Aeroengine Safety

Fig. "Comparison of sizes from particles in oil" (Lit. 22.3.3.1-1 and Lit. 22.3.3.1-6): For the practitioner an imagination of the particle sizes of oil contaminations may be helpful. This enables primary hints for the evaluation of deposits from filters and magnet plugs about its failure relevance.
The properties of the particles are of high interrest. These enable as characteristic features an identification. With this the precondition for remedies and risk evaluations would exist.
The potential deteriorating effect of a particle does primarily depend on indentations (Fig. "Reduction of bearing lifetime by particles") and abrasive wear. In this connection besides the size, the form (Fig. "Deterioration effect of particles in oil") plays an important role.
In the following, the influence of these features at the deteriorating effekt of foreign particles will be considered.

Particle size: The designations primarily relate at the diameter of a virtual ball which enwraps the particle. Does this surpass the usual thickness of the oil film between sliding surfaces or pitch surfaces (races of anti friction bearings, gear tooth flanks), the precondition for a dangerous mechanical deterioration exists (Fig. "Micropittings and life time"). Brittle particles, which bridge the bearing oil film fragment in a size, corresponding the gap width. With this some hundred smaller particles can arise. They can intrude into the gap and increase the damaging effect.

Particle surface: With the crushing of a particle, the total (summarised) surface increases markedly. So for example, 100 equal fragments have a 4,5 times larger surface than the original particle, from which they have developed.
This is significant for the settle velocity. It rises exponential with the particle size and the particle surface, related to the particle weight (Fig. "Size depending settling of particles in oil"). The settling velocity of thypical particles is markedly slower than the flow velocities in the oil system. With this it is hardly enough, to form deposity without inertia, like from centrifugal forces at flow deflection during operation. However during stand still and in storage tanks there is an other situation.
The size of the surface promotes potentially the damaging effects like:

• Water in the oil (chapter 22.3.3.2) is bonded by a bigger particle surface. So an emulsifying effect develops.
• Catalytic effects in the oil are intensified.
• The effect of additives like for minimising of wear or as corrosion protection is decreased.
• The particles serve as condensation nucleus for air bubbles. With this the proneness of the oil for foaming is favoured, at the same time the separation of air and oil is hindered.
  

The form of the particle is from the view point of the tribology respectively of the wear/abrasion effect of high influence. Understandabliy spherical particle (Fig. "Spherical oil contaminations") behave less abrasive than edged (Fig. "Deterioration effect of particles in oil"). Edged particles form primarily by fragmetation. However if this takes place into the oil in the oil circuit or already before the insertion, like blasting grit/shot (aluminium oxide, glass, Fig. "Ol filter helping for diagnistics") is not clear. Of importance is the adjustment of the fracture facets of hard, brittle particles. So it is easy to see, that from this abrasive processes are concerned (Fig. "Deterioration effect of particles in oil").

The number of particles rises the potential damaging capabilty. Withn this, the risk of damaging indentations and erosive abrasion rises linear. The amount of wear can easy surpass a multitude (a magnitude) of the particle mass. This risk exists in regions of the oil system, which are not protected by a filter to the „contamination source“. At the same time particles can promote by erosion and chemical reactions the formation of new particles. Corrosion inhibitng additives can minimise this effect.

Fig. "Deterioration effect of particles in oil" is concerned with further, besides the the appearance, however informative particle properties. To these belong hardness, density, composition, polarity, magnetic behaviour and electric conductivity.

Fig. "Ol filter helping for diagnistics" (Lit. 22.3.3.1-3): Oil filter (sketch below left) can, contrary to magnet plugs, catch all kinds of contaminations. They merely must be bigger than the mesh/pore size of the filter. Even the possibility exists, to contain smaller particles. For this, filter fibres must have a certain surface activity and/or a fibrous structure (e.g., paper filters). Filters are periodical maintained, respectively the filter elements (cartridges, sketch above) exchanged. For this reason filter residues must be related to this time period.
Particles above 0,1 mm can be investigated with the binocular. An assessment is possible with sufficient expertise and experience. For this, if necessary, picture examples in maintenance manuals and specification must be used.
Is the pore size/mesh of a filter about 0,03 mm, particles in the size range of 0,015-0,025 mm are a first sign for accelerated wear of a component within the oil system. Are ferro magnetic particles concerned, it must be assumed, that sizes above 0,1 mm settle at magnetic plugs. With filters which contain particles of about 0,003 mm, the investigations of the surface, structure and form get more importance. Used are modern microscopes. Optical with the use of computer programs (e.g., for a better depth of field) and/or elektron mikroskopic (SEM, volume 4, Ill. 17.3.2-7), even tiny particles can be evaluated. So important conclusions about the formation mechanism get possible.
Typical macroscopic and microscopic (SEM) assessable particles (detail) are:

