Calvi Students Liege 2011 6(1)

download Calvi Students Liege 2011 6(1)

of 126

Transcript of Calvi Students Liege 2011 6(1)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    1/126

    Spacecraft Loads Analysis - A. Calvi 1

    This presentation is distributed to the students of the University of Liege

    Satellite Engineering Class November 21, 2011

    This presentation is not for further distribution

    Spacecraft Loads AnalysisAn Overview

    Adriano Calvi, PhD

    ESA / ESTEC, Noordwijk, The Netherlands

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    2/126

    Spacecraft Loads Analysis - A. Calvi 2

    Spacecraft Loads Analysis - Contents

    1. Introduction and general aspects

    2. Mechanical environment

    3. Requirements for spacecraft structures

    4. Mathematical models and structural analyses5. Spacecraft mechanical testing

    6. Mathematical models validation

    7. Conclusions

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    3/126

    Spacecraft Loads Analysis - A. Calvi 3

    Example of satellite structural design conceptExample of satellite structural design concept

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    4/126

    Spacecraft Loads Analysis - A. Calvi 4

    What is Loads AnalysisWhat is Loads Analysis

    The task of loads analysis

    Loads analysis substantially means establishing appropriate

    loads for design and testing.

    The goal or purpose of loads analysis

    Nearly always to support design or to verify requirements fordesigned or built hardware.

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    5/126

    Spacecraft Loads Analysis - A. Calvi 5

    Spacecraft loads analysis processSpacecraft loads analysis process disciplinesdisciplines

    It is 3 years that I work in thiscompany. Now, finally I have

    understood what I do, butstill I have to understand why.

    Legal aspects

    Requirements, contracts

    Philosophical aspects

    General logic, verification approach, criteria

    Physics

    Structural dynamics, validation of mathematical

    models, criteria

    Mathematics

    Computational models, verification of models

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    6/126

    Spacecraft Loads Analysis - A. Calvi 6

    OrganizationsOrganizations andand Levels of AssemblyLevels of Assembly an examplean example

    Launcher Authority

    Spacecraft Authority

    Spacecraft Prime Contractor

    Payload Contractor

    Other Contactors

    Spacecraft + launcher

    Spacecraft

    Spacecraft

    Instruments/sub-systems

    Units/components/parts

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    7/126

    Spacecraft Loads Analysis - A. Calvi 7

    Levels of AssemblyLevels of Assembly

    RFFE

    RPM

    RFFE

    RPM

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    8/126

    Spacecraft Loads Analysis - A. Calvi 8

    Design Loads CyclesDesign Loads Cycles

    A load cycle is the process of:

    Generating and combining math models for a proposed design

    Assembling and developing forcing functions, load factors, etc. to

    simulate the critical loading environment

    Calculating design loads and displacements for all significantground, launch and mission events

    Assessing the results to identify design modifications or risks

    Then, if necessary, modifying the design accordingly or choosing to

    accept the risk

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    9/126

    Spacecraft Loads Analysis - A. Calvi 9

    Design loads cycle processDesign loads cycle process

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    10/126

    Spacecraft Loads Analysis - A. Calvi 10

    Loads and Factors

    Expendable launch vehicles,

    pressurized hardware and

    manned system Test Logic

    Common Design LogicSatellites

    Test Logic

    Limit Loads - LL

    Design Limit Loads

    DLL

    x Coef. A

    DYL

    x Coef. B

    DUL

    x Coef. C

    x KQ x KA

    QLAL

    x KQ x KA

    QL

    AL

    Increasing

    LoadL

    evel

    ECSS E-ST-32-10

    Protoflight Prototype

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    11/126

    Spacecraft Loads Analysis - A. Calvi 11

    2. Mechanical environment2. Mechanical environment

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    12/126

    Spacecraft Loads Analysis - A. Calvi 12

    Mechanical loads are caused by:Mechanical loads are caused by:

