GB2563247A - Improved temperature measurement - Google Patents

Abstract

Disclosed is a temperature sensing method for use in a control system for a gas turbine engine. The method enables use of a faster thermal response thermocouple in a hotter area closer to the inlet of the turbine after the combustors of the gas turbine engine by correcting the faster response thermocouple for temperature induced sensor degradation, such as drift, using a slower thermal response thermocouple, with a stable temperature response, exposed to the same environment. The method includes providing two thermocouple sensors 8, 9 in a single housing 10; the first thermocouple 8 having a first faster thermal response time and the second thermocouple 9 having a second slower thermal response time and having a stable temperature response. The output from the two thermocouples 8, 9 is compared periodically whilst they are in situ in the environment having its temperature monitored and the first thermocouple 8 is recalibrated such that its sensed temperature output matches the sensed temperature output from the second thermocouple 9. The second slower response sensor may comprise a self-calibrating thermocouple. The second slower response sensor may comprise a cooled thermocouple. A temperature sensing apparatus is also claimed, comprising a faster thermal response time temperature sensor and a slower thermal response time temperature sensor located alongside each other, the slower thermal response time temperature sensor having a stable temperature response.

Description

IMPROVED TEMPERATURE MEASUREMENT

The present invention is concerned with method and apparatus for measuring temperature. In particular, embodiments of the invention are concerned with making temperature measurements for use in dynamic control systems requiring sensors with fast response times. Embodiments of the invention are particularly useful in engine control systems for gas turbines such as jet engines (i.e. aircraft turbines).

Modern aircraft turbines include an engine control system whose function is to allow the engine to perform at maximum efficiency for a given condition. Papers discussing fundamentals of aircraft turbine engine control can be seen online at https://en.wikipedia.org/wiki/FADEC and https://www.grc.nasa.gov/www/cdtb/aboutus/Fundamentals of Engine Controi.pdf.

To optimise engine control to give maximum efficiency in component life it is desirable to measure the gas temperature in the turbine as close as practical to the inlet of the turbine after the combustors.

Currently this is not possible due to deficiencies in the poor performance of temperature sensors (typically thermocouples) at the high temperatures found in this area of the engine. One of the issues preventing accurate measurement is so-called measurement drift of the thermocouple which is known to increase in rate with increasing temperature.

Measurement drift occurs because of metallurgical changes of the thermoelements during the operation of the thermocouple. These metallurgical changes are time dependent such that thermocouples degrade with high temperature. A explanation of measurement drift in thermocouples is set out by Michael Scervini on the Cambridge University page setting out his research activity which is at https://www.msm.cam.ac.uk/utc/thermocouple/pages/MicheieScerviniMainPage.html and which links to https://www.grc.nasa.gov/www/cdtb/aboutus/Fundamentais of Engine Controi.pdf.

Typically the very high temperature conditions at the ideal turbine location for measuring temperature means that the measurement thermocouples are located in a cooler area further away from the combustor. The temperature of the inlet of the turbine is then estimated by extrapolating from this measurement at the cooler location using a mathematical model of the engine’s temperature profile.

This estimate of the inlet temperature is inaccurate and becomes more inaccurate the further back the thermocouple is positioned. Increased margins on maximum operating temperature are therefore required to allow for uncertainty in actual sensed temperatures. This means that the engine is run at non-optimum conditions in order to protect components from possible overheating.

The present invention in a first aspect provides a temperature sensing method for use in a control system, the method comprising the steps of: providing two temperature sensors, a first sensor having a first faster thermal response time and a second sensor having a second slower thermal response time and having a stable temperature response; exposing both sensors to the same environment whose sensed temperature is for use in a closed loop control system; supplying the output from the first sensor to a data processor; and periodically comparing the output from the two temperature sensors whilst they are in situ in the environment having its temperature monitored and re-calibrating the first sensor such that its sensed temperature output matches the sensed temperature output from the second sensor.

The inventors of the subject application have appreciated that it is possible to allow the positioning of temperature sensors in hotter areas of the engine by providing a means for correcting the sensors for the errors arising from temperature- induced or affected sensor performance degradation such as drift.

