The parts used in turbine engine construction are often very expensive. The costs of a finished part can often be comparable to a vacation (turbine blades) or a luxury car (exit housing, rotor). In addition, such complex parts require very long processing times, from months to years. If there are delivery delays, additional high costs can be incurred. This makes the reworking and use of flawed parts an attractive option where it is possible.
In the following, “reworking” is understood to be work done in the finishing process that makes parts with deviations conform to specifications; i.e. makes them suitable for delivery. Steps in the normal specified finishing process, such as burr removal, are not considered “rework” in this context.
Reworking removes the deviation or flaw using suitable methods that are to be designed for the specific case by the responsible technical departmentments. Reworking may involve a complex procedure (Ill. 7.5-1). In certain cases, such as if the part is considered acceptable by experts but deviates from the usual series standard (e.g. minor local dimensional deviations), approval from rating authorities and clients may be required.
There are cases which merely involve the correction of an unusual optical appearance, such as discoloration, that obviously does not influence the operating behavior of the part. In other cases, such as arc burning (Fig. "Causes of local overheating"), part safety may be affected by this influence. In this case, reworking is absolutely necessary from the standpoint of part safety.
Even reworking that seems unproblematic at first glance can expectedly affect the operating behavior/functioning of the parts (Fig. "Surface reworking influencing operation behavior i").
Figure "Minimizing scrap rates throuch reworking": Reworking can be a very demanding task. Depending on the deviation/flaw, a large number of steps may be necessary. The ones discussed here are a case-specific selection that should be the responsibility of the proper technical departments. Specific approaches have already been discussed in problem- and process-related chapters:
- Fig. "Measures preventing fouling": Surface changes related to fouling.
- Fig. "Reworking at local overheating": Damages through electrical arcs.
- Fig. "Preventing cracking and remedies during finishing": Micro cracking.
Fundamentally, spontaneous attempts at mechanical surface reworking must be avoided. These can only be done with the approval of a specialist who has seen the area in question in its uninfluenced state. Often, sand paper or polishing materials are used to determine the depth of a flaw/deviation. This destroys important information necessary for determining the cause and threat level (Ref. 17.5-1), such as tarnishing or microscopic damages (e.g. melting, diffusion, corrosion). This makes evaluation and decisions regarding reuse more difficult.
Even if the anomaly was apparently successfully removed without noticeably altering the part, there is still a danger of unacceptable damage. For example, structural changes may have reduced part strength or tensile residual stresses may remain. This may be the case when metal droplets or welding sparks strike a surface. Larger damage zones may be present under seemingly harmless “baked-on” particles (Ills. 220.127.116.11-2 and 18.104.22.168-3). These zones can be recognized and assessed with regard to their reworkability with the aid of metallographic procedures that are virtually non-destructive (Fig. "Metallography and SEM").
Important steps in the framework of reworking:
Prerequisites: Even if a deviation/flaw is discovered, possible personal consequences as well as time and cost pressures mean that there is no motivation to report the findings. Therefore, additional motivation is required. The primary concern must be objective problem-solving and not personal consequences (Fig. "Negative motivation"). Another requirement is a simple, widely known, documented “reporting system”. This includes the steps of the process and the responsible decision makers.
Every spontaneous attempt at surface treatment can at the very least impede the determination of the causes and the possibility of acceptable reworking!
The damaging effect of a flaw can be considerably larger and deeper than it first appears. This must be taken into consideration during reworking!
Conclusions regarding deviations: The prerequisite here is the identification of the damage type and the determination of the damage cause. This can be done with the aid of problem analysis (Ill. 17-11). In this context, the geometric extent of a potential damage is especially important. Useful aids include metallography (Fig. "Metallography and SEM") and SEM inspections (Fig. "Scanning electron microscopy (SEM)"), which can be done non-destructively, at least in the initial phases (impressions, inspection of the surface on location, Fig. "Non destructive microscopic inspection").
Risk estimation: Here, as well, there is a systematic approach with suitable analyses (Ill.17-11), which must of course be conducted in cooperation with the responsible or affected technical departments.
The primary concern is part safety. Research in handbooks and maintenance and repair specifications can be very useful in this regard. These sources can also provide initial ideas concerning possible reworking options.
Conceiving the rework process: This is also a task for the responsible technical departments. Criteria such as possible influencing of the operating behavior, time frames, costs, and acceptance (clients, authorities, etc.) must also be considered.
Acceptance of the reworking: In most cases, especially if there is an obvious connection with safety issues, the approval of the decision-makers must be obtained. The sufficient explanation of the damage cause and process are very helpful in this regard. In addition, the extent and location of already delivered parts should be known (also see Fig. "Minimizing finishing problems by documentation"). It may be necessary to reinforce the safety of a reworking procedure with technical test-based verifications.
“Lessons learned” - preventive measures: First, one must determine whether or not delivered parts are already in operation. For these, suitable testing and repair measures must be developed and coordinated with the responsible authorities in accordance with the relevant regulations.
In addition to reworking, preventive measures include ensuring that no further parts have these deviations/flaws. This means that the finishing process must be stabilized and/or additional verifying tests are necessary. An effective tool for detecting weak points in the finishing process is FMEA (Ills. 17-10 and 17-11).
Figure "Negative motivation" (Ref. 17.5-1): Psychologists understand that positive motivation is more successful than negative motivation.
Negative motivation means forcing behavior through the threat of punishment. Fear and frustration are important factors in this process. This type of repressive approach suppresses the essential creative cooperation of the affected persons towards solving the problem. Positive motivation refers to using encouragement and information to motivate people to actions that are understood and actively supported. The modern approach of a superior should orient itself to this latter method in order to prevent and solve problems. It is based on teamwork.