    Transportation

    Rocket Motor Ignition Overpressure Lift-off Loads

    Engine/Motor Generated Acoustic Loads

    Engine/Motor Generated Structure-borne Vibration Loads

    Engine/Motor Thrust Transients Pogo Instability, Solid Motor Pressure Oscillations

    Wind and Turbulence, Aerodynamic Sources

    Liquid Sloshing in Tanks

    Stage and Fairing Separation Loads Pyrotechnic Induced Loads

    Manoeuvring Loads

    Flight Operations, Onboard Equipment Operation

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    13/126

    Spacecraft Loads Analysis - A. Calvi 13

    AccelerationsAccelerations some remarkssome remarks

    The parameter most commonly used (in the industry) to define themotion of a mechanical system is the acceleration

    Good reasons: accelerations are directly related to forces/stresses andeasy to specify and measure

    In practice accelerations are used as a measure of the severity of themechanical environment

    Some hidden assumptions Criteria for equivalent structural damage (e.g. shock response spectra)

    Note: failures usually happen in the largest stress areas, regardless if theyare the largest acceleration areas!

    Rigid or static determinate junction (e.g. quasi-static loads) Important consequences

    Need for considering the actual (e.g. test or flight) boundary conditions(e.g. for the purpose of notching)

    Need for a valid F.E. model (e.g. to be used for force and stress recovery)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    14/126

    Spacecraft Loads Analysis - A. Calvi 14

    Launch mechanical environmentLaunch mechanical environment

    Steady state accelerations

    Low frequency vibrations Broad band vibrations

    Random vibrations

    Acoustic loads

    Shocks

    Loads (vibrations) are transmitted to the payload (e.g. satellite)

    through its mechanical interface

    Acoustic loads also directly excite payload surfaces

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    15/126

    Spacecraft Loads Analysis - A. Calvi 15

    SteadySteady--statestate and lowand low--frequency transient accelerationsfrequency transient accelerations

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    16/126

    Spacecraft Loads Analysis - A. Calvi 16

    Acoustic LoadsAcoustic Loads

    During the lift off and the early phases ofthe launch an extremely high level of

    acoustic noise surrounds the payload

    The principal sources of noise are: Engine functioning

    Aerodynamic turbulence

    Acoustic noise (as pressure waves)impinging on light weight panel-like

    structures produce high response

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    17/126

    Spacecraft Loads Analysis - A. Calvi 17

    Broadband and high frequency vibrationsBroadband and high frequency vibrations

    Broad band random vibrations are produce by:

    Engines functioning Structural response to broad-band acoustic loads

    Aerodynamic turbulent boundary layer

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    18/126

    Spacecraft Loads Analysis - A. Calvi 18

    ShocksShocks

    Mainly caused by the actuation of pyrotechnic devices:

    Release mechanisms for stage and satellite separation Deployable mechanisms for solar arrays etc.

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    19/126

    Spacecraft Loads Analysis - A. Calvi 19

    ShocksShocks

    Time [t]Frequency [Hz]

    [g][g]

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    20/126

    Spacecraft Loads Analysis - A. Calvi 20

    Static and dynamic environment specification (typical ranges)Static and dynamic environment specification (typical ranges)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    21/126

    Spacecraft Loads Analysis - A. Calvi 21

    Static and dynamic environment specification (typical ranges)Static and dynamic environment specification (typical ranges)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    22/126

    Spacecraft Loads Analysis - A. Calvi 22

    QuasiQuasi--Static Loads (accelerations)Static Loads (accelerations)

    Loads independent of time or which vary slowly, so that the dynamicresponse of the structure is not significant (ECSS-E-ST-32). Note: this is

    the definition of a quasi-static event! Combination of static and low frequency loads into an equivalent static

    load specified for design purposes as C.o.G. acceleration (e.g. NASA RP-1403, NASA-HDBK-7004). Note: this definition is fully adequate for thedesign of the spacecraft primary structure. For the design of components

    the contribution of the high frequency loads, if relevant, is included as well!

