Study of transmissibility between an aircraft engine and its accessories

(1)

Marco Ottolini

Study of transmissibility

between an aircraft engine

and its accessories

Tesi di Laurea Specialistica

Università di Pisa

29 Aprile 2008

(2)

Università di Pisa Facoltà di Ingegneria

Corso di Laurea Specialistica in Ingegneria Meccanica

Study of transmissibility between

an aircraft engine and its

accessories

Relatori Il candidato Prof. Ing. Leonardo Bertini Marco Ottolini Universita’ di Pisa . . . .

. . . . Prof. Ing. Marco Beghini

Universita’ di Pisa

. . . . Prof. David J. Ewins

Imperial College London

. . . .

Sessione Laurea 29 Aprile 2008 Anno accademico 2006/2007

(3)

(4)

Study of transmissibility between an aircraft engine and its accessories

Sommario

Lo scopo di questo lavoro è stato la creazione di modelli matematici in grado di simulare e studiare il diverso comportamento dinamico di una sottostruttura tra la configurazione reale e quella di prova, vale a dire tra quando essa è montata sull’assieme per la quale è stata progettata o su di una macchina di prova per il collaudo. Questo problema nasce generalmente quando i componenti vengono testati per la loro affidabilità dando luogo ad un sovra o sotto dimensionamento degli stessi. Un’analisi delle principali tipologie di test, dipendenti dal tipo di problema investigato da fenomeni di sollecitazione massima a quelli di fatica, è stata condotta nella fase preliminare. Successivamente sia le due configurazioni principali, reale e di prova, sia le tipologie di collaudo maggiormente utilizzate, sono state implementate con modelli matematici. Sui modelli matematici sono state eseguite sia analisi modali sia analisi armoniche per permetterne il confronto utilizzando il software Matlab. Infine sono state condotte le stesse tipologie di analisi su modelli più complessi, considerando come assieme, una parte di un motore aereo, e come sottostrutture, generici accessori utilizzati in campo aerospaziale. Contrariamente al caso precedente, per questo studio sono state fatte simulazioni agli elementi finiti utilizzando il software Ansys.

Abstract

The aim of this work is to create a mathematical model that enables the simulation and investigation of the different dynamic behaviour of a substructure, when it is attached to its assembly or placed on a shaker. This problem arises regularly when components are tested for their reliability and can lead to significant over or under testing, thereby introducing uncertainty into the test results. The present work includes a review of the applied test procedures depending on the problem at hand from high amplitude to fatigue failures. In a first step the two main configurations, real - on the structure - and test - on the shaker - are modelled, and different kinds of tests are simulated with a theoretical model in Matlab. Modal and harmonic analysis are carried out for these models to allow the comparison between the configurations. Afterwards a similar analysis is carried out with an FE code for more complex models, where a part of an aircraft engine is consider with it’s generic accessories as a substructures.

(5)

Acknowledgement

I thank

my tutor Dr. Christoph Schwingshackl my school mates Giuseppe and Andrea

my landlord Michele my room mate Federico

my English teachers Ramtin and Ilona my snack mates Federico and Nicolas the best secretaries ever Arianna and Nina

my Italian supervisors Professor Leonardo Bertini and Professor Marco Beghini my English supervisors Professor D.J. Ewins and Dr. Hilmi Kurt-Elli

my sponsor Programma Leonardo Francigena my sponsor Rolls Royce plc

(6)

List of Figures

1.1 Envelope of vibration environment available. B, E and S are box,

engine and shaker respectively . . . 2

1.2 Vibration environment available . . . 3

1.3 Displacement control method . . . 4

1.4 Simulated ideal state . . . 4

2.1 Single element . . . 9

2.2 The engine . . . 10

2.3 Displacement of the mass 15 from the engine . . . 11

2.4 Real configuration, box attached at one point to the engine . . . 12

2.5 Test room, box attached at one point to the engine . . . 12

2.6 The box attached at one point . . . 13

2.7 Behaviour of the box fixed using hard mount at different point to the engine, from 6th DOF up to 9th DOF . . . 13

2.8 Behaviour of the box fixed using hard mount at different point to the engine, from 10th DOF up to 15th DOF . . . 17

2.9 Behaviour of the box fixed using soft mount at different point to the engine, from 10th DOF up to 15th DOF . . . 18

2.10 Output from the box, hard mount and soft point . . . 20

2.11 Output from the engine, hard mount and soft point . . . 20

2.12 Check output from the box, hard mount and soft point . . . 21

2.13 Output from the box, soft mount soft point . . . 23

2.14 Output from the engine, soft mount soft point . . . 23

2.15 Check output from the box, soft mount soft point . . . 24

(11)

