19:192:1921:1921

19.2.1 Problems with sensors/probes, lines, pipes and connectors.

Problems with sensors and monitoring units in aeroengines (Ill. 19.2-1.1) can affect safety respectively flight safety very different. A breakdown influences directly the function of signal processing and signal depending components like c ontrol units. Those again lead to an undesirable and unexpected operation behaviour of the aeroengine.

Even without direct influence at the aeroengine, for example during a defective contact of the signal transfer, the behaviour of the pilot can be essential influenced. If it is known, that a sensor functiones not reliable or frequently shows errors respectively fails, in a real potrntial hazard the display can misleadingly interpreted as insignificant.

In many cases indication errors of sondes can be identified on-site by the pilot if there is sufficent knowledge of the system or later during a maintenance by plausibility considerations (Ill. 19.2.1-5). For this displays and measurements of other instruments and components can be used (Ill. 25.1-6 and Ill. 25.1-7).

With the „health monitoring“ (chapter 25.1), the life time observation of aeroengine components, parts and electronic control units, which can consider a multitude of operating parameters (Ill. 25.1-10), increases the number of serially required measurement devices, sondes and sensors. So obviously fighter aircrafts of the next generation will be equipped with hundreds of sensors/probes. These are:

  • Temperature measurement instruments like thermal couples (Ill. 19.2.1-7) and pyrometer (Ill.19.2.1-6).
  • All sorts of pressure probes.
  • Measuring instruments for the flow rate of gases and liquids.
  • Flow measuring device/flow meter.
  • Probes for liquid levels.
  • Sensors which register the failure of parts/components (e.g., fracture of shafts with overspeed).
  • Distance measurement (e.g., gaps).
  • Sonds for interlocking devices (e.g., thrust reverser).
  • Load/overload measurement (e.g., at thafts),
  • Acceleration probes from vibration pick ups.
  • Smoke detector.
  • Fire detector/alarm (Ill. 19.2.1-10), ignition detector (combustion chamber, afterburner).
  • Impulse meter/speed sensor for shafts.
  • Measurement devices for electrical parameters (chapter 25.1)
  • Measurement devices for electronics. (chapter 25.1)
  • Magnetic chip detectors (chapter 22.3.3).
  • Particles in lubrication oil.(chapter 25.2.1).
  • Particle identification in exhaus gas (capacitive).

The transfer is carried out electrical, mechanical or optical by conductor respectively cable with appropriate connector systems/plugs. With the number of sondes/probes also increases the likelihood of a failure. This is tried to identify and balancing with logic circuits (Ill. 25.1-7).

The maintenance can in different ways, passive or active, influence sensors/probes and its transmission systems. First of all the maintenance has to care that deficits are discovered and eliminated. To this also belong symptoms as wear, contamination and corrosion. In this connection the cleaning of pyrometer lenses is to mention (Ill. 19.2.1-6)

But the manitenance itself can trigger problems with extensive consequences. As passiv the deferring of cleaning tasks or the inspection of the systems must be mentioned. As an active example can serve a maintenance caused installation error in the area of magntic chip detectors (Ill. 19.2-2.2 and Ill. 19.2.1-9) or the damaging by mistake wetting of susceptible components (Ill. 19.2.1-2). The chafing of cables with the danger of a disrupt signal transmission, short circuit and fire ignition by sparks frequently is connected with maintenance.

Illustration 19.2.2-1.1 (Lit. 19.2.1-15): Sensors/probes as well as electric cables and connectors are direct or indirect responsible for a high percentage of problems (Ill. 19.2.1-1). Not seldom they are in connection with the maintenance. In the following the problems with the main components as there are sensors, connectors and cables are discussed. Naturally plenty potential damaging operation influences affect the components.

Some examples for clarification (Lit. 19.2.1-3):

(„A”) Adaptors, connectors and contacts (Ill. 19.2.1-2):
Cables for the electrical power supply and the data transmission can be very different. Equal also the sensivity for operation influences. Generally a connector should be checked before the use for problems like damaged contacts, corrosion, contamination and insufficiend safety.

  • Conectors attatched at the outside of the engine can be intensified exposed to corrosion by condensation, especially in marine environment. As a result the oxidation of the pins acts electrically isolating (Lit. 19.2.1-3).
  • Contaminations with media like oil, hydraulic fluid, cleaning agent (Ill. 19.2.1-2).
  • High vibration load can cause cable fractures and/or wear when there is a contact to other surfaces (chapter 23.5).
  • Damage during inappropriate handling, e.g., forced plugging.
  • Wear of the contacts by vibrations/micro movements (fretting, Lit. 19.2.1-3) and frequent use.

