AC 33.70-5
• Surface Credit for Process Monitoring cannot be claimed unless Fluid Monitoring and Machine Condition, Fluid & Fluid Condition, Tooling, and Setup Validation are also in use.
F.3.2 Restrictions.
• Use of Surface Credits requires a Process Validation and Control Plan.
• Use of Edge Credits requires a Process Validation and Control Plan.
• Surface Credits cannot be used for edges.
• Edge Credits cannot be used for surfaces.
• Non-destructive inspection(s) at manufacture cannot be accounted for in the analysis to lower the POF value before application of manufacturing credits.
F.4 Credit Application Process.
1. Decide which controls are to be applied.
2. Sum the credit values for the combination of controls.
3. Calculate the number of effective slots.
4. Calculate the probability of fracture for the axial slot feature in the disk.
5. Divide the probability of fracture by the sum of the credit values.
6. This reduced probability of fracture should be compared to the DTR.
F.5 Slot Surface Credit Examples.
F.5.1 An axial slot feature with 100 slots, N:
a. Applicable Credits:
Process Validation and Control Plan 5 Total Credit 5 Number of Effective Slots: a = 0, Neff = 90 + (1-0)*(N-90) = N = 100 Initial design POF using Neff: 5.0E-4 events per service life POF with controls is now: 5.0E-4 / 5 = 1.0E-4 events per service life This is not adequate against Axial Blade Slot DTR.
b. Applicable Credits:
Process Validation and Control Plan 5 Machine Condition, Fluid & Fluid condition, Tooling and Setup validation 5 F–8
AC 33.70-5 Fluid Monitoring 5 Surface Condition inspection – nondestructive 5 Total credit 20 Number of Effective Slots: a = 10/20=0.5, Neff = 90 – (1-0.5)*(N-90) = 95 Initial design POF using Neff: 4.5E-4 events per service life POF with controls is now: 4.5E-4 / 20 = 2.25E-5 events per service life This is not adequate against Axial Blade Slot DTR
c. Applicable Credits:
Process Validation and Control Plan 5 Machine Condition, Fluid & Fluid condition, Tooling and Setup validation 5 Fluid Monitoring 5 Process Monitoring 10 Surface Condition inspection – nondestructive 5 Total credit 30 Number of Effective Slots: a = 0.9, Neff = 90 – (1-0.9)*(N-90) = 91 Initial design POF using Neff: 4.0E-4 events per service life POF with controls is now: 4.0E-4 / 30 = 1.3E-5 events per service life This is acceptable to Axial Blade Slot DTR.
F.6 Slot Edge Credit Examples.
F.6.1 An axial slot feature with 100 slots, N:
a. Applicable Credits:
Process Validation and Control Plan 5 Total Credit 5 Number of Effective Slots: a = 0, Neff = 90 + (1-0)*(N-90) = N = 100 Initial design POF using Neff: 2.5E-4 events per service life F–9
AC 33.70-5 POF with controls is now: 2.5E-4 / 5 = 5.0E-5 events per service life This is not adequate against Axial Blade Slot DTR.
b. Applicable Credits:
Process Validation and Control Plan 5 Edge Processing: Manual Finishing 8 Total credit 13 Number of Effective Slots: a = 8/15, Neff = 90 + (1-8/15)*(N-90) = 95 Initial design POF using Neff: 2.0E-4 events per service life POF with controls is now: 2.0E-4 / 13 = 1.54E-5 events per service life This is acceptable against Axial Blade Slot DTR F–10
AC 33.70-5 APPENDIX G. SURFACE (MANUFACTURING INDUCED ANOMALY) DAMAGE
TOLERANCE ASSESSMENT FLOWCHARTS G.1 Overview.
Note: For more details on the steps described in the following figures, refer to chapter 3 and the appendices.
For the purposes of this section, the definition of a "blade slot set" is consistent with the description in appendix E, paragraphs E. 1 and E.2.
G.2 Process.