2.17 Output from the engine, hard mount and hard point . . . 26

2.18 Output from the box, soft mount and hard point . . . 28

2.19 Output from the engine, soft mount and hard point . . . 28

2.20 Whole system, box attached at two points . . . 29

2.21 Test room, box connected at two points, two different input for the shaker . . . 30

2.22 Test room, box connected at two points . . . 33

2.23 Comparison output from the box connected at two points, output from the whole system as input for the shaker . . . 33

2.24 Comparison output from the box connected at two points, output from the engine as input for the shaker . . . 34

2.25 Comparison the first output from the engine with the output from the whole system, x . . . 35

2.26 Comparison the second output from the engine with the output from the whole system, y . . . 35

2.27 Comparison output from the box connected at two points, first output from the whole system as input for the shaker . . . 36

2.28 Comparison output from the box connected at two points, second output from the whole system as input for the shaker . . . 36

2.29 Real and test configuration, box attached at four points . . . 37

2.30 Whole system, box attached at four points, disposition of DOFs . . . 38

2.31 Target response . . . 40

2.32 Monitoring response . . . 41

2.33 Comparison output from the box attached at four points, output from the whole system as input for the shaker . . . 42

2.34 Stiff mounting point . . . 45

2.35 Soft mounting point . . . 46

2.36 Check with Ansys, output from the engine . . . 47

2.37 Check with Ansys, output from the whole system connected by one point . . . 47

2.38 Check with Ansys, output from the box on the shaker connected by one point . . . 48

2.39 Check with Ansys, target point, box on the shaker connected by four points . . . 48

3.1 Pegasus . . . 51

(12)

3.3 CCOC constrained on one side, zero nodal circles family of the nodal

diameters . . . 53

3.4 5th and 6th mode shape of the CCOC clamped one side, third nodal diameter and zero nodal circles . . . 54

3.5 Zero nodal diameter and zero nodal circles . . . 55

3.6 Tenth nodal diameter and zero nodal circles . . . 55

3.7 Drawing of the probe . . . 56

3.8 System, CCOC plus probe . . . 57

3.9 Natural frequencies concerned modes shape of CCOC, System con-strained on one side . . . 60

3.10 Zero nodal circles and third nodal diameter for the System . . . 61

3.11 Shell model of the CCOC, the mesh is fit for purpose . . . 61

3.12 CCOC constrained on one side, the mesh is fit for purpose . . . 62

3.13 CCOC unclamped, free-free . . . 63

3.14 CCOC clamped both side . . . 63

3.15 System used for the harmonic analysis, force along y . . . 64

3.16 First mode shape of the probe clamped by rigid brackets . . . 65

3.17 Second mode shape of the probe clamped by rigid brackets . . . 65

3.18 Natural frequencies of CCOC, probe and system . . . 66

3.19 Real configuration and test configuration, translation along y for both points: monitoring and mounting. Uy is vertical displacement with respect to the CCOC . . . 67

3.20 Real configuration and test configuration, translation along x for both points: monitoring and mounting. Ux is axial displacement with respect to the CCOC . . . 69

3.21 Real configuration and test configuration, translation along x and y for the monitoring and mounting points respectively . . . 70

3.22 Real configuration and test configuration, translation along x for both points: monitoring and mounting. The maximum value of M ux₁ is used as input for the shaker . . . 71

3.23 Draft of the displacement control on the shaker . . . 72

3.24 Real configuration and test configuration, displacement control on the shaker, axial direction of monitoring and mounting point . . . 73

3.25 System used for the harmonic analysis, forces along Z . . . 74

3.26 Real configuration and test configuration, translation along z for both points: monitoring and mounting . . . 75

(13)

3.27 Deformed and underformed shape of the system, 1 Hz . . . 75 3.28 Comparison between the input for the shaker, mounting point response

from the system M uz1, and output from the probe on the shaker, P uz1 76

3.29 Deformed and underformed shape of the probe on the shaker, 75 Hz and , 283 Hz . . . 77 3.30 CCOC clamped by simple supports . . . 78 3.31 CCOC clamped by simple supports, zero nodal circle family of the

nodal diameters . . . 80 3.32 Comparison between CCOC clamped by simple supports (CLOSEST)

and as a cantilever . . . 81 3.33 Comparison between CCOC clamped by simple supports (CLOSEST)

and on both sides . . . 82 3.34 Real configuration and test configuration, translation along z for both

points: monitoring and mounting . . . 83 3.35 Real configuration and test configuration, translation along z for both

points: monitoring and mounting . . . 84 3.36 Mounting point response along z, M uz₁ . . . 85 3.37 Mounting point response along z, M uz₁, 0 to 5000[Hz] as frequency