(„K“) Cables / lines are conformed to the designated use. Therefore the type can be very different. Normally the signal transferring or energy transferring element is enclosed by a protecting and stabilizing jacket. This can consist of metallic tubes or fabric. Electric and optic conductors use rather plastic/elastomere for the jacket. This is according to the damaging operation influences.

Mechanical actuation and feedback cables as they are used at elder engine types:

  • Wear in guidances and force transferring spiral coils (system bowden).
  • Fracture as result of force or vibration fatigue.
  • Blockage/jamming by insufficiens lubrication, corrosion, freezing and contamination. Electrical cables to transfer power and/or impulses/measurement data:
  • Short circuit, annoyances because of fretting on the insulation at contacts with other components.
  • Cable fracture by vibration fatigue or electrical overload (e.g., short circuit).
  • Failure of the shielding against electric or magnetic fields.

Optical cables (fiber optic, light cables):

  • Fracture due to mechanical overload. - Fracture because of humidity triggered stress corrosion (stress corrosion cracking = SCC, Ill. 19.2.1-6). Particular dangerous is a bending or folding at the entrance into the connector.
  • Problems with light guiding contact surfaces.

(„S”) Sensors, probes, measurement devices:
Damaging operation conditions are mostly `sensor spezific', that means its effect must be seen in connection with type and principle of the sensor.

Optical sensors: To those belong pyrometers (Ill. 19.2.1-6), smoke and flame detectors in fire detector systems (Ill. 19.2.1-10; volume 2, chapter 9.5) and for the monitoring of combustion chambers. Common problem is the unallowable impact of contamination or matting due to erosion of lenses and windows that influence the entering light (Ill. 25.2.1-10). Further a damage or change of the light sensitive photo cell is possible.

Electric and magnetic sensors have as common problem, if existing, the failure of coils by mechanical overload (vibrations, thermal expansion) and temperature respectively damaging/aging of the isolation.

Sensors for measuring the flow and its rate: Here especially problematic is the blocking of measuring openings. It can be also about contaminations (e.g., insects) or icing (volume 1, Ill. 5.1.4-3).

Pressure sensor: To those belong especially those to measure the ambient pressure at the engine inlet. They can be blocked by ice, foreigen objects or mechanically damaged.

Systems with electrical coils: To those belong vibration pick-ups and tachometers for shafts. Those probes can, according to the place of installation, encounter extreme vibrations and temperatures. Thereby it must be reckoned with damages of the insulation (e.g., aging/deterioration of resins). Results are short circuits and mechanical failures. The winding wires can oxidize or corrode. That weakens the cross section up to a failure. So it comes to a lingering change or blackout of the signal.

Vibration wear (Fretting) at wires rubbing against each other can lead to decrease of the cross section and fracture.

(„B“) Attatchements of lines serve different purposes. They protect against the contact with other surfaces and wear damages. A further task is the avoidance of vibrations and static mechanic overload, as it occurs with thermal expansion. Used are suited lined clamps or brackets. It's important that the designated type is attached to the right position (Ill. 23.5.2-2). Does this not happen, there exists the danger of a damage of the line by operation loads and/or the attatchment itself. Loosened attatchments, or leaved by mistake, can also produce fretting and/or act as foreign object on other locations.

Illustration 19.2.1-1.2 (Lit. 19.2.1-21 bis -23): Connectors/plugs are relative frequent cause for a fault or hampered transmission of sensor signals (Ill. 19.2.1-11). In many cases this is concerned with leakages by which contaminations can reach contacts/pins in the inside of connectors (Ill. 19.2.1-2). Typisch is the entrance of humidity during stand still (condensation water, humid air) or durig the flight (Ill. 19.2.1-4). This leakages can be based on a combination of several mentioned influences. The disturbance of the signal tansfer not only develops effects like corrosion (Ill. 19.2.1-12) by reactions acting as insulation. Also can be simultaneously acting fretting (wear, Ill. 19.2.1-12) markedly aggravated by vibrations. Electrical conducting contaminations like seawater (Ill. 19.2.1-11 and Ill. 19.2.1-12) can trigger shorts or at least creepage currents between then contact parts.
Are connectors not sufficient protected against electromagnetic stray radiation, there are false alarms possible in the region fo strong radio transmitters (Ill. 19.2.1-3).
In the end also problems are to mention, caused by maintenance. To those belong not sufficient tensioned/torqued electrical connections (Bild 19.2.1-11).

