326 Effects of Damage on HCF Properties While the maximum serviceable limits for FOD in some jet engines in operation today range from 0.36 to 0.76 mm (0.014"–0.030"), the damage depths found in service vary significantly (see Appendix G). Of major concern is if FOD as small as sometimes found in service provides a significant fatigue strength debit in simulated engine blade specimens. Whether the damage should be blended out or the component be removed from service and replaced is very difficult to ascertain in many cases. Another item of concern is the determination of whether damage depth is a good indicator of the actual remaining fatigue life of an FOD impacted blade. 7.4. BACKGROUND There has not been a substantial amount of work done relating FOD to fatigue strength. Peters et al. [2] researched the effects of FOD on HCF thresholds by shooting steel spheres onto the flat surface of modified K b (edge cracked) specimens. They found that the overall effect of FOD markedly reduced the fatigue life, compared to that obtained on undamaged smooth-bar specimens, by providing preferred sites for the premature initiation of fatigue cracks. Mall et al. [3] tested both diamond cross section and uniform rectangular cross section samples to study the effects of FOD. The samples were either ballistically impacted or damaged by quasi-static indentation or shearing. They found that the different damage methods created distinctly different damage mechanisms. It was suggested that a total damage depth parameter could be utilized to allow the use of inexpensive and easily controlled methods of simulating FOD, such as the quasi-static chisel indentations, to replace more difficult and expensive means, such as ballistic impacting. Due to the continuing concern of FOD, studies characterizing the damage sites have been conducted in order to understand the effects of FOD on the HCF strength of engine blades. There are several ways of simulating FOD in a laboratory environment which do provide an impact site and fatigue strength debit; however, each method produces different types of damage which could lead to different crack initiation mechanisms. Characteristic material, geometries representing the leading edge of engine blades, and “realistic” impact conditions have been explored. In this section we attempt to quantify the various measurable damage parameters and to determine if they play a role in controlling the fatigue strength in simulated blades. 7.5. FOD DATA MINING AND INVESTIGATION While FOD is not solely an HCF problem, many incidents involve FOD damage and, at some later time, vibratory loading that can lead to HCF failure. It is important therefore, from a damage tolerance perspective, to find damage from FOD before the component is exposed to sufficient HCF loading cycles to lead to failure. For this reason, military turbine Foreign Object Damage 327 engines are inspected routinely to preclude the existence of FOD damage as well as under- stand the nature of FOD before they are returned to service. The steps in a typical FOD investigation have been documented by Franklin and Kleinakis in a NATO report [4]. The procedure includes, in general, the following steps which are extracted from that report: • Visual examination of the inlet and the compressor for missing parts or indications (marks) of foreign material with the use of light source. The first step is to collect every available component from the engine. It is best if the entire engine is available, but investigators can work with less. The most important components are those two stages upstream and two stages downstream of the failed stage. If the FOD is repaired prior to investigation, this will eliminate nearly all of the evidence and make the investigation very difficult if not impossible. Therefore, the first step of collecting all of the available components and debris from the engine is very important. • Visual examination of the area and the stages upstream and downstream of the FOD dam- aged blades. This inspection may be performed using flexible boroscopes in accordance with the manufacturers’ recommendations. Once the debris is collected, the investigators will start looking at the two damaged stages that are farthest forward in the engine. Looking atthesetwostageswillprovideusefulinformationthathelpsseparateprimaryandsecondary damage. The first step is to look at the witness marks. Basically the direction of the damage origination can be determined from the witness marks and that will be used to determine where to focus investigation efforts. If the majority of FOD sites are on the pressure side of the airfoils,there isa good chancethat the damagecame fromthe firstdamaged stageor from in front of that. If the damage is on the suction side, this usually means the damage came from behind which would indicate some sort of stall in the airflow. If none of the engine components actually failed, the witness marks have to tell the entire story. • Disassembly of the engine and examination of the blade. Once the primary damage component is located, the witness marks will be subjected to microscopic inspection using both optical and scanning electron microscopes. The shape of the damage can be determined from these inspections and potential components (such as bolts, nuts, washers) that match the damage geometry can be isolated. After the detection of the most likely damage origination, the broken surfaces must be examined. Hopefully, at least one of the fracture surfaces will have ‘beachmarks’, crack growth striations or the general appearance of brittle fracture – as these are all indications of fatigue. If all of the components have signs of tensile rupture, the cause of failure is difficult to determine. Once the components with fatigue damage are identified, a more complete microscopic examination can be performed. Using a combination of experimentation and numerical analysis, both ‘beachmarks’ and striations can be counted to determine approximately when the damage was initiated. Basically, the prediction is based on the identification of the stresses, in conjunction with experiments in the lab using specimens to simulate the loadings if the necessary data are not available from the bibliography or the engineering data provided from the manufacturer. If there is the appearance of brittle fracture (a relatively flat and featureless fracture surface) without crack growth striations, this can indicate high R-value fatigue as a damage mechanism. This could be indicative of HCF 328 Effects of Damage on HCF Properties or LCF, but the appropriate type of fatigue must fall within the realm of what is predicted numerically. • In addition, a spectrographic inspection of the fracture surface can be used to determine if any material has been deposited on the blades from the impactor. This is crucial to determining which of the geometrically compatible components is most suspect. For example, if the