Industrial Gas Turbines

4.3.2 The problematic of cheaper spare parts

The car owner puts the scene concretely before our eyes. The exhaust gives up the ghost and an expensive new part is necessary. There are often astoundingly cheap spare parts available, that do not stem from the original producer! This can be obvious or it can have slipped our attention that only an apparently original part is present. In the meantime one discovers, already at the time of assembly, that measurements do not exactly correspond with each other and one has to use force during assembly. Already then, doubts are expedient. If one made the effort to insert such a part in exchange, one would not rarely have to mourn the failure (against the original part) after an extremely short time of operation.

Why should industrial gas turbines escape the misfortune of qualitative deficient spare parts of a dubious origin (Suspect Unapproved Parts = SUPs)? Even in the area of especially well monitored flight engines, one has to fight against this problem. From this we can learn.

Also all parts of a gas turbine can be purchased from obscure sources. According to the experience, especially machine elements like bolts, clamps, pipe line connections and bearings are concerned.

There are various reasons for the existence of parts, that are not approved by the manufacturer. Such parts are:

not produced according to specification,
not repaired according to recommendation,
do not correspond to the permitted type,
do not correspond to general specifications and norms.

The origin of not accepted parts (suspect unapproved parts = SUPs) can be very different:

parts that have been produced without being sanctioned,
rejected parts which where not scrapped,
parts from military sources (not needed anymore, ‘surplus’). Such parts can have distinctive features which can unexpected affect the operation behavior,
parts from bankruptcy assets,
re worked life limited parts,
falsified parts,
parts from cannibalized engines with non identified damages.

Such parts can, by all means, be serviceable. It is, however, to be accepted that in the case of failure the manufacturer of the engine could refuse a guarantee. Perhaps insurance is also impaired.

Outward features ( "Ill. 4.3.2-1") of suspect parts are:

identity markings, that are false or indecipherable or totally lacking,
markings that deviate from the marking procedure of original parts, or such that are brought onto deviating positions,
unusual surfaces (roughness, brightness, discoloration, machining marks etc.) that deviate from the original part.

The surest method to avoid such problems is always the guarantee of the manufacturer (OEM).

Note:
Repair parts from untraceable sources or those with an unusual appereance should only then be acquired when, prior to that, because of the availability of typical trial components, it has been established that the parts correspond to the manufacture’s instructions!

"Illustration 4.3.2-1": (Lit. 4.2-4): There are part-specific features that lead to a SUP-suspicion. Naturally by far not all anomalies are an evidence for a SUP. If the suspicion is legeitimated will show not until a detailed investigation. Generally counts:

Unusual is suspect!

Geometric features: This is a question of deviations, outside the design tolerances, and specifications. Even within tolerances, that lie differently than so far usual, at the upper or lower limit can attract our attention. So the mating forces of centerings can differ markedly from the usual.That enhances the detectability but also can trigger during mating damages by galling. Eye catching radii geometries and sharp edges are able to raise with a damage unrealized local loads (notch effect). That are able to dangerous lower the fatigue strength. This can be a hint at SUPs if such failures occur If passed through openings are concerned, sharp edges and burrs are quite in the position to influence the operation behavior (e.g., combustion chamber and control unit) impermissible.

Machining features: Besides the already addressed burrs and badly rounded edges there are further symptoms. To those belongs an unusual orientation of machining marks/scratches. Normally the OEM will pay attention in his production specifications, that those don’t run transverse to the critical loads. A deviation can be evaluated as a sign to look closer at the origin of the part. Roughness, even within the specification, can differ quite markedly in the topography (structure), even within the specifications. A feature can be a dull or reflecting/bright surface. That is not only a outwardness. This can be a hint at a changed production process and a distinct influence at the strength. A dull surface can be caused by an abrasive blasting process which shall cover a rework. Dull surfaces develop also by etching processes. Are those not planed/specified, there is the danger that a forbidden etching attack exists, which can lower the fatigue strenth unacceptable.

Shot peening surfaces are normally required to guarantee locally the necessary fatigue strength. Are typical indications for a shot peening treatment missing, this concerns a dangerous aberration. For example the roots of many turbine rotor blades are shot peened. The boundary or the peening pattern of the shot peened region can be visuell evaluated by an expert. It is most alarming if the finding does not accord with the original parts. To this belongs, if the (brittle) diffusion coating as oxidation protection is also blasted/peened. In such a case an examination because of SUP-suspicion is required.