Illustration 19.2.1-2 (Lit. 19.2.1-1): The failure warning of the fire warning system was in this case traced back to an electric conducting contamination (electric short) of the contacts of a connector. Obviously there was a mixture of aeroengine lubrication oil and hydraulic fluid. Not clear remained, how the contaminations got into the hermetic tight plug-connection. The faulty function could not repeated in the shop of the operator on the disassembled plug.
To a similar failure case refers a canadian „service difficulty advisory” (Lit. 19.2.1-3). This advisory was triggered by a multitude proved incidents during operation. They were traced back to contaminated and corroded elrectric connector plugs. There was reported amongst others of a wrong torque measurement at turboshaft aeroengines, fauly digital control units (full authority digital engine control = FADEC), malfunctions of bleed valves, problems with electtric fuel control units (engine electrical control = EEC) and fuel pumps. Cut off signals can deceive and are difficulf to determine, what leads to long repair work while the airplanes can not be used.

Illustration 19.2.1-3 (Lit. 19.2.1-2): Near the mast of a radio station with high power (100 kW) it came to the crash of a fighter aircraft. Because of unsufficient „electromagnetic tolerance (EMV) the control flaps of the airplane were uncontrolled actuated. After that the screenings of the computerised devices, especially the installation of new cable harnesses correspondent to the militars standard 1760 were carried out. After this measure no more a comparable case occurred in the following decades.

Note: This elder fighter aircraft type has a weight at the limit of its performance. Even when a relatively small drop of the aeroengine power occurres (e.g., opening of the variable thrust nozzle), the airplane is no more safe to handle and the danger of an immediate crash exists.

Example 19.2.1-1.1 (Lit. 19.2.1-4):
Citation (translated): Failure of the aeroengine because of a cable fracture at the power lever and crashed. Both pilots… escaped.

Example 19.2.1-1.2 (Lit. 19.2.1-4):
Citation (translated): During an aerodrome cycling, fracture of the power lever and therefore no more control of the aeroengine. Both pilots deboarded.
Comment: In all cases it is a single engined fighter airplane of an elder type (Ill. 19.2.1-4)

Example 19.2.1-2.1 (Lit. 19.2.1-4):
Citation (translated):Landing in snow flurry … ovbiously with open thrust nozzle, canceled and touch-and go… pilot subsequent because of safety teasons deboarded.

Example 19.2.1-2.2 (Lit. 19.2.1-4):
Citation (translated): …during approach… with open thrust nozzle in heavy snowfall 1,5 km before the border of the port crashed on free field …both pilots dead.

Example19.2.1-2.3 (Lit. 19.2.1-4):
Citation (translated): …during approach… in a severe storm crashed on free field with open thrust nozzle. Pilot deboarded.

Illustration 19.2.1-4 (Lit. 19.2.1-4): The examples 19.2.1-2.1, -2.2, -2.3 are valid for the in the picture shown one-engine fighter aircraft type. In several cases during flying through a rainfall area the variable thrust nozzle opened and lead so to an air accident. Thereupon extensive aeroengine tests in simulated rain were carried out at a test bed. At the beginning without success to simulate the failure of the thrust nozzle. Not until the casing of the electric amplifier for the thrust nozzle control was splashed from the front with a water hose it came to the expected opening of the thrust nozzle. A closely investigation showed, that the water gets to the inside electric through a narrow ringshaped gap between the, from the outside tunable trim button (detail) and the casing. This lead to a temporary malfunction of the system. The potential leak was tightened by adequate means.

Illustration 19.2.1-5 (Lit. 19.2.1-5): Concurrently causative for the crash was the icing and with this the blocking of the probe for the measuring if the intake air pressure. The probe is located in the nose cone (sketch below). It was missed, to deice the airplane shortly before the start, in spite of heavy snowfall. So the wings and the fuselage iced, what at least would have demanded an especially high take off power. The measurement value of the pressure probe blocked by icing, is of expecial meaning for the adjustment of the aeroengine power. The situation was aggravated, because the deicing of the aeroengine was also not activated, but the probe has a bore at the nose cone. Lacks the pressure of the deicing air, the display of the aeroengine power is markedly higher than actually. This was not noticed by the pilot, although the plausibility with other disyplay values like rotation speeds, speaks in contrast against the high displayed power.
So the power of the aeroengine was no more sufficient for the icy airplane with its deteriorated flight qualities. It came to the crash.