    CONCLUSION: quasi static loading means under steady-stateaccelerations (unchanging applied force balanced by inertia loads). Fordesign purposes (e.g. derivation of design limit loads, selection of the

    fasteners, etc.), the quasi-static loads are normally calculated bycombining both static and dynamic load contributions. In this context thequasi static loads are equivalent to (or interpreted by the designer as)static loads, typically expressed as equivalent accelerations at the C.o.G.

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    23/126

    Spacecraft Loads Analysis - A. Calvi 23

    3. Requirements for spacecraft structures3. Requirements for spacecraft structures

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    24/126

    Spacecraft Loads Analysis - A. Calvi 24

    Typical Requirements for Spacecraft Structures

    Strength

    Structural life

    Structural response

    Stiffness

    Damping

    Mass Properties

    Dynamic Envelope Positional Stability

    Mechanical Interface

    Basic requirement: the structure shall support the payload andspacecraft subsystems with enough strength and stiffness topreclude any failure (rupture, collapse, or detrimental deformation)that may keep them from working successfully.

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    25/126

    Spacecraft Loads Analysis - A. Calvi 25

    Requirements evolutionRequirements evolution

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    26/126

    Spacecraft Loads Analysis - A. Calvi 26

    Some definitionsSome definitions

    Design:

    The process used to generate the set information describing theessential characteristics of a product (ECSS-P-001A)

    Design means developing requirements, identifying options, doinganalyses and trade studies, and defining a product in enough detail soit can be built (T. P. Sarafin)

    Verification: Confirmation by examination and provision of objective evidence thatspecified requirements have been fulfilled (ISO 8402:1994)

    Verification means providing confidence through disciplined steps thata product will do what it is supposed to do (T. P. Sarafin)

    Note: we can prove that the spacecraft satisfies the measurable criteriawe have defined, but we cannot prove a space mission will be successful

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    27/126

    Spacecraft Loads Analysis - A. Calvi 27

    Design requirements and verificationDesign requirements and verification

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    28/126

    Spacecraft Loads Analysis - A. Calvi 28

    Examples of (Mechanical) Requirements (1)

    The satellite shall be compatible with 2 launchers (potential candidates:

    VEGA, Soyuz in CSG, Rockot, Dnepr)...

    The satellite and all its units shall withstand applied loads due to the

    mechanical environments to which they are exposed during the service-life

    Design Loads shall be derived by multiplication of the Limit Loads by a design

    factor equal to 1.25 (i.e. DL= 1.25 x LL)

    The structure shall withstand the worst design loads without failing orexhibiting permanent deformations.

    Buckling is not allowed.

    The natural frequencies of the structure shall be within adequate bandwidths

    to prevent dynamic coupling with major excitation frequencies

    The spacecraft structure shall provide the mounting interface to the launch

    vehicle and comply with the launcher interface requirements.

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    29/126

    Spacecraft Loads Analysis - A. Calvi 29

    Examples of (Mechanical) Requirements (2)

    All the Finite Element Models (FEM) prepared to support the mechanical

    verification activities at subsystem and satellite level shall be delivered in

    NASTRAN format

    The FEM of the spacecraft in its launch configuration shall be detailed enough to

    ensure an appropriate derivation and verification of the design loads and of the

    modal response of the various structural elements of the satellite up to 140 Hz

    A reduced FEM of the entire spacecraft correlated with the detailed FEM shall bedelivered for the Launcher Coupled Loads Analysis (CLA)

    The satellite FEMs shall be correlated against the results of modal survey tests

    carried out at complete spacecraft level, and at component level for units above

    50 kg

    The structural model of the satellite shall pass successfully qualification sine

    vibration Test.

    The flight satellite shall pass successfully acceptance sine vibration test.

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    30/126

    Spacecraft Loads Analysis - A. Calvi 30

    Spacecraft stiffness requirements for different launchersSpacecraft stiffness requirements for different launchers

    Launch vehