GB2484135A

GB2484135A - Testing of hollow rotary components

Info

Publication number: GB2484135A
Application number: GB1016503.3A
Authority: GB
Inventors: Eric Pierre Ravey; Himanshu P Chhabra; Shunsuke Washiya; Yizhi Wu; Wayne Van Rooyen
Original assignee: Rolls Royce PLC
Current assignee: Rolls Royce PLC
Priority date: 2010-10-01
Filing date: 2010-10-01
Publication date: 2012-04-04
Also published as: GB201016503D0

Abstract

A method of testing a hollow rotary component for which centrifugal loading is the principal in service load comprises internally pressurising the component to a pressure at which the stresses caused by in service centrifugal loading are replicated within the component. Apparatus for testing comprises a pressure rig 20 able to fixedly mount the component 10 and pressurising means. The apparatus comprises base plate 22, top plate 24, seal 28, struts 26, and support ring 30. The component may be a turbine engine nose cone or other hollow component subjected to rotation in use.

Description

TESTING OF HOLLOW ROTARY COMPONENTS

The present invention relates to testing of hollow rotary components for which the principal in service loading is centrifugal loading. Such components, typically high speed rotary components, are employed in all types of turbo machinery including gas turbine engines. The present invention is particularly applicable to testing such components that are formed from composite materials.

Background

Organic matrix composite materials can offer significant advantages over metallic alternatives for component parts of a gas turbine engine, principally in reducing the overall weight of the engine. Component parts of the fan system, including containment casings, fan blades, and secondary structures, lend themselves to composite construction owing to the relatively low temperatures at which they operate. Over these operating temperature ranges, organic matrix composites can provide the required levels of robustness, durability, strength and strain to failure. One such component that may be advantageously formed form a composite material is the nose cone.

As with other engine components, exhaustive testing of the nose cone is required to ensure adequate performance at normal operating levels and at increased operating levels, as well as to determine the component lifetime, its ultimate endurance, and the mechanism by which it fails. Low cycle fatigue testing is an important part of this testing program. As a high speed rotary component, the major in service load experienced by the nose cone is the centrifugal load that is caused by its rotation.

Other one off loads may be experienced, for example in the event of a bird strike or other foreign object impact, but the principal stresses that govern the in service performance of the nose cone are those arising from its high speed rotation.

Consequently, the established method for low cycle fatigue testing of a nose cone is spin testing.

Low cycle fatigue performance of the nose cone is determined by spin testing of the entire fan assembly, as illustrated in Figure 1. Spin testing is a costly procedure; the loading conditions required causing each individual test to be both time consuming and labour intensive. The complete fan assembly, which may include both metallic and composite components, is rotated under a cyclic loading pattern designed to replicate both normal and abnormal operation of the engine. Such testing cannot be optimised for each individual component of the fan assembly. The loading conditions for each test, including ramp rate, maximum rotational speed and dwell time, are of necessity determined by the limitations of the metallic components and the apparatus control unit, and thus may not be optimal for testing of a composite nose cone. Additionally, as fan blades or discs will tend to fail before the nose cone, it is generally the case that only a single nose cone can be spin tested, providing a poor statistical sample with which to determine the life of the nose cone. The requirement for full assembly testing also means that it is not possible to establish the life of a nose cone in isolation.

Sum mary According to the present invention, there is provided a method of testing a hollow rotary component for which centrifugal loading is a principal in service load, the method comprising internally pressurising the component to a pressure at which the stresses caused by in service centrifugal loading are replicated within the component.

For the purposes of the present specification, centrifugal loading refers to the load experienced by a component as a result of rotation of the component. Examples of components for which this load may be a principal in service load include high speed rotary components found in turbo machinery. The centrifugal load on the component may be the principal in service load experienced by the component.

By pressu rising a hollow rotary component to a level at which in service stresses resulting from centrifugal loading are replicated, it is possible to conduct fatigue, lifetime and burst testing on a rotary component more quickly and economically than has previously been possible.

The method may further comprise determining stresses generated in the component by in service centrifugal loading, calculating the internal pressure at which such stresses will be replicated in the component, and pressurising the component to the calculated internal pressure.