Illustration 19.2.1-6 (Lit 3.6-5): Against thermocouples, pyrometers have a big advantage. They can measure contact-free the temperature of rotating components like turbine blades. Thereby it becomes also possible to detect over temperatures that rely not on an increase of gas temperatures. This is for example the case when it comes to the blocking of cooling air channels in hot parts. But pyrometers have also weak points that can mean increased maintenance effort. To these belong:

(„1”) Contamination of the front lens (lens fouling) pretends the control unit a lower temperature niveau. That can have a strong influence on the life time of the hot parts. 15°C increase of the material temperature lead at the normal hot parts operation temperatures to bisection of the lifetime (Ill. 2.3-2). Besides the optical transparency of the lens the calibration of the pyrometer is changed. This shortens the required maintenance interval. Therefore a clean front lens of the pyrometer is an important maintenance task. So the maintenance intervals may not excessed in any case.
The reasons of the lens contamination are particles in the gas stream which originate from the combustion chamber and end up in the sight tube. To minimize this effect pyrometers are pressurized with cleaning air (purge air) from the high pressure compressor. It acts as sealing air against the hot gas and escapes from the sight tube into the gas stream. But this air can favor the contamination of the lens, just in contrast to the intended effect. This is the case when particles with sufficient high kinetic energy (velocity, size) breakthrough the flow of the purge air and hit the lense by the swirl before.

(„2“) Fractures of glass fibers in the light cable (frame below). If the light is transferred from the photo cell by a glass fiber bundle there is the danger of stress corrosion cracking (SCC) in the fibers. It was observed that apparently during the time in the stand still phases condensation can accumulate in the region of the fiber bundle. Are the glass fibers subject to a certain tensile stress niveau, they can break in humidity with a delayed crack growth (volume 3, Ill. 14-12). For example dangerous stresses can develop at the fitting of the glass fiber bundle behind the lense and/or before the photo cell. Also a too narrow bending radius of the glass fiber bundle, if it is exposed over a longer time to condensation and/or humid air, can trigger the cracking of fibers. That causes a slow drift of the measurement data to seemingly lower temperatures. Thereby the cosly exchange of the pyrometer system will be inevitable.

(„3”) The problem of the change of the emission behavior of the monitored area on the component is not attributed to the pyrometer itself. However oxidation, contamination, erosion or interactions with foreign objects can change the radiation spectrum and the drift off of the data.
Also a falsification by glowing soot particles can not be ruled out.

(„4“) Haze of the lens by erosion was not referred until now, but is a probable damage. It attains actuality with the use of hard particles at the bladetips of turbines and compressors and ceramic thermal barrier coatings in combustion chamber and turbine. Those particles can have quite enough kinetic energy to reach the lens against the cleaning airstream at the lens. Even a very hard sapphire (alumina) lens could be affected by the erosion effect.

Illustration 19.2.1-7 (Lit 3.6-5): Thermocouples (thermal elements) are used in gas turbines not only to measure hot gas temperatures. Further examples of use are

  • oil and fuel temperatures,
  • temperature at the engine intake,
  • temperature in compressors.

There are several constructions to guarantee the reliability over long operation times. Generally in the hot parts region the metal combination nickel („Alumel”) / chromium nickel („Chromel“) is used. Its application temperature lies in the industry at maximal 1200°C.
Thermocouples using PT/PtRh have the disadvantage that a catalytic effect with rest hydrocarbons in the hot gas, rises the temperature of the surface. That creates false measurement data.

Thermocouples can be damaged under operation influences in several ways:
A damaged thermocouple lets always expect a drop of the voltage. This can mean that a too low temperature will be shown. For elder engines the temperature may be raised by the control unit or by the operator and so overheating damage can occur over a long operation time. In the most cases the hot junction has visible changed. 15°C increased component temperature can bisect (see page 2.2-9) the creep life!

Empirical the most frequent problems occur at the junction (joint) of the sensor and at the adjacent links to the line wires.

Already during the visual check, if necessary under the binocular, damages can be seen. Typical is cracking with discolorations (e.g., green) and pustule formation (tumor). In some cases the metal can be significant eroded. This is normally to be seen in connection with corrosion/oxidation. Then a repair is no more possible.

A failure mechanism is the diffusion of contaminants from the hot gas and/or a reaction (e.g., hot gas corrosion/sulfidation) with the wires of the element. For example an intensified oxidation can prefer special alloy elemments and so chance the composition of the wires. Also a diffusion of contaminants from insulating ceramic (Fe, Si) into the wires and the region of the junction between the couple materials is possible.