1. Assess the Probability of Fracture for One Blade Slot.
2. Calculate the Probability of Fracture for the Entire Blade Slot Set.
Blade slot set list Identify all blade slot sets in the engine Slot set 1 Slot set 2 Slot set 3 For each blade slotset:
Improve:
propagation Crack Blade slot set risk -Slot manufacturingor Slot analysis/design or analysis assessment Blade slot set risk < DTR ? Slot in service (see Figure G-2) (see Figure G-3 inspection or and Figure G-4) no Inspection trequency yes End J Figure G-1. Surface Damage Risk Assessment Work Flow G-1
AC 33.70-5
Blade slot set Identify critical locations on blade slot edges and surface Determine crack propagation model (plate, crack prop.
law) Temperature & Stress results Run crack propagation analyses on edges and surface Crack size vs number of flights for corner crack "a vs Ne" Crack size vs number of flights for surface crack "a vs Ns" Figure G-2. Crack Propagation Analysis
3. Begin by selecting one slot from the blade slot set.
4. Calculate the probability of fracture for this single blade slot using the established damage tolerance criteria.
G–2
AC 33.70-5
Crack size vs number of flights for corner crack "a vs Ne" Crack size vs number of flights for surface crack "a vs Ns" Calculate probability of fracture (figure G-6) Slot edge probability of fracture "sepf1" for 1 slot of set i (figure G-6) Expected component service life Inspection plan and appropriate POD curves Anomaly distribution (appendix B) Slot surface probability of fracture "sspf1" for 1 slot of set i (figure G-6) Figure G-3. Feature Risk Assessment (Part 1 of 2)
5. Use the probability of fracture for one slot (from s tep 1) to calculate the probability of fracture for the entire blade slot set.
6. Apply the appropriate manufacturing credits and effective slot numbers as defined in previous sections to complete the assessment.
G–3
AC 33.70-5 Manufacturing Credits for Effective number of slots for edges "emc" edges "nese" (see Figure G-5) Determine the probability of fracture "sepf" for the Slot edge probability edges of the slot set of fracture "sepf1" for (see Figure G-6) 1 slot of seti sepf = 1 - (1 - sepf1 / emc) nese Keep the worst probability of fracture between sepf and sspf values for the slot Slot surface Determine the probability "sspf" of set to obtain the probability of fracture "sspf1" for 1 slotof fracture for the surface of the slot set blade slot set risk.
set i (see Figure G-6) sspf = 1 - (1 - sspf1 / smc) ness Manufacturing Credits for Effective number of slots for surfaces "smc" surfaces "ness" (see Figure G-5) Figure G-4. Blade Slot Set Risk Assessment (Part 2 of 2) G-4
AC 33.70-5 Manufacturing Credits associated with Automated Edge Processing being applied "emac" Determine the effective number of slots for edges "nese" if n <= 90 : nese = n Effective number of slots for if n > 90 : edges "nese" nese = 90 + (1-(emac/emac_max)* (n - 90) emac_max = max possible manufacturing credit for Automated Edge processing Number of slots "n" Determine the effective number of slots for surfaces "ness" if n <= 90 : ness = n ness = 90 + (1-(smac/smac_max)* (n - 90) if n >90 : Effective number of slots for surface "ness" smac_max = max possible manufacturing credit for Automated Surface processing Manufacturing Credits associated with Automated Surface Processing being applied "smac" Figure G-5. Determination of the Number of Effective Slots in the Blade Slot Set G-5
AC 33.70-5 The POF calculation is described here following a Monte Carlo simulation process. A large sample of anomalies should be simulated. Each simulation in the sample follows the flowchart below to determine if it:
- fractures before the expected component service life meets or exceeds the expected component service life
- is detected with a crack Probability of fracture before N flight cycles (sepf1, sspf1 from Figure G-3 = (number of fractured slots before N flight cycles) / (total sample size) manufacturing anomaly? Is there a yes What "ao"? size life from initial flaw size, Calculate crack growth inspection(s)? frequency of Number and no Success Any inspection no >= the expected Is anomaly life yes during anomaly component Success life? service life?
no yes Fracture (1) Is anomaly detected at Calculate anomaly size inspection(s) using yes at each inspection(s) applicable POD curve? Detection (1) Notes : no
- Each requires a random pick from the pertinent random variable distribution Fracture (1)
- (1) record the number of flights at fracture or detection for each slot simulated Figure G-6. Probability of Fracture Calculation G-6
AC 33.70-5 APPENDIX H. DE FINITIONS The following definitions apply for the purpose of this AC.
H.1 Axial Blade Slot.
Axial blade slots are nominally identical features, with typically one per blade root.
They are located at the same nominal radius, usually in the rim of a disk, and their geometry provides a mechanical constraint that positions the blades radially. Appendix E of this AC provides more information.