„1” Needle like metal chips may be procuced durig sliding wear. Typical are rubbing processes or dry run at soft materials like brass, copper (e.g., bearing cage) or light metals of casings (Al-, Mg-alloys). At steel chips the suspicion of an intense rubbing process with overheating and the drop of hardness exists.

„2“ Metal burrs can be identified at features like sharp edged cross sections and crack like notches. At first the suspicion of unsufficient deburring during the production process exists (volume 4, Ill. 16.2.2.2-1 and Ill. 16.2.2.2-6). At such burrs cross grooves at the surface can be expected. Burrs can also develop during a rubbing. In this case attention must be payed at signs for high temperatures (annealing colour) and lengthwise grooves.

„3” Machining chips/milling chips have their origin in the production (new parts, repair). They get into the oil circuit because of unsufficient cleanness or carelessness. Such a possibility is the accumulation in a badly visible oil bore.The chipa have a lengthwise curled shape with crosswise oriented scales. Smooth chips with longitudinal grooves formed as flow chip.

„4“ Flaked coating particles are normally flat, sometimes with structures of the production process. To these belong features of not reworked galvanic surfaces (e.g., silver) and pores of thermal spray coatings (hard facing of labyrinth tips, rub in coatings). Pores and certain structures must also be expected at fracture surfaces in the coating. Grinded hard coatings like Cr-coatings/platings are especially used as sliding surfaces for seal rings and/or as dimensional correction during repair. The structure of its fracture surfaces are characteristic features for the expert (e.g., column structure). On a surface grinding marks on the adherence side a non machined structure can be expected.

„5“Nonmetallic particles which developed in the aeroengine itself (Fig. "Deterioration effect of particles in oil") show frequently a brittle appearance. Typical are crumbing hard facings at the tips (volume 2, Ill. 7.2.2-3.1) of labyrinth seals from the main bearing chambers. Usually concerned is aluminium oxide (also alumina, Al₂O₃). Also tungsten carbide as wear protection can spall. Particles of thermal barrier coatings (zirconia) of the hot parts can be transported by sealing air into the oil.

„6” Nonmetallic particles from the production (Fig. "Deterioration effect of particles in oil"): Typical are alumina blasting particles of cleaning processes. Also cullets of glass beads from a cleaning process or strain hardening process can be found. Particles which widely kept their original shape (glass bead, crystals of corundum/alumina), may stay during the new parts production or repair in components of the oil system (e.g., gear box/casing). Splints have a charging effect at abrasive blasted surface (volume 1, Ill. 5.3.1-7). Thereby splints of the blasting material become stuck in soft metals (e.g., gear casings made of light metals). These are released during the operation time and are transported by the oil stream.

„7” Beads/shot can have very different origins (Fig. "Spherical oil contaminations"). For example, left peening material from a strain hardening process (peening with glass beads, blasting, shot peening) or slugs (melting drops) from welding processes or from thermal drilling processes (e.g., laser drilling, EDM = electric discharge machining). The surface of the beads which formed from a melt often show a characteristic dendrite structure which can be easy identified in the SEM.
Very small beads can also develop inside a fatigue crack (Fig. "Spherical oil contaminations").

„8“ Residues of sealing componds are usually jelly, coloured (frequently blue) particles. They originate from excessive sealing compound (Fig. "Excessive use of sealant") between the flanges. This paste is oozed out of the sealing gap into the interior of the casing during the tightening of the flange bolts. At flanges of gear casings/boxes, the special danger exists, that rests of the sealing compond get into the oil circuit. If these are markedly amounts, the danger of blocking is especially high for the oil sieves/screens (Fig. "Correct amount of sealant is important").

„9” Coke particles originate from coke accumulations in hot zones of the oil system. They are frequent and can get quite dangerous as cause for a blocking/clogging of small cross sections (e.g., oil jets/nozzles, Fig. "Formation conditions shown by coke deposits" and Fig. "Problems of bearing chambers near hot parts").