Fracture of the wires can be based on the decrease of the cross section (hot gas corrosion/ sulfidation Ill. 3.4-2, oxidation, erosion), overload by foreign object damage and/or embrittlement. Diffuse embritteling elements like aluminium (rub in coatings in the compressor) or silicon (dust), cracks can occur. Also a foreign object or own object like coke from the combustion chamber (carbon impact, see also Ill. 3.3-12) can trigger a fracture.

Deteriorated insulator: In elder engine types thin metallic bridges were observed in the insulator (Mg-oxid?) between the wires of the thermocouple. They showed deviations which affected the temperature measurement unacceptable. Possibly they formed during a faulty, much too high calibration or `healing' temperature in the overhaul. Unfortunately this phenomenon was never satisfactory settled.

During long stand still (weeks) the insulation can reversible absorb condensation / air humidity. Than, by a leak current the voltage declines and the gas temperature seems to drop.
That can result in over heating during the start phase. In such a case normally the OEM prescribes a `healing' of the thermocouples before a start. This can take place in a suitable oven under air at operation temperature.

Short-circuit in the thermocouple or between its wires. At the fracture also a thermocouple develops that produces a signal corresponding to the here acting temperature. Causes for short-circuits in the element itself are

  • a broken ceramic insulator,
  • metal in the shield tube and
  • a cracked bonding.

In the wires itself it is to be about an unprotected, jammed area.
The insulation can considerably degrade if the porous insulator is soaked with humidity.
Damage of the protective pipe: Enables leakage the entrance of hot gas or molten metal (e.g., abrasion of a severe rub in) to the inner of the element, it is to reckon with pustle formation and unnormal discoloration.

Illustration 19.2.1-8 (Lit. 19.2.1-11): This is an impressive example for the operation caused for the susceptibility ot the function of sensors. In the case on hand obviously several influences play together:

  • Design with brazings as bond and seal.
  • Corrosion by atmosphere, especially in sea atmosphere.
  • „Creeping” change of the sensor.

Obviously since longer time problems with the fuel control pressure bellows appeared. This can be suggested from a bulletin of the aeroengine OEM that mentiones the change from the Cu-Be-sheet of the bellows to stainless sttel sheet. It is prescribed that the exchange must be executed at the next overhaul/repair, when the area of the pressure bellow is disassembled. However at the latest during the next overhaul of the fuel control unit. The advantage of the new bellows should be that they are not joint from three parts and have a better corrosion resistance. The elder bellows from Cu-Be material could be repaired because of their three part design. Obviously this was necessary because of their leakages but represented a increased risk. This is understandable if we know how problematic the proof of a small leak on such components is (lower sketch).

Ill. 19.2.1-9 (Lit. 19.2.1-13): Since long time magnetic chip detectors are used for the monitoring of the lubrication oil and detecting failures as early as possible (bearings, gears, chapter 22.3.3). Mostly are several of such probes positioned at suitable locations of the oil system. The check of the collecting probes (sketch above left) takes place in by the OEM in the maintenance manuals specified aeroengine specific intervals. For this the probes are unscrewed and removed. Then it is possible to assess amount and type of the build-up accordant to specifiied requirements.

At other probe versions a settled chip leads to a resistance change respectively bridgeover of a electric circuit (e.g., „two pole, two wire“ type). This activates a warning light in the cockpit. So this system will be a sensor with continuous function.

Unfortunately the experience shows, that this monitoring system not always possesses the necessary reliability. In some cases there were so called „dormant failures” which were caused by the production and lead to „periodic“ outages. Obviously this were failures in the associated electronics which showed not until during operation. Such failures may also in furure not be confined to an exception. Rather it can be supposed, that such a danger potential exists at other sensor systems. A tendency to complexity of the analysis systems may be rather benificial for this.
In one case the failure of the chip display lead to secondary damages and the fracture of a rotor component with penetration of the airplane nacelle.

Corrosion, dwindling magnetism and disconnected electric circuits are typical operational system failures.

To improve the reliability of continued chip monitoring there are the following recommendations:

  • Check of the function in suitable in periodic intervals and at every assembly process at components in the oil system.
  • Check of the sensors for creeping changes of the functionality. The check must take place in a specified manner. For example with an electric conductive bridging of the magnetic poles and inspection of the cabeling. To this belong special plug connections.
  • Vibration of the pick ups can lead to fatigue fractures and wear with a deteriorated signal transfer at the plug connectors of the connection lines.