spectrographic inspection of the blade indicates trace amounts of cadmium, and one of the bolts is cadmium coated, then the process of determining what caused the FOD will focus on those bolt locations. When the suspected impactors are narrowed down to the smallest possible list, the locations of the suspects are inspected. If a failed component is found, the fracture surface is inspected to determine if it failed before the larger incident. If it did, this will probably be called the cause of the incident. Nearly all of the investigation procedures are useless if heating is involved in the incident. Heating can and will distort fracture surfaces, melt material, and just make a general mess of things. This is the general FOD incident investigation process. For determining what caused a damaged component to fail, optical and scanning electron microscopes are used almost exclusively (Figure 7.2). If these workhorse tools do not indicate what went wrong with the part, then the investigators will use any possible tool to determine the cause of damage, but the majority of these tools are mostly only proven in a laboratory setting. 7.6. DEFINITION OF FOD Similar to the use of the terminology HCF, there is no formal definition of FOD other than “foreign object damage.” The foreign object is typically either a soft object like a bird or ice, or a hard object like a stone, wrench or other similar object ingested into an engine. While these could formally be classified as FOD events, the mid-air collision of J85-GE-4A 1st stg blades (FOD) Figure 7.2. SEM photo of a damaged blade (FOD). Foreign Object Damage 329 Figure 7.3. F/A-18 aircraft after suffering severe FOD from mid-air collision. two F/A-18 aircraft is not treated in any analysis and is not considered as potential FOD in the design process. Figure 7.3 shows that the radome, radar and all of the avionics equipment, everything, are gone. This “FOD incident” created several problems for the pilot: aerodynamics, eventual loss of hydraulics due to loss of fluid, navigation, and probably the most amazing, as the pieces fell away, some debris had to be ingested by the engines (real FOD) and he still was able to bring it home. The story behind this is that two F/A-18 Hornets made a head on pass, just a bit too close. One got home with part of the left wing and left vertical fin and rudder missing, while the other jet, shown in Figure 7.3, is missing everything forward of the cockpit pressure bulkhead – and is a flying convertible because the canopy is shattered too. This illustrates the extreme nature of FOD and the possible consequences. In this chapter, we will deal primarily with small hard objects that are commonly ingested into engines. 7.7. TYPES OF DAMAGE When an engine blade experiences FOD, the damage can be in many different forms. Some of these, like curling, denting, and distortion, have little effect on the fatigue properties of the component. In a recent report [5], in Annex C, the definitions of the 330 Effects of Damage on HCF Properties various forms of engine blade damage from FOD have been summarized along with photos showing examples. Figures 7.4–7.15 are taken directly from that report, defining the different types of damage. Burred: Rough edge or sharp projection on edge or surface of parent material. Most commonly associated with surge damage. Figure 7.4. Example of a burred blade. Chipped: Breaking away of surface of parent material usually caused by heavy impact (not flaking). Most commonly associated with snubber abutment face damage. Figure 7.5. Example of a chipped blade. Foreign Object Damage 331 Cracked: Partial separation of material which may progress to a complete break, either visible or detected by NDT. Figure 7.6. Example of a cracked blade. Curled: Condition where edges of outer portion of blade have been rolled over. Often associated with rubbing on casing or vanes, or sometimes soft body. Figure 7.7. Example of curled blades. Dented: Indentation with rounded bottom, usually on leading/trailing edge, sometimes on surface. Parent material is displaced, seldom separated. 332 Effects of Damage on HCF Properties Figure 7.8. Example of a dented blade. Deposits: Build-up of material (on blades and vanes). Figure 7.9. Example of blades with deposits. Disintegrated: Separated or decomposed into fragments. Excessive degree of fracturing (breaking) as with disintegrated bearings. Complete loss of original form. Distorted: Extensive deformation (whole or major part of aerofoil) of original contour of part. Foreign Object Damage 333 Figure 7.10. Example of a distorted blade. Erosion: Gradual removal of metal due to continuous action of impact from very small particles. Gouged: Scooping out of material, usually associated with surface of aerofoil. Figure 7.11. Example of a gouged blade. Nicked: Relatively short, sharp indentation, usually associated with leading/trailing edges. Parent material is displaced, sometimes removed. 334 Effects of Damage on HCF Properties Figure 7.12. Example of a nicked blade. Peeled: Breaking away of surface finishes such as coating, platings, etc; peeling would be flaking of very large pieces; blistered condition usually precedes or accompanies flaking. Piece Out: Complete material removed from leading/trailing edge. Figure 7.13. Example of a blade with a piece out. Pitted: Small irregular shaped cavities in surface of parent material. Scored: Relatively long, deep scratch or scratches made by sharp edges of foreign parti- cles. Usually associated with aerofoil surfaces. Foreign Object Damage 335 Figure 7.14. Example of a pitted blade. Scratched: Light, narrow, shallow mark or marks caused by movement of sharp object or particle across surface. Material is displaced, not removed. Torn: Separation by pulling/ripping apart. Figure 7.15. Example of a torn blade. As noted above, not all of these damage modes are related to HCF. The specific types of damage treated in this book are, using the definitions above, burred, chipped, cracked, nicked, pierce out, scored, scratched, and torn. These, in turn, can be analyzed using some of the tools dealing with notches, particularly including the extension to cracks and short crack/shallow notch behavior discussed in Chapter 5. In some cases, these tools are inadequate because of the many complex mechanisms associated with FOD. . deformation (whole or major part of aerofoil) of original contour of part. Foreign Object Damage 333 Figure 7.10. Example of a distorted blade. Erosion: Gradual removal of metal due to continuous action. necessary data are not available from the bibliography or the engineering data provided from the manufacturer. If there is the appearance of brittle fracture (a relatively flat and featureless fracture. sharp projection on edge or surface of parent material. Most commonly associated with surge damage. Figure 7.4. Example of a burred blade. Chipped: Breaking away of surface of parent material