Calculation of the internal pressure may comprise conducting a finite element analysis of the component, running the analysis at a range of pressures and using the results to produce a correlation between internal pressure and stresses generated in the component. This correlation may be used to select the appropriate pressures to replicate in service stresses in the component.

The method may further comprise cycling the internal pressure of the component between a minimum value and the calculated value for replicating in service stresses.

The method may further comprise measuring strain in the component during pressurisation. The measured strain values may be converted to stress values and may be used to verify that the internal pressure to which the component is pressurised replicates stresses caused by in service centrifugal loading.

The method may further comprise identifying critical regions of the component and critical stresses within the critical regions. The internal pressure may be calculated to replicate the critical stresses in the critical regions. Identifying the critical stress for the component and replicating that stress enables performance of the entire component to be verified by reference only to the most critical stress generated in the most critical area of the component.

The method may be employed to conduct low cycle fatigue testing, to test for Overspeed rotational conditions and/or to test for Overspeed to burst rotational conditions. In the case of testing for Overspeed to bust rotational conditions, the method may further comprise pressuring the component to a pressure above that at which the stresses caused by in service centrifugal loading are replicated within the component and supporting the component during such increased pressurisation.

The method may further comprise measuring acoustic emissions during pressurisation.

Such acoustic emissions may provide information about micro crack formation and also evidence of the failure mechanism of the component.

The method may further comprise measuring temperature during pressurisation.

Temperature information may be used to validate and inform component analysis and may be employed in pressure calculations for materials having temperature sensitive properties. Temperature information may also be used to confirm that operating conditions for the test are representative of in service conditions.

The component may be formed from an organic matrix composite material.

The component may comprise a nose cone of a gas turbine engine.

According to another aspect of the present invention, there is provided apparatus for fatigue testing a hollow component, the apparatus comprising a pressure rig and means for pressu rising a hollow component loaded in the rig, wherein the pressure rig comprises a base plate, a first end plate operable to close a first opening in the hollow component, a seal operable to seal the interface between the component and the first end plate and a first plurality of struts, fixedly connectable between the first end plate and the base plate and operable to carry an axial load.

The apparatus may further comprise a support ring, operable to support the component within the rig, and a second plurality of struts, fixedly connectable between the support ring and the base plate and operable to carry an axial load. The apparatus may further comprise a support sleeve mounted on the support ring for mating engagement with the component over a support surface.

The apparatus may further comprise a second end plate, in opposed relation to the first end plate and operable to close a second opening in the hollow component, and a second seal, operable to seal the interface between the second end plate and the component.

The means for pressurising a hollow component loaded in the rig may comprise a hydraulic pump.

Brief description of drawings

For a better understanding of the present invention, and to show more clearly how it may be carried into effect, reference will now be made, by way of example, to the following drawings, in which:-Figure 1 is an illustrative representation of a spin test for a fan assembly; Figures 2a and 2b are views of a nose cone; Figure 3 is a partial view of a nose cone indicating stress in a critical region; Figure 4 is a graph showing pressure against time for a low cycle fatigue test; Figure 5 is a graph showing pressure against time for an Overspeed test; Figure 6 is graph showing pressure against time for an Overspeed to burst test; Figure 7 is a section through a perspective view of a pressure rig; Figure 8 is a section through a perspective view of the pressure rig of Figure 7, illustrating additional features; and Figure 9 is a schematic representation of a test set up for pressure testing a nose cone.

The present invention is described in detail with respect to the low cycle fatigue testing of a composite nose cone for a gas turbine engine, as illustrated in Figure 2. However, it will be appreciated that the method and apparatus described may be applied to the testing of a range of hollow components across the aerospace and other industries.

The present invention provides a method in which in service stresses generated by component rotation can be replicated in a test environment through internally pressu rising the component, rather than rotating it. An exemplary embodiment of such a method, together with apparatus with which the method may be conducted, is described below.

Initially, the stresses to which the nose cone is subjected in service are established.