Illustration 19.2.1-10 (Lit 19.2.1-12): The fire warning in the area of the aeroengine is of highest importance. Malfunctions can have serious consequences. Either a real case of emergency is not indicated or it will be wrong interpreted because of (possible) malfunctions in the system.

There is a multitude of thinkable sensor principles (e.g., optic, electric, pneumatic and combined) of which however only two became accepted in the application.

The electrical respectively thermistor prinziple (detail bottom left) is based on the temperature dependent resistance of metallic conductors. Used is an extruded, wire shaped brittle semiconductor, which possesses an especially strong effect. Problematc is, that the resistance does not change suddenly and therefore no doubtless detectable impulse occurrs. With that the system has problems zu indicate early creeping changes of the resistance, e.g., by corrosion of the connector.

At high vibration load the semiconductor can crumb, escape and cause electric shorts with failure warnings („4”).

The pneumatic principle (detail bottom right) uses as a sensor a thin metal tube filled with helium and titanium hydride. Heating decomposes the hydride and releases larger quantities of hydrogen. With this is linked a rise of the gas pressure inside the small tube which triggers the warning signal. This principle has the advantage,that a clear impuls type signal occurrs. A problem arises, when by crack formation under vibrations titanium hydride trickles out of the sensor jacket and can get into the electric contacts of the switches („5“).

The cable like sensors can react to over temperatures over the whole length. The sensors must meet as well the airworthiness requirements as also the characteristics of the particular application. The experience showed several weak points of the warning devices. Certain failures occur at all systems.

Disruption of the cabling or deterioration of the insulation caused by temperature can lead to crack formation („2”) and electric shorts. Wires can crack by vibration fatigue. Contacts can corrode and change sporadic the circuit continuity. Also the sensors itself can age and affect the electric resistance (impedance) of the system.

The tin sensors are supported by a backing tube (sketch above) and damped to avoid a vibration overload. A further advantage of the backing tube principle is an easier exchange of the whole sensor unit. In the cases of resonance, which can be causative connected with the fitting of the sensors (fixing) also fatigue fractures of the backing tube itself can occur („3“). The handling of the plug connections to the pin wires seems not always „fool proof” („1“).

In the course of time the described weak points of the systems are largely, but not fully eliminated. To minimise this, the remaining risk sensor signals are monitored. Those data are if necessary readout and evaluated during maintenance.

It is a special problem to find targeted failures respectively „poor” components in sensor systems. Contrary to pneumatic systems with a single warning impulse, the identification of component failures in electrical systems is extremely difficult. An advantage have systems which combine sensors of both principles.

To raise further the safety of the display by redundancy during resonance, several systems, each with several senors will be used parallel. For example every aeroengine in a typical modern application has four sensor systems, each with two sensors (pneumatic). In contrast the APU area is protected by three sensor systems, each with one sensor only.

Illustration 19.2.1-11 (Lit 19.2.1-9 and Lit. 19.2.1-14): In this case a loose connection at a plug contact (pin) for the output of the torque signal was found. This lead to a temporary disconnected signal transfer. The signal output device was changed and the airplane cleared for operation.

At two of the previous cases the connector was contaminated with water. During a time interval of about 8 years 19 cases registrated at which problems ar the cabling and the connectors lead to an insifficient dignal were global. The most cases pccurred during take off or in the first climb phase. The OEM traced those failures back to ab intermittent signals. After this actions followed which can give an impression of the problems for the remedy of this seemingly simple problem:

  • The OEM recommended to the concerned operators an inspection of cabling and connectors for the entire fleet and an improvement for the tightening of the connector.
  • 8 years before the current incident the OEM demanded in a service bulletin (SB) the attaching of shrink hoses at the cable attachments to keep humidity off and to avoid a loosening.
  • 6 years before a SB dealed with adjustments at the torque sensorsystem.
  • 5 years before a SB outlined fretting wear at the connector pins. Then followed the introduction of a cannection cable with plug connectors.
  • 4 years before the OEM informed about two new cases which were traced back to an unsufficient tightening of the connectors. The connectors must be checked for wrong screwings. When loose they must be controlled for inner contaminations and humidity.
  • 1 year before again an information of the OEM about recent occurred incidents and the hint at recent in the flight manual published procedures in connection with a malfunction of the propellers during the start. Further it was pointed at a possible situation at which the danger exists, that also the intact engine will be shut off.