Discolorations are an important feature for an abnormality. To this belongs also the distribution, intensity and colouring of tempering colours. They must not itself influence the operation behavior, but can be a hint at aberrations in the production. Indications of rust can stand in connection with an unsuitable packaging which itself causes suspicion.

Traces of rework at new parts and repaired parts shold draw our attention if they are unusual. This can show at a locally changed appearance of the surface (machining marks, polished areas, flat sinkings). There always is the suspicion, that here existed a damage that was not removed in accordance to the specifications. It is especially alarming when those marks are on new parts at particular high stressed zones which determine the life time. At least such features must be in accordance with the relevant specifications (drawing, manuals etc.).

Weldings and brazings: Anomalies in the appearance are noteworthy. A very rough brazing can hint at an insufficient fusing due to aberrant brazing parameters or braze characteristics. Also eye-catching wide seams are suspect. This is also true for unusual, bad looking weldings, which possibly are connected with not adequate competent welding employees.

Coatings can be an important indicator. Primarily aberrations in color and roughness attract attention. But also we have to look at unusual failures and flaking. They can be a hint at adhesion deficits and thereby at process deviations. Possibly with a not accredited producer. But also licensed producers don’t rise above every suspicion. So, before a longer time, a case emerged when instead of a coating (diffusion?), turbine blades where only painted.

Assembly behavior: Under this should be understood abnormalities and problems during pushing together /sliding parts.Tight tolerances even in the specified limits can increase the assembling forces markedly and so strike the assembler if he had not yet experienced this effect in such a pronounced manner. Also the sliding behavior, respectively the friction due to changed machining structure can produce such effects.

Magnetic features will be rather noticed by chance. Possibly magnetic tools or consumables will stick at a component that actual should be of a nonmagnetic nickel alloy or a titanium alloy. Contrary, a part known as magnetic, does not show this feature as pronounced as expected. Such effects are indications for material deviance and with this of an unacceptable part.

Identification mark as „identity card“ of a component/part may not show suspicious aberrations. Regard at addings (numbers, signs) from which can be concluded at rework or changed versions. Also the marking technique that was used can give hints. It is alarming if the marking is localized at an unsual place on the part.

Packings/wrappers are indeed only an indirect indication for the legitimacy of the contained part/component. However, deviations from the usual, are at least worth an enquiry. Are new parts concerned, the integrity of the original packing is important. Unnormalities like an conspicous conservation oil (e.g., smell, colour, consistency) should also attract our attention.

Example 4.3-1: Failure limits allowed by the manufacturer (cracks, deformation) as well as the accepted extent of repairs, are high for turbine guide vanes, because of the specific operation loads, as compared to rotor blades. Thus there are virtually artists at repair shops, who undertake repair brazing, in the framework of the specifications of the manufacturer. Naturally, the same strength, as from a new blade, is not to be expected, even if the manufacturer permits it and a further overhaul interval is given. Further repair cycles are either not permitted or strongly limited. The employed, high temperature solder has, indeed, an adequate high melting point (diffusion of the melting point reducing element additions during brazing) but creep stability and ductility (plastic deformation capability) are less than in new parts. Consequently, a thermal fatigue crack initiation is quicker and to be expected more often. The brazing is to be seen more as a sealing and filling up and not as a regeneration of the new part strength. In this connection, one has to be wary of cheap offers of repair parts, where the exact origin is not definitely traceable. In one case, repaired turbine guide vanes were offered cheap on the international market. The inspection of some samples showed, ironically, that only the serial number was not brazed and, therefore, the blades did not fulfill the specifications of the manufacturer.

Example 4.3-2: After long service, a gas turbine suffered during pipeline operation a bearing damage in the front compressor area with extensive secondary damages. Examination showed that the bearing carrier of the compressor shaft made from titan was not, as required, coated with wear protection as in the aeroengine version already introduced for some time.Through this missing protective coating, (tungsten carbid = TC) an extreme fretting with a loosening of the inner bearing ring, rotating on the shaft, overheated it so much that a shaft fracture took place.

Example 4.3-3: gas turbine on an oil drill platform broke down a relatively short time after the overhaul in a shop, through the fracture of a rotor blade in the high pressure turbine. Examinations showed that apparently the failed blade was deformed plastically, through mechanical effects during static, and in the root area cracked through a forced rupture (not apparent in the assembled condition). This could only have happened through a previous overhaul. A plausible explanation was, that the rotor abutted against the guide vanes, in the shop and was damaged without being noticed during the assembly work.

← previous page

start

next page →

Table of Contents

4.3.2 The problematic of cheaper spare parts