Fig. "Deterioration effect of particles in oil" (Lit. 22.3.3.1-6): The influence of the shape of brittle fracturing crystalline particles shows the diagramm above. It can be seen, that the effect of wear/abrasion depends from the so called edge frequency, i.e. the number of facets. Astonishingly obviously not always the hardness is essential. For example the wear effect of a diamond, in spite of its extreme hardness is, markedly lower than the much „softer“ minerals Alumina (corundum) and quartz crystal (silica).
Besides the particle size and shape (Ill. 22.3.3.1-1) the deteriorating effect is also influenced from other properties of foreign particles in the oil.

Particle hardness influences the wear effect and deterioration by indentations in functional surfaces (Fig. "Reduction of bearing lifetime by particles") of the components in the oil system. In the chart the hardness is specified in Mohs. It is determined with the scratching ability of the material with the lowest value, by the material with the higher value. The compression strength is essential, if brittle materials like dust, particles from cleaning blasting, thermal spray coatings like hard facings and thermal barrier coatings are concerned.
Also hard metals like steels of anti friction bearings can be assigned ductile materials. This applies the aspect of a deformation between two thrust faces. These particles are only deformed without fracturing and such can produce manifold consecutively damaging indentations.
It must be noted, that metallic materials strain harden durig cold deformation. This can raise its hardness markedly. So the hardness of seemingly soft ductile metals increases by cold strain hardening dangerously. From this point of view, the indications of hardness charts can confuse.
To leave in running surfaces/pitch surfaces/bearing races dangerous indentations, a high hardness of the particles is not absolutely necessary. Is the plastic deformation during indentation hampered, incompressibility can get effective.

The density of particles respectivels the specific weight is of importance for the floating/suspension in the oil (Fig. "Size depending settling of particles in oil"). With this, the tendency for settling is influenced. For example it needs minutes till a babbitt (white metal, sliding layer from lead and tin) of about 0,02 mm size settles in turbine oil. So the particle weight improves under centrifugal forces the cleanig of the oil. This is especially important, because else more heavy particles increase the wear in the oil system. The cause is, that the particles have an intense contact with the surface of the part during deflection (oil pipe line) or impinging oil jets.

Composition of the particles: The deterioration potential of a particle is influenced by the chemical reaktivity. Ceramic particles like dust, glass and blasting grit already don't react chemical, they are inert. This applies as long as they don't bear contaminations like salt deposits. However fresh metallic sufaces by erosion behave different, this applies for components and abrasion. These fresh metallic microsurfaces are not protected by oxide layers. They are highly reactive. Iron and copper promote the aging (oxidation) of the oil. Acids develop deposit layers and sludge (Fig. "Formation of depositions in hot oil systems" and Fig. "Formation mechanisms and oil coke features").

Polarity is caused by the electric charge of molecules, respectively ions (charged molecules). These can deplete oil around polar additives. They serve the protection against corrosion (rust inhibitors) and wear (antiwear agents). Its cleaning effect prevents an agglomeration of contaminants. Other concerned additives shall increase the compression strength of the oil.
So it can be avoided, that particles which tend to agglomerate form deposits. These could clog narrow cross sections of the oil system. Thereby water in the oil can act as a binder. This supports, with the depletion of additives, the formation of emulsions and sludge.

Magnetic properties of particles are utilised to separate them with magnet plugs from the oil (chapter 22.3.4). Removed magnetised particles are transported by the oil stream. With this they tend to adhere and to form agglomerations on magnetic surfaces. So the danger exists, that fine openings will close or sliding surfaces wear. This is especially alarming at control units and valves in hadraulic facilities. To these belong actuators for the thrust reverser or variable compressor guide vanes. Often these adjusting movements are carried out by electromagnets (solenoids). Its magnetic field can attract and accumulate magnetic or magnetised particles from the oil.
It must be noted, that also high alloy austenitic steels which normally are non-magnetic can get magnetic, caused by a heavy plastic deformation.

Electric conductivity gained impotance with the higher pureness of the oil, concerning the formation and effect of of contaminations. In the oil flow, the innternal friction can produce static electricity. Does this trigger small flashs, those high temperatures cause coking. This effect is prevented by electrically conducting particles.