This may be through the use of recorded in service component data or using any other sources. A critical stress analysis is conducted to determine the particular locations and individual stresses that are critical for the nose cone. Critical locations are generally those regions having the lowest material reserve factor and will commonly include areas of discontinuity around which stress "hot spots" are generated. In the exemplary nose cone 10 of Figure 2, the critical locations are the regions 12 surrounding the bolt holes, where the nose cone is attached to surrounding components via a series of bolts. The critical stress in this region is the hoop stress, indicated as load 2 on Figure 3. Other stresses include the axial stress (load 1) and the bolt compression stress (load 3).

Having established the critical stress for the nose cone 10, a Finite Element Analysis (FEA) is performed to determine the internal pressure to which the nose cone 10 must be subjected in order to replicate the critical stress at the critical bolt hole regions.

Known analytical tools can be used to model the conditions under which the nose cone 10 is to be pressurised, in order to give an accurate representation of the stresses in the nose cone 10 under pressurisation. Established FEA models for the component may require modification to account in particular for the different boundary conditions in the test apparatus compared to the in service situation. The FEA model is run at a range of pressures in order to establish a correlation between internal pressure and stress at the critical regions. The pressure required to induce the required value of critical stress in the critical regions is then deduced from this correlated relationship.

It will be appreciated that it is generally necessary to replicate more than one different value of critical stress in order to fully represent the range of engine conditions for which the nose cone must be tested. An exemplary test schedule requires three different conditions to be replicated. The first condition is the limit of normal operating conditions, known as Red line speed (RL), at which the engine must operate without damage. The second condition is Overspeed, typically 125% of Redline speed, which is representative of engine performance after a foreign object strike or other significant event. Finally, it may be necessary to test the nose cone at Overspeed to burst conditions, typically greater than 140% of Redline speed, enabling the ultimate endurance and failure mechanism of the nose cone to be established. Determining the failure mechanism and energy release during failure of the nose cone ensures that surrounding structures, such as the containment casing, can be designed with sufficient robustness to withstand failure of the nose cone.

For a complete test schedule, the value of the critical stress experienced around the bolt holes in the nose cone at each of the three conditions is established, and the FEA is used to determine the three test pressures PRL, P0 and FOB, that will replicate these stress levels when the nose cone 10 is pressurised.

Once the test pressures for the three test conditions have been calculated, the nose cone 10 is mounted in a pressure test rig, the details of which are described in full below. Strain gauges are mounted on the nose cone 10 in the location and direction of the critical stress. Additional strain gauges may also be mounted in the directions of other principal stresses at the critical regions of the nose cone.

The first test to be conducted is the low cycle fatigue, or Red line test, during which pressure is cycled between zero and PRL, as illustrated in Figure 4, for an appropriate number of cycles. If determining the lifetime of the nose cone 10 the number of cycles may be in excess of 100,000. The low cycle fatigue test ensures that there is no component damage during normal operating conditions and may be used to establish the lifetime over which the nose cone may safely function under those conditions.

The second test to be carried out is the Overspeed test, illustrated in Figure 5. During this test pressure is increased at a constant ramp rate to P0, at which it is maintained for a dwell time of for example 5 minutes, before being reduced back to zero. This test ensures that there is no component failure at Overspeed, and hence that mechanical integrity will be maintained in the event of a bird strike or other significant event.

The final test to be conducted is the Overspeed burst test, illustrated in Figure 6.

During this test pressure is increased at constant ramp rate to and above ROB until the nose cone 10 fails, allowing the failure mechanism of the nose cone to be studied.

During the Overspeed burst test, it may be desirable to provide additional support to the nosecone, as discussed more fully below.

Values registered during testing on the strain gauges mounted in the critical stress directions around the bolt hole are recorded. Strain values at test pressures are then converted to stress values and compared to the in service critical stress values that the testing is designed to replicate. The comparison between recoded values during testing and the in service values enables verification and adjustment of the FEA model used to generate the test pressures. It will be appreciated that conversion from strain to stress may involve a range of calculations and additional measurement, depending on the material properties of the nose cone.