Illustration 19.2.1-12 (Lit 19.2.1-16): At least six crashes of this fighter airplane were allocated to a corrosion problem at the plug connections. Obviously the shut off valve between the main fuel pump on the side of the fuselage and the aeroengine could shut off unintended. Obviously this could happen also during flight. The valve can be shut off by a switch in the cockpit when the following screen must be dismounted and cleaned.

The cause for the malfunction of the shut-off valve was corrosion at the tin plated pins which at the opposite side were plugged into gold sleeves. Apparently this was a matter of „fretting corrosion“. This damage develops by micro movements during the vibration of the plug connector in combination with the influence of corrosion. Typical is the access of air humidity, especial to sea atmosphere. The corrosion may have intensified the formation of a corrosion cell between the gold bushes and the tin coating at the connectors. A dangrerous deterioration was so small, that it could be hadly identified with the naked eye.

The electrical conductive corrosion products (abrasion) seemingly lead to a current transfer to the other contacts. This triggered the shutting off of the valve.

As favourable fast and effective interim solution, proved a corrosion protection spray, according MIL-L-87177A Grade B. With this treatment at least the corrosion could be prevented.

Generally the corrosion load in the military use must be estimated as high. This is due to relatively long shut down periods, compared with airliners. Thereby condensate has enough time to form and to act. Additionally missions in low hight above ground, expose the aeroengines intensified to sea air or dust. This is especially true for the mission at aircraft carriers and in helicopters.

Illustration 19.2.1-13 (Lit 19.2.1-15): The total cable length of the aeroengines is especally compared with the fuselage considerably shorter. Modern airliners have cable lengths up to 150 km, Ffighter airplanes up to about 40 km. However the load by vibrations, temperature, atmospheric air pressure differences and humidity is in military use comparatively high.

Already during the assembly failures can occur. Those will be found first at a function check.
Especially dangerous are (electric) insulation failures which were not found during the function test and then show not until the beginning of the operation.

The testing, locating and correcting of cable failures at the complete aeroengine is more demanding as before the assembly. Anyway the chance to identify all insulation problems is limited if surrounding/environment conditions (e.g., temperature or humidity) play a role. From experience the testing of the aeroengine cabling (harness) is relatively unproblematic and is carried out by overhaul shops. Concerned are mostly simple resistance tests with an ohmmeter and/or a control lamp. Those manual tests have understandably in spite of a remarkable expenditure of time only a limited information value.

Therefore automatic cabling analysis systems are more and more used. They are very versatile usable that means as well military as civil. The software can be adapted to the requirements.
Advantages and disadvantages of manual and automatic tests of cabelings:

A disadvantage of manual testing is, that during plug in and plug out of the connection a contact is established which in not guaranteed during operation. Such failures are especially with control lamps not to detect. In spite of bad contact clamping or cold soldering junction the electric connection can be indicated.

The automatic test uses certificated test programs and connection cables/plug connectors. This minimises the likelihood of failures. The evaluation of the test result undertakes the software comparing it with specified limit values. So subjective evaluations are omitted. With the programming of the test parameters also possible overloads during operation can be simulated and electric safety margins can be verified.

Especially complex prove to be the testing of insulations. Here all concerned lines must be checked against each other, when necessary against a shielding and against the aeroengine. Manually this multitude of combinations contrary to an automatic testing is hardly feasible. Noticeable cables and connectors can be identified with a so called scan program in a split of a minute.

Empirically during a manual check it quite happens that plug connectors of the test device don't match sufficient to the application. Thereby during inserting, the contacts/pins can be untraceable damaged. However a so arising high resistance can have a dangerous effect on the operation. For automatic test devices the use of the correct connection cable rather is ensured.

The manual documentation of all test results and events is always problematic. This can be improved by preprogrammed report forms which are completed by the system. In those for example special test instructions of he operator can also be considered.

With an automatic test system the possibility exists to control the function of acessories and single components.Thereby besides the electric resistance, capacity and tension can be considered. If the electric power supply is sufficient the function of components like valves, solenoids and ignition can be checked.

A cabeling should be controlled at the incoming aeroengine before the disassembly. So the highest chance exists to detect latent operation caused malfunctions. After a repair or modification and the assembly before a test/test run again a test should take place. For example an autpomatic test has in practice reduced the expenditure of time under one-tenth of the manual test (e.g., from 40 h down to 2 h). This, not at least economic advantage, motivated the operator to introduce automatic electric test systems.