Fig. "What deposits of magnet plugs can tell" (Lit. 22.3.3.1-4 and Lit. 22.3.3.1-5): Magnet plugs (magnet chip plug, sketch above) and magnetic particles sensors (magnet chip detector, chapter22.3.4) collect ferromagnetic particles which are bypassed by the oil stream. Thereby the detector triggers a warning signal in the cockpit during the bridging of the poles. Because of the necessary magnetism particles of steels of different origins are concerned. Typical features (Fig. "Ol filter helping for diagnistics") enable the expert conclusions. Besides the potential deteriorating effect at oil wetted components, special attention apply particles which point at developing or already dangerously developed failures (frame below).
Concerned are primarily fatigue failures at pitch surfaces. The fatigue caused particles are typical for the races of anti friction bearings (Fig. "Fatigue pittings at bearings") and tooth flanks in gears and gear pumps (Ill. 23.2.1-5). Its characteristics develop during the special failure mechanism of cyclic (vibration) fatigue (chapter 23.1 and chapter 23.2).

Fig. "Spherical oil contaminations" (Lit.22.3.3.1-2): In filter residues and magnet plug deposits always beads/tiny balls or parts of it can be found. These can differ characteristic in type, composition, surface structure and size. Most suitable is a microscopic investigation with the scanning electrom microscope (SEM).

Typical causes for beads/tiny balls in oil:
Fatigue processes in metals: Inside a (cyclic) fatigue crack, tiny abrasion particles and fracture particles can be formed to balls (< 0,005mm diameter). Acting is the plastic deformation between the moving crack faces and/or abrasion/wear. They are transported out of the crack with the sucked and squeezed oil, or escape into the oil during the fracture of the crack (fatigue pits/pittings). The expert sees it as an additional hint at a beginning or advanced fatigue failure. Such tiny balls are found in the oil, already after 60 % of the bearing life. After 80 % of the life the tiny balls are frequent found in a so called ferrogram. However, if there are no tiny balls a beginning or present fatigue failure can not be ruled out.
Rub processes and sliding processes can produce tiny balls (< 0,005mm diameter) already shortly after the run in phase. It is supposed, that they form from metallic abrasion in the roughness cavities. Aso tiny balls seem to develop during wear processes with oscillating sliding movements (fretting).
Obviously rubbing processes during high temperatures form primarily by fusing tiny balls in the region < 0,005mm diameter.
In this connection shall be mentioned the formation of also very small spherical oxides during the corrosion of steels (Lit. 22.3.3.1-7). Larger but also tiny balls (> 0,01 mm diameter) are especially traced back to skidding pitch surfaces (Fig. "Deterioration by cage slipping so called skidding"). They develop during galling and scoring. With this also could be explained, why the surface of the tiny balls often show signs of overheating like melting structures.
Sliding surfaces (e.g., in gear pumps) made from porous sintering materials „bleed” during running hot tiny balls. Concerned are infiltrated low melting metals (lead). However these tiny balls are with a diameter up to 1 mm markedly larger than those, mentioned before (volume 1, Ill. 5.1.5-2). Electric sparks can produce small balls. This is as well possible in the oil circuit by electric continuity (lightening?) as also during processes in production or repair. Already micro sparks can melt/fuse agglomerations of metallic abrasion into tiny balls. Also out of the round or edged particles can get ball shape during heating by sintering, already below (!) the melting point. Here the high surface energy of the abrasion supports the forming process.
In production processes, which proceed with spark formation it must always reckoned with the formation of tiny metallic balls. Sparks can occur desired or undesired. For example during thermal spraying (volume 4, Ill. 16.2.1.8.2-4), welding (volume 4, Ill. 16.2.2.6-5) or laserdrilling. Loose adhering, they get into the oil system because of carelessness.
Galvanic deposits (electrodeposition) like chrome plating and nickel plating can develop ball shaped warts (volume 4, Ill. 16.2.1.8.3-5). These processes are applied for repair of worn slide faces of sliding ring seals or hydraulic pistons. Is the deburring not enough careful (volume 4, Ill. 16.2.2.2-6), ramained warts can get loose during operation.
Nonmetallic balls in the diameter range up to some hundredth millimeters, develop in the oil from additives (white tiny balls, so called „oil balls“). Astonishingly high hardness of 48 Rc was measured, similar a tool steel. With this a high damage potential exists.