode I and stress levels Volume subsurface before breaik - out✓transition after bl'eak-outltrarns-ition surface AC 33.70-34/17/23 Appendix A Figure A-1. Typical Zone Types Figure A-2. Guidelines for Crack Growth Analysis A-6
4/17/23 A.4 A.4.1 A.4.2 A.4.3 A.4.4 AC 33.70-3 Appendix A Test Case Analysis Exam ple.
Problem Description.
The test case geometry consists of a titanium ring disk under simple cyclic loading for 20,000 cycles. The maximum speed is 6,800 RPM, and an external pressure load of 50 MPa (7.25 ksi) is applied on the outer diameter to simulate blade loading. The disk probability of failure will be calculated, assuming no in-service inspection and with a single in-service inspection at 10,000 cycles (see figures A-3 and A-4).
Step 1 - Component Stress Analysis.
Component stresses are determined to perform crack growth analysis, define zones, and calculate zone volumes. The stress analysis results are shown below as a component stress contour plot for the maximum principal stress in each band. Since crack growth calculations are to be performed, maximum principal (hoop) stresses are used (see figure A-5).
Assumptions:
1. Disk is at a constant temperature.
2. No thermal stresses.
Step 2 - Stress Volume Calculation.
Incremental volumes are used to determine the probability of having an anomaly in a particular region of the part. The disk is partitioned into zones where within a zone, the residual life is nearly constant. Next, the volume of each zone is calculated. The disk shown in figure A-6 has been partitioned into 36 zones. The FAA provides guidelines for defining the volume of each zone in paragraph A.3. Stress volume results are shown in Table A-2.
Assumptions:
1. Stress volumes partitioned at 5 ksi (34.5 MPa) increments are good starting points to perform the risk integration.
2. A 0.020 in (0.5 mm) thick onion skin provides adequate definition of the surface volume.
Step 3 - Crack Growth Model Definition.
Crack growth models are constructed for each of the zones defined in paragraph A.4.3.
Examples for zones 17, 22, and 10 are shown in figure A-7 of this AC. The FAA provides guidelines for crack growth analysis in figure A-2 of this AC.
Assumptions:
1. The crack is positioned in the most life-limiting location within the zone.
2. Surface anomalies are modeled as semicircular cracks.
3. Surface corner anomalies are modeled as quarter circles.
4. Subsurface anomalies are modeled as circular cracks.
A-7
4/17/23 A.4.5 A.4.6 A.4.7 A.4.8 AC 33.70-3 Appendix A Step 4 - Crack Growth Calculations.
Crack growth calculations are performed using the predicted stresses and crack growth rate data to determine the residual life associated with each zone (figure A-8). The calculations are conducted for a range of initial crack sizes to ensure that the component service life is covered.
Assumptions:
1. All anomalies act as sharp propagating cracks and are orientated normal to the maximum principal stress: hoop stress.
2. The crack growth rate curve is the same for both surface and subsurface calculations.
3. Average air crack growth data.
4. No surface enhancement effects.
Step 5 - Relative Risk Calculation - No In-Service Inspection.
The probability of failure for each stress volume is calculated, integrating the volume, anomaly distribution, and residual life information from the previous steps (figure A-9).
The results for each zone are statistically summed to determine the total component probability of failure. The calculated probability of failure without an in-service inspection is 1.9E-09 events/cycle.
Step 6 - Relative Risk Calculations - With a Single In-Service Inspection.
The “with inspection” probability of failure calculations are performed in the same manner as in paragraph A.4.6, except the ultrasonic inspection POD data and cycles to inspection are included in the risk integration (figure A-10). The calculated probability of failure with a mid-life inspection is 1.4E-09 events/cycle.
Assumptions:
1. The ultrasonic inspection POD curve is applicable for 100 percent of the component volume (surface connected and subsurface).
2. Inspection is performed at 10,000 cycles.
3. The anomaly area in the inspection plane is equivalent to the anomaly area in the stress plane.
Step 7 – Results.
Numerous manufacturers have performed this test case. A statistical analysis of the results, given in failure risk for each cycle, demonstrated the following statistical values.
A-8
Appendix A T able A-1. Failure Risk Data Failure risk for each cycle Mean value: m m – 1.65s m + 1.65s Without in-service inspection
1.57E-09 1.27E-09 1.93E-09 With in-service inspection
1.13E-09 8.36E-10 1.53E-09 Note 1: All results in the “m-1.65s and m+1.65s” range are considered acceptable for both conditions. A graphical representation is shown in figure A-12.
Note 2: This range defines the interval, centered on the mean value, covering 90 percent of the result population, assuming a log-normal distribution.
A-9