References

19.2.1-1 NTSB Bericht FTW891A06, microfiche number 39760A, „Scheduled 14 CFR 121 operation of Delta Air Lines, Incident occurred Mar-14-89, Aircraft Boeing 727”. (352, page 60)

19.2.1-2 Deutscher Bundestag, Drucksache 14/4867 vom 1.12.2000 „Flugzeugabstürze der Bundeswehr im Ausland und militärische Flugzeugabstürze in der Bundesrepublik Deutschland“ page 9.

19.2.1-3 Transport Canada, Civil Aviation (TCCA), Service Difficulty Advisory AV 2006-01, 8 February 2006, „Pratt & Whitney PW 100 Engine - Electrical System Care”, page 1-3.

19.2.1-4 G. Fischbach, „916 Deutsche F-104 Starfighter, ihre Bau- und Lebensgeschichten“, page 256 und 727.

19.2.1-5 NTSB Aircraft Accident Report NTSB-AAR-82.8 „Air Florida, Inc. Boeing 737-222, N62F, Collision With 14th Street Bridge Near Washingon National Airport, Washington. D.C. January 13, 1982”, page 1-136.

19.2.1-6 I.E.Traeger, „Aircraft Gas Turbine Engine Technology, Second Edition“, Glencoe Verlag, ISBN 0-07-065158-2, 1994, page 152.

19.2.1-7 Firmenprospekt, Rosemount, „ Thermoelemente für Hochtemperaturen”, Produktdatenblatt 00813-0405-2654, Rev. 2. September 2002, Seite 1-10.

19.2.1-8 AMC Publication 05-105/MSG-211, June 15, 2005, „Avionics Maintenance Conference 2005, AMC Report April 25-28, 2005, Atlanta, Georgia“, page 192 - 201.

19.2.1-9 M.J.Kroes, T.W.Wild, „Aircraft Powerplants, Seventh Edition”, Glencoe Aviation Technology Series, Glencoe Verlag, ISBN 0-02-0801874-5, 1995, page 567, 633-666.

19.2.1-10 C.Kerr, P.Ivey, „Numerical Predictions for the Performance of Pyrometer Purge Air Systems“, AIAA, ISABE-2003-1194 ISBN 0-02-0801874-5, 2003, page 1-5.

19.2.1-11 NTSB-Report LAX981LA053 , 1997, page 1-3.

19.2.1-12 R.Meyer, „Fire detection at its most sophisticated”, Zeitschrift „Aircraft Technology & Maintenance - Engine Yearbook“, 1999, page 108-111.

19.2.1-13 Transport Canada, Airworthiness Notice - B009, Edition 1-21 June 1993, „Chip Detectors in Aircraft Engines, APUs, Transmissions and Reduction Gearboxes”, page 1-3.

19.2.1-14 Australian Transport Safety Bureau, Occurrence Number 200 105 173, Aviation Safety Investigation Report-Final, „de Havilland Canada DHC-8-315, VH-TQY“, 19.12.2002, page 1-3.

19.2.1-15 G.Osborne, „Automated engine harness testing”, „Engine Yearbook 2006“, page 90-95.

19.2.1-16 „Examples of Aircraft Corrosion”, www.corrosion-doctors.org, page 1 and 2.

19.2.1-17 K.Bauerfeind, „Steuerung und Regelung der Turbotriebwerke“, Birkhäuser Verlag, 1999, ISBN 3-7643-6021-6, page 141-156.

19.2.1-18 „Aircraft Engine Manufacturer Reduces Scrap With Correct Thermocouples”, Case History # 23 Thermocouples, SIC 3724 - Aircraft Turbine Engines/Heat Treating, www.watlow.com, page 1.

19.2.1-19 Technical Information Nr. 1110-PD-001-0-00, „Identifying and Correcting Temperature Control Problems“, Fa. Barber-Colman Co., page 8-2 up to 8-12.

19.2.1-20 K.Bauerfeind, „Steuerung und Regelung der Turbotriebwerke”, Birkhäuser Verlag, 1999, ISBN 3-7643-6021-6, page 144 and 145.

19.2.1-21 Airworthiness Directive No. 96-ANE-35-AD, Amendment 39-14339, AD 2005-21-01, „Pratt & Whitney JT8D-200 Series Turbofan Engines“, page 1 -8.

219.2.1-22 AMC Reference 02-059/MSG-177 , „'Hot Start' - Turbine Engines”, Seite 233 - 243.

19.2.1-23 Reference 05-105/MSG-211, „Engine Systems“, page 192 - 201.

© 2021 ITTM & Axel Rossmann
19/192/1921/1921.txt · Last modified: 2020/06/25 22:43 (external edit)

Page Tools