Preferably the step of providing two sensors includes providing two thermocouple sensors, a first thermocouple sensor having a first faster thermal response time and a second thermocouple sensor having a second slower thermal response time and having a stable temperature response.

In this application the expressions “a stable temperature response” and/or “thermally stable” are used to describe the material property that the material’s response characteristic to temperature does not significantly alter over time (i.e. degrade because of, for example, the temperature drift effect described above).

Preferably the providing of the second sensor includes providing a self-calibrating thermocouple.

Alternatively the providing of the second sensor includes providing a cooled thermocouple.

The method may be for monitoring the temperature inside a gas turbine.

Preferably the control system is a digital engine control unit.

The invention in a second aspect provides temperature sensing apparatus for use in a control system, the temperature sensing apparatus including two temperature sensors located alongside each other, a first sensor having a first faster thermal response time and a second sensor having a second slower thermal response time and having a stable temperature response.

Preferably the two temperature sensors are located within a single probe housing.

Preferably the second sensor is a self-calibrating temperature sensor.

Alternatively the second sensor is a cooled temperature sensor.

One or both sensors may be thermocouples.

Preferred embodiments of the invention will now be described by way of non-limiting example with reference to the attached figures in which:

Figure 1 is a schematic illustration of a turbine;

Figure 2 is a schematic illustration of the cross-section of a temperature sensor embodying the invention;

Figure 3 is a detailed schematic illustration of the tip of the self-calibrating thermocouple in figure 2; and

Figure 4 is a schematic cross-section through a second temperature sensor illustrating the current invention.

Referring to figure 1, an aircraft turbine 1 (i.e. a gas or jet turbine) includes a fan 2, a compressor 3, a combustor or combustion chamber 4 and a turbine 5. The fan 2 at the front of the engine 1 sucks cold air into the engine and forces it through the inlet. The compressor 3 (which is effectively a second fan) then compresses the air (i.e. increases its pressure) and thereby significantly increases its temperature. Fuel is supplied to the combustor 4 which is just behind the compressor 3. The fuel mixes with the compressed air and burns fiercely, giving off hot exhaust gases and producing a significant increase in temperature.

The exhaust gases rush past the turbine blades, spinning them. Since the turbine 5 gains energy, the gases must lose the same amount of energy—and they do so by cooling down slightly and losing pressure.

The turbine blades are connected to a shaft 6 that runs the length of the engine 1. The compressor blades and the fan are also connected to this shaft. So, as the turbine blades spin, they also turn the compressor 3 and the fan 2.

The hot exhaust gases exit the engine through a tapering exhaust nozzle. The backwardmoving exhaust gases power the jet forward.

The behaviour of an engine may modelled and used in an engine control system. Modern digital engine control systems aim to allow the engine or turbine to operate at maximum efficiency for a given condition. An engine control system works by receiving multiple input variables of the current flight condition including air density, throttle lever position, engine temperatures, engine pressures, and many other parameters. The inputs are received by the control system may be analysed up to 70 times per second. Engine operating parameters such as fuel flow, stator vane position, bleed valve position, and others are computed from this data and applied as appropriate. A digital engine control system may also control engine starting and restarting. A digital engine control system may also allow the manufacturer to program engine limitations and receive engine health and maintenance reports. For example, to avoid exceeding a certain engine temperature, the engine control system can be programmed to automatically take the necessary measures without pilot intervention.

An efficient and accurate engine control system requires accurate measurements including of temperature in the very hostile environment inside a gas turbine. The measurements must also be fast so that the measurement system and the engine control system can react to a step change in temperature. Slow responding measurements systems could lead to the exceeding of desired or limit system parameters (e.g. temperature) and cause inefficient operation, damage to the engine or even failure.

An important temperature needed for an effective digital engine control system is the temperature at the inlet to the turbine. However, in practice this is not possible due to deficiencies in the performance of temperature sensors (currently thermocouples) at the high temperatures found in this area of the engine. One of the issues is measurement drift of the thermocouple, which is known to increase in rate with increasing temperature. To avoid this issue the thermocouples are located in a cooler area (position C shown in figure 1), further away from the combustors. The temperature at the inlet of the turbine (position A shown in figure 1) is estimated by extrapolating from this measurement using a mathematical model of the engine’s temperature profile.

This estimate of the inlet temperature is inaccurate and becomes more inaccurate the further back the thermocouple is positioned. Increased margins on maximum operating temperature are required to allow for uncertainty in actual temperature. This means that the engine is run at non-optimum conditions in order to protect components from possible overheating.

Referring to figures 2 and 3, a sensor assembly 7 contains two thermocouples 8,9 of mineral insulated cable construction mounted in a single housing 10. The housing would be of standard construction: typically a Nickel alloy such as Haynes 230 (Haynes and 230 are trade marks of Haynes International Inc). Haynes 230 alloy is a nickel-chromium-tungsten-molybdenum alloy that combines excellent high temperature strength, outstanding resistance to oxidizing environments up to 2100°F (1149°C) for prolonged exposures, resistance to nitriding environments, and long-term thermal stability. It is readily fabricated and formed, and is castable. Other attractive features include lower thermal expansion characteristics than most high-temperature alloys, and a pronounced resistance to grain coarsening with prolonged exposure to high-temperatures.

The housing has an inlet 11 and outlet aperture 12 to guide the flow of gas in the turbine over the two sensing elements 8,9 to aide uniform flow. In an alternative embodiment (not shown) the housing could be omitted. The two temperature sensors 8,9 or thermocouples would however be mounted on a common flange while having exposed tips.

One of the thermocouples is a measuring thermocouple 8 whose output is used as an input to the engine control system. The measuring thermocouple 8 is optimised for response time by reducing its diameter towards the tip to reduce the thermal mass and thermal path to the thermocouple hot junction, while being suitably thicker further along its length to provide mechanical strength. In this embodiment the thermocouple is of Type K construction (i.e. Nickel-Chromium/Nickel-Alumel). (Alumel is a trade mark of Concept Alloys, Inc). The thermocouple hot junction may be grounded or isolated from the external sheath.

The second or reference thermocouple is a self-calibrating thermocouple 9. Self-calibrating thermocouples use self-validating thermocouple technology, which incorporates 'fixed-points' (materials which melt at well-defined temperatures) made of, for example, metal-carbon eutectic alloys. This means that the thermocouple is able to self-validate, or check, that it is measuring the temperature correctly, by adjusting its calibration in response to the melting of the fixed point alloy.

The second, self-calibrating or reference thermocouple 9 is again of K-type construction but larger in diameter to improve its long term stability while incorporating a self-calibration fixed point cell. One such method is to use a metal-carbon eutectic fixed-point cell 14. The thermocouple junction 13 is embedded into the fixed point cell with sufficient thermal conductivity that the thermocouple tracks the temperature of the fixed point cell 14. The outer case of the fixed point cell may be manufactured from a suitable inert material e.g. Alumina, sealed to the metallic sheath of the thermocouple. The phase change material 15 could be Iron to form a Fe-C reference point with an inner graphite crucible. The Fe-C reference point would give a self-calibration point at ~1153°C. In calibration mode, the inflection point of the melting point represents the accurate reference temperature to establish the drift of the reference thermocouple which in turn can be compared to the first measuring thermocouple.

The measuring temperature sensor 8 (a thermocouple in the embodiments of figures 2 to 4) and the reference temperature sensor 9 (a thermocouple in the embodiments of figures 2 to 4) needs to read the same temperature. They should therefore be at the same height and radial position within the engine they are monitoring.

Alternative embodiments of the invention may replace the self-calibrating reference thermocouple with another reference thermocouple 16 which is thermally stable (i.e. not prone to significant degradation or impaired performance at higher temperatures).