H.2 Component.
A part that may contain multiple sets of component features.
H.3 Component Feature.
A unique location, structural shape, or surface of a component or part. For a disk, component features are blade slots, bore, or web locations. Other examples are grooves, slots, or holes. Usually, different machining processes and practices are required to produce each unique feature.
H.4 Damage Tolerance.
An element of the life management process that recognizes the potential existence of component imperfections that are the result of inherent material structure, material processing, component design, manufacturing, or usage. Damage tolerance addresses this situation through the incorporation of fracture resistant design, fracture mechanics, process control, or non-destructive inspection.
H.5 Default Probability of Detection (POD) Values.
Values representing mean probabilities of detecting anomalies of various types and sizes. These values are accepted as valid provided the inspection is conducted consistent with applicable industry specifications.
H.6 Design Target Risk (DTR).
The relative risk of failure caused by material, manufacturing, and service induced anomalies and the standard against which probabilistic assessment results (stated in terms of feature, component, and engine level event rates) are compared.
H.7 Event.
A rotor structural part separation, failure, or burst, with no assessment of consequence.
H–1
AC 33.70-5 H.8 Exceedance Curve.
The distribution of anomaly size and the frequency at which anomalies occur.
H.9 Focused Inspection.
A component inspection in which the inspector has been instructed to pay special attention to specific component feature(s) or location(s) and the necessary specialized processing instructions have been provided.
H.10 Full Field Inspection.
A component inspection without special attention to any specific component feature(s) or location(s).
H.11 Hard Time Inspection Interval.
The number of engine flight cycles since new or since the most recent inspection after which a rotor part must be made available and must receive the inspection specified in the airworthiness limitations section (ALS) of the instructions for continued airworthiness (ICA).
H.12 Hazardous Engine Effects.
Any of the effects listed in § 33.75(g)(2). Failure of rotating parts usually results in the non-containment of high-energy debris, which is the primary hazardous engine effect addressed by this AC. Disks, hubs, impellers, large rotating seals, and other similar large rotating components represent potential sources of high-energy debris.
H.13 Inspection Opportunity.
An occasion when an engine is disassembled to at least the modular level and the hardware in question is accessible for inspection, whether or not the hardware has been reduced to the piece part level.
H.14 Linear Elastic Fracture Mechanics.
For the purposes of this AC, linear elastic fracture mechanics is a method used to predict crack growth rates and component lives using the stress intensity factor (K) to characterize t he stress state near the tip of a crack caused by a remote load or residual stresses.
H.15 Low Cycle Fatigue (LCF).
The process of progressive and permanent local structural deterioration occurring in a material subject to cyclic variations, in stress and strain, of sufficient magnitude and number of repetitions. The process will culminate in detectable crack initiation typically H–2
AC 33.70-5 within 105 cycles. A detectable crack initiation is defined as 0.030 inches in length by
0.015 inches in depth.
H.16 Maintenance Exposure Interval.
A statistical distribution which defines the number of flight cycles between engine shop visits. This interval provides an opportunity to inspect the engine internal parts because the module or parts have been disassembled as a result of normal maintenance activity.
H.17 Manufacturing Anomaly.
A surface-related imperfection introduced during the manufacturing process (subsequent to melting) that is considered potentially detrimental to the structural integrity of a life-limited rotating part during its service life.
H.18 Mean Probability of Detection (POD).
The 50-percent confidence level POD versus anomaly size curve.
H.19 Module.
A combination of assemblies, subassemblies, and parts contained in one package, or arranged to be installed in one maintenance action.
H.20 Operationally-Induced Anomaly.
A surface-related imperfection introduced during service operations (both flight operations and maintenance) that is potentially detrimental to the structural integrity of the life-limited rotating part.
H.21 Part Available.
A part that can be inspected, as required by the ALS of the ICA, without any further disassembly. Depending on the inspection requirements, some parts may require a fully disassembled “piece part” condition, while other parts may be available for inspection while still in the assembled module.
H.22 Probability of Detection.
A quantitative statistical measure for detecting a particular type of anomaly (flaw) over a range of sizes using a specific non-destructive inspection technique under specific conditions. Typically, the mean POD curve is used.
H.23 Probabilistic Risk Assessment.
A fracture-mechanics based simulation procedure that uses statistical techniques to mathematically model and combine the influence of two or more variables to estimate H–3