A range of data can be collected from the pressure testing in addition to the strain data that allows verification of the FEA model. For example, temperature gauges can advantageously be applied to the nose cone 10 during testing, providing an accurate reading of the temperature of the component during testing. This may also be beneficial in validating and improving the FEA, particularly in the case of certain composite materials for which the material properties are highly temperature dependent. Temperature measurements facilitate confirmation that the material properties used in the analysis are representative of the test conditions, and also that the test conditions are representative of the actual in service conditions experienced by the nose cone.

An additional data collection tool that can be used to advantage is an acoustic sensor.

A common failure mechanism in composite components is cracking. The formation of a micro crack gives off an acoustic emission that can be detected by an acoustic sensor for later analysis. Different types and levels of cracking give off different levels of acoustic emission, all of which data can be used to give an indication of the level of damage sustained by the component during testing, and the mechanism by which the component will eventually fail. Additional strain gauges can also provide valuable data for damage and failure analysis.

It is desirable to conduct an examination of the nose cone 10 both before testing is started and after each individual test. Examination for both internal and external damage provides additional information about how the nose cone has performed in the testing, as well as ensuring that the component does not contain any initial flaws that might affect the test results. Visual inspection is preferably supported by analysis such as C scanning to check for internal micro racks and other damage.

Apparatus for pressurising a nose cone 10 is illustrated in Figures 7, 8 and 9. The apparatus comprises a pressure rig 20 and means 30 for pressurising the rig 20. With reference to Figure 7, the pressure rig comprises a base plate 22, a first or top end plate 24 that is substantially parallel to the base plate 22 and a plurality of outer struts or columns 26 extending between the base plate and the top end plate. The top end plate 24 includes a groove suitable to receive the larger diameter end of the nose cone 24. A seal 28 in the form of a polyurethane ring seals the interface between the end plate 24 and the nose cone 10. The rig 20 further comprises a support ring 30 connected via a plurality of inner struts or columns 32 to the base plate 22. The support ring 30 has an internal taper to closely receive the nose cone 10 and provide support during pressurisation. The smaller diameter opening at the tip of the nose cone is sealed by a second or bottom end plate 34 in contact with the base plate 22 and the interface between the nose cone 10 and the bottom end plate 34 is sealed by an o-ring seal (not shown). The top end plate, carrying the nose cone 10 is supported by a retaining ring 36 that is held on the plurality of outer struts 26 via a series of bolts 38.

Additional bolts 40 hold the top end plate in place and resist pressure as the rig is pressurised. The entire rig 20 is constructed within a bund (not shown) and may be surrounded by a protective wall to contain any projectiles released in the event of failure of the nose cone.

In use, as the nose cone 10 within the rig is pressurised, loads are safely transmitted through the sealing end plates, support ring, base plate and columns. In this manner, unnecessary loads, which may cause unrepresentative stresses, are not transmitted through the nose cone. Only the hoop stresses, which are representative of the in service centrifugal loading, are carried through the nose cone 10. -11 -

With reference to Figure 8, the pressure rig 20 may further comprise a support sleeve mounted on the support ring 30 for mating engagement with the nose cone 10 over a support surface. As illustrated, the support sleeve 70 comprises a conical sleeve dimensioned to closely accept the region of the nosecone adjacent to the support ring 30. The support sleeve 70 may be formed from a metal or composite material and may be removably mounted on the support ring 32. The support sleeve 70 offers a relatively large support area to the nose cone, distributing and dissipating very high pressure loads. The support sleeve 70 ensures that undesirable stress concentrations do not arise owing to point or edge contact between the nose cone 20 and the support ring 30.

Such stress concentrations may be of particular concern during the very high pressurisation associated with Overspeed burst testing, and the presence of the support sleeve thus enhances the capabilities of the pressure rig 20 under such extreme conditions.

It will be appreciated that while the rig described and illustrated in Figures 7 and 8 is designed for testing of a nose cone, non tapered components such as casings, hollow vanes and pump housings may also be tested in the rig with minor modification of the rig components. The sealing end plates may retain their simple geometry and the polyurethane sealing rings can be designed to function as adaptors to conform more complex component geometry to the simple end plate geometry.

The pressure rig 20 is represented schematically in Figure 8 together with supporting test equipment. Hydraulic oil from a supply 50 is pumped through a servo trolley 52 and valve trolley 54 to the pressure rig 20 and nose cone 10. A relief valve 56 may be used to protect the pressure rig and a pressure gauge 58 employed to monitor the pressure within the nose cone 10. Bleed valves 60 ensure that air is bled out of the system as well as providing safety features to prevent over pressurisation of the nose cone 10. The test apparatus sits within a bund 62 and a scavenger pump 64 pumps excess oil out of the bund 62.

It will be appreciated that the pump and pressure transducers used to pressurise the rig may be selected to be capable of high frequency loading, meaning that the loading cycle of the pressure rig is not constrained by the test equipment, in contrast to existing spin test equipment. Consequently, the loading cycle duration in the pressure rig can be minimised to give a shorter test and quicker determination of component life. In addition, the nose cone 10, or other component loaded in the rig, can be individually tested and its lifetime can be determined in isolation, rather than as part of a larger assembly. With shorter test times and individual testing, several nose cones can be tested to give a larger statistical sample and hence a more reliable indication of component life. The pressure rig is also simple to manufacture and to operate, again in contrast to the conventional spin testing apparatus.

Claims

CLAIMS1 A method of testing a hollow rotary component for which centrifugal loading is a principal in service load, the method comprising internally pressurising the component to a pressure at which the stresses caused by in service centrifugal loading are replicated within the component.
2 A method as claimed in claim 1, further comprising determining stresses generated in the component by in service centrifugal loading, calculating the internal pressure at which such stresses will be replicated in the component, and pressurising the component to the calculated internal pressure.
3 A method as claimed in claim 2, wherein calculation of the internal pressure comprises conducting a finite element analysis of the component, running the analysis at a range of pressures and using the results to produce a correlation between internal pressure and stresses generated in the component.
4 A method as claimed in claim 2 or 3, further comprising cycling the internal pressure of the component between a minimum value and the calculated value for replicating in service stresses.
A method as claimed in any one of the preceding claims, further comprising measuring strain in the component during pressurisation.
6 A method as claimed in claim 5, further comprising converting the measured strain values to stress values and verifying that the internal pressure to which the component is pressurised replicates stresses caused by in service centrifugal loading.

-14 -
7 A method as claimed in any one of claims 2 to 6, further comprising identifying critical regions of the component and critical stresses within the critical regions and wherein the internal pressure is calculated to replicate the critical stresses in the critical regions.
8 A method as claimed in any one of the preceding claims further comprising measuring acoustic emissions during pressurisation.
9 A method as claimed in any one of the preceding claims, further comprising measuring temperature during pressurisation A method as claimed in any one of the preceding claims, further comprising pressuring the component to a pressure above that at which the stresses caused by in service centrifugal loading are replicated within the component and supporting the component during such increased pressurisation.11 A method as claimed in any of the preceding claims, wherein the component is formed from an organic matrix composite material.12 A method as claimed in any one of the preceding claims wherein the component is nose cone of a gas turbine engine.13 Apparatus for fatigue testing a hollow component, the apparatus comprising a pressure rig and means for pressurising a hollow component loaded in the rig, wherein the pressure rig comprises a base plate, a first end plate operable to close a first opening in the hollow component, a seal operable to seal the interface between the component and the first end plate and a first plurality of struts, fixedly connectable between the first end plate and the base plate and operable to carry an axial load.14 Apparatus as claimed in claim 13, further comprising a support ring, operable to support the component within the rig, and a second plurality of struts, fixedly connectable between the support ring and the base plate and operable to carry an axial load.Apparatus as claimed in claim 14, further comprising a support sleeve mounted on the support ring for mating engagement with the component over a support surface.16 Apparatus as claimed in any one of claims 13 to 15, further comprising a second end plate, in opposed relation to the first end plate and operable to close a second opening in the hollow component, and a second seal, operable to seal the interface between the second end plate and the component.17 Apparatus as claimed in any one of claims 13 to 16, wherein the means for pressurising a hollow component loaded in the rig comprises a hydraulic pump.18 Apparatus substantially as described herein, with reference to and as shown in the Figures 7 and 8 of the accompanying drawings.