range . . . 85 3.38 System with harmonic displacement along z, first mode shape . . . . 86 4.1 Ansys model of the box meshed . . . 90 4.2 Ansys model of the box, first mode shape . . . 91 4.3 Ansys model of the board, with thickness, mass elements and constraints 92 4.4 First mode shape of the first board . . . 93 4.5 Modes shape of the chips on a stiff board . . . 94 4.6 Electronic control unit attached to the CCOC using rigid brackets . 95 4.7 Real configuration and test configuration, box by rigid brackets on

the CCOC, translation along y for the heavier chip on the lighter board 96 4.8 Ansys model of the bracket . . . 97 4.9 Electronic control unit attached to the CCOC using brackets . . . . 98 4.10 Real configuration and test configuration, box by brackets on the

CCOC, translation along y for the heavier chip on the lighter board . 99 4.11 Real configuration and test configuration, box by brackets with AV on

the CCOC, translation along y for the heavier chip on the lighter board100 4.12 Displacement control on top of the box with brackets . . . 102 4.13 Displacement control on top of the box with brackets, board 1 . . . . 103

(14)

4.14 Displacement control on top of the box with brackets, board 2 . . . . 104 4.15 Displacement control on board 1, box with brackets . . . 105 4.16 Displacement control on board 1, box with brackets, board 1 . . . . 106 4.17 Displacement control on board 1, box with brackets, board 2 . . . . 107

(15)

List of Tables

2.1 Connective matrix . . . 9

2.2 Forces . . . 11

2.3 Parameters of the engine . . . 14

2.4 Connective matrix data for the engine . . . 15

2.5 Natural frequencies of the engine . . . 16

2.6 Parameters of the box . . . 16

2.7 Connective matrix data for the box attached at one point . . . 16

2.8 Natural frequencies of the box attached at one point . . . 16

2.9 Natural frequencies of the system hard mount soft point . . . 21

2.10 Natural frequencies of the system soft mount soft point . . . 22

2.11 Natural frequencies of the system hard mount hard point . . . 25

2.12 Natural frequencies of the system soft mount hard point . . . 27

2.13 Parameters of the box attached at two points . . . 30

2.14 Connective matrix data for the whole system connected at two points 31 2.15 Natural frequencies of the system connected at two points . . . 32

2.16 Parameters of the box attached at four points . . . 39

2.17 Connective matrix data for the whole system attached at four points 43 2.18 Natural frequencies of the system attached at four points . . . 44

3.1 Geometric dimensions of the CCOC . . . 50

3.2 Material properties of the CCOC . . . 50

3.3 Material properties of the probe . . . 56

4.1 Properties of the Ansys model of the box . . . 89

4.2 Properties of the Ansys model of the first board . . . 91

(16)

(17)

Introduction

This thesis is result from a research project at Imperial College London in collaboration with Rolls Royce PLC. The aim of this work is to quantify the difference in the dynamic behaviour of a substructure when it is mounted on its assembly and when it is attached to a shaker for testing. The testing of components is a significant part of the designing processes, both for validation of FE models and as check for high amplitude or fatigue failure. The main issue is that the test, carried out on the shaker for different reasons cannot reproduce accurately the complete vibration environment generating from the assembly. Therefore each test will only be an approximation of the real configuration in which the substructure will operate. There are several reasons for the difference between test and real behaviour:

• If the test is carried out at different company from the one which produces the assembly, this can only obtain the envelope of the mounting point responses because of intellectual property. The mounting points are the points where the accessory is attached to the assembly.

• Even if the complete mounting point responses is available this cannot be applied one to one by the shaker to the substructure. This is because in reality one mounting point has 6DOF s, but the shaker can apply only one of these to the accessory, generally one translational displacement orthogonal to the plane where the shaker is fixed.

• The assembly applies to the substructure a different vibration level for each mounting point, but the shaker can only apply the same mounting point response for every attachment point.

• The shaker is a very stiff body with its own dynamic, while the assembly has a totally different dynamic behaviour which influences the behaviour of the

(18)

mounted substructure.

For all these reasons it is paramount to understand what the differences between the real and test configurations are, in order to minimise over and under testing and improve the final test results.

Study of transmissibility between an aircraft engine and its accessories

Marco Ottolini

Study of transmissibility

between an aircraft engine

and its accessories

Tesi di Laurea Specialistica

Università di Pisa

29 Aprile 2008

Study of transmissibility between

an aircraft engine and its

accessories

Sommario

Abstract

Acknowledgement

Contents

List of Figures

List of Tables

Introduction