Referring to figure 4, the second reference thermocouple 16 shown there has more insulation than the first thermocouple and therefore has a larger thermal mass. The higher thermal mass means that the second reference thermocouple is more thermally stable but has a slower response time: it takes longer for the thermocouple to heat up sufficiently for both thermal degradation and for an increase in temperature to take effect.

Further alternative reference thermocouples include cooled thermocouples (including air or liquid cooled thermocouples), thermocouples having a geometry or shape which encourages cooling (e.g. a ridged exterior), or thermocouples made of materials making them more thermally stable at the likely temperature range in which the apparatus is to operate.

Alternative embodiments of the invention may use other forms of temperature sensor such as bare wire thermocouples and/or RTDs (resistance temperature detectors). The reference thermocouple might be any temperature sensor technology for measuring the reference (low drift) temperature, for example but not limited to thermocouples, optical sensors, cooled temperature sensors.

The low drift/self-calibrating thermocouple 9 itself is not able to be used as an engine performance parameter due to its high thermal mass and resulting slow thermal response. The offset between the fast response time measuring thermocouple and the low drift/self-calibrating reference thermocouple are known at TO when the probes are first manufactured via characterisation, for example using a wind tunnel. The change in offset (i.e. drift) of measuring probe can be calculated from the accurately known low drift/self-calibrating reference thermocouple probe. The measurement of drift using the carbon-eutectic freeze point is a well-known science and inflection point of the melting point can be observed, accurately giving a known reference temperature to allow self-calibration of the reference thermocouple.

During the inflection point of the melting point (i.e. the melt and freeze plateau) of the fixed-point cell the output of the reference probe 9 or thermocouple 8 will lag that of the engine and the measuring probe or thermocouple. The thermocouple offsets should not be calculated during this time; it is assumed that this offset between measuring probe and reference probe is calculated as near to the melt point as possible, before this thermal lag takes place when isothermal conditions are present.

Possible enhancements include: • Depending on the stability of the engine temperature, the characterisation may need to be conducted during a thermal transient. Due to this transient the response time differences between the measuring and reference probes will also need to be taken into account via characterisation in a wind tunnel at known flow conditions. • The reference thermocouple in the freeze point cell could be optimised for drift performance in a number of ways using drift reduction technology such as dual wall configurations illustrated in WO 2011/121313. • This method could also be applied to a number of fast response time thermocouples around the circumference of the engine if the offsets are known at TO and they are individually measured.

Claims (10)

maims

1. A temperature sensing method for use in a control system, the method comprising the steps of: i) providing two temperature sensors, a first sensor having a first faster thermal response time and a second sensor having a second slower thermal response time and having a stable temperature response; ii) exposing both sensors to the same environment whose sensed temperature is for use in a closed loop control system; iii) supplying the output from the first sensor to a control data processor; and iv) periodically comparing the output from the two temperature sensors whilst they are in situ in the environment having its temperature monitored and re-calibrating the first sensor such that its sensed temperature output matches the sensed temperature output from the second sensor.

2. A method according to claim 1, wherein the step of providing two sensors includes providing two thermocouple sensors, a first thermocouple sensor having a first faster thermal response time and a second thermocouple sensor having a second slower thermal response time and having a stable temperature response.

3. A method according to claim 2 wherein the providing of the second sensor includes providing a self-calibrating thermocouple.

4. A method according to claim 2 wherein the providing of the second sensor includes providing a cooled thermocouple.

5. A method according to any preceding claim for monitoring the temperature inside a gas turbine.

6. A method according to any preceding claim wherein the control system is a digital engine control unit.

7. Temperature sensing apparatus for use in a control system, the temperature sensing apparatus including two temperature sensors located alongside each other, a first sensor having a first faster thermal response time and a second sensor having a second slower thermal response time and having a stable temperature response.

8. Apparatus according to claim 7 wherein the two temperature sensors are located within or held by common probe mount or mounting flange.

9. Apparatus according to claim 7 or claim 8 wherein the second sensor is a selfcalibrating temperature sensor.

10. Apparatus according to any of claims 7 to 9 wherein one or both sensors are thermocouples.