Therefore, quality is a product that relies on team spirit and trust. In order to prevent damages and high scrap rates, the approach should be similar to the healing of diseases: the earlier the problem is recognized and the correct diagnosis made, the greater the chances of a full recovery and the lower the costs and time required for therapies/corrective measures. For this reason, the undoubtedly uncomfortable situation of an early “doctor visit” is especially important. If it turns out that there is no problem, then we can congratulate ourselves. Accusations of excessive sensitivity are completely counterproductive.
The following recommendations are based on these considerations:
- Superiors should realize that objective behavior and trust are prerequisites for the sufficiently early reporting of quality problems. They should, for example, explain to the technicians the importance of their work for the quality and operating performance of the parts.
- The technical solution of the problem and the development of corrective measures should also be the focus of our actions. The involvement of personnel in the solutions can be an important motivator.
- No one should experience negative repurcussions for reporting flaws or deviations in finishing processes or parts. Superiors should motivate in a positive way.
- If in doubt, it is always better to raise ones hand too often than not often enough.
- We learn from our mistakes and problems. Solutions and corrective measures should be recoreded in appropriate guidelines and documents.
Figure "Surface reworking influencing operation behavior i": Some types of reworking are intended to correct apparently secondary, finishing-related deviations in order to make parts conform to specifications. Even if the responsible agencies, authorities, licensor, or customer agree to this process, there can be problematic cases. These can often be “hidden” and easily overlooked. These cases typically concern changes in the operating behavior of the part.
This diagram contains several examples.
Rubbing surfaces (“1”) are generally designed to minimize the clearance gap at blade tips. There is usually a hard, thermally-sprayed ceramic coating (Al2O3) on rotor spacers opposite compressor stator vanes without inner shrouds. For aerodynamic reasons, the coating surface should be as smooth as possible. However, tribological demands call for a rough, cuttable coating. This allows rubbing to cause material removal at the blade tips, but without the following risks:
- dangerous heat development in the blade tip and rotor;
- accelerated material buildup on the coating (Volume 2, Ill. 7.1.3-7) with dangerous intensifying of the rubbing process;
- mechanical overstressing of the blades, which could be either static (Volume 2, Ill. 7.1.3-8) or dynamic (Volume 2, Ill. 7.1.3-4). If a tested, approved, and proven rough coating is mistakenly smoothed during incorrect working or reworking, the risk of serious damage in case of rubbing will increase (Volume 2, Ill. 7.1.3-6).
Labyrinth tips ( “2”) are usually armored with a hard, thermally-sprayed ceramic coating. If the clearance gap is bridged, the opposite surface, which is usually a soft abradable coating, will be worn down. The resulting friction heat should be as low as possible, at least at the labyrinth tips, in order to prevent dangerous self-increasing heating process in the system, especially the rotating labyrinth ring (Volume 2, Ills. 7.1.3-8 and 7.2.2.-4). The rougher (i.e. more cuttable) the armor, the less heat is created. If, for example, an armored coating is reworked (diamond file) and smoothed for dimensional reasons, its rubbing behavior can worsen dangerously. For this reason, even minor reworking of rubbing surfaces and armored coatings is problematic. If rubbing occurs, smooth armoring can cause unusually high temperatures with cracking, structural changes, smearing, and in extreme cases total failure of the labyrinth.
An additional damage-promoting factor is armoring that has become unusually thin in local areas following reworking. It loses its insulating effect and can no longer help to prevent overheating of the base material.
If armoring was removed from around the labyrinth tips, it can be assumed that the rubbing behavior will be considerably worse. In addition, there is a danger that oxidation will increasingly penetrate along the bond surface from the tips during operation. This can result in high thermal stresses, separation of the armor, and further consequential damages.
Blade tips reinforced with hard particles (“3”): This technology is considered the state-of-the-art in rotor blades in the high pressure compressor and high pressure turbine (Volume 2, Ill. 7.1.4-14). Reworking of these particles, such as through grinding in the rotor, compromises their cutting performance and therefore also their intended function with regard to clearance gap minimization.
Reworking hardened surfaces: Parts are shot-peened in order to increase or ensure dynamic fatigue strength (Fig. "Increasing fatigue strength by shot peening"), as well as to minimize any fretting influences (Volume 2, Ill. 6.1-19). This process induces positive compressive residual stresses and hardening in a thin surface zone (about 0.1 mm; Fig. "Blasting (Almen) intensity"). The unique surface topography of shot-peened contact surfaces (e.g. on blade roots) can contribute to improved fretting behavior (Fig. "Fretting damage loweing by shot peening"). Material removal during reworking can eliminate these advantages and have a dangerous effect on operating behavior.
Smoothing coated surfaces (“4”): Baked-on material and raised areas (“warts”) on diffusion-coated surfaces can impede their function, especially if the surfaces are aerodynamically significant or designed to create contact seals. Material-removing reworking can lead to micro cracking and material breaking out from brittle coatings. This compromises their designed function, i.e. protection from oxidation or corrosion, and can affect the life span of the parts.
Removal of the “e layer” from salt bath nitrided (carbo-nitrided) parts (“5”). In contrast to gas nitriding, the nonmetallic characteristics of this very thin coating ensure emergency running properties that prevent the parts from seizing under mixed friction. This can be a requirement for the functioning of regulator valves with pistons that rotate in fuel at high rpm rates in fuel.
17.5-1 A.Rossmann, “Unser Beitrag zur Qualitätssicherung”, 1998, Turboconsult.
17.5-2 Metals Handbook “Volume 11, Failure Analysis and Prevention”, ASM 1986, ISBN 0-87170-007-7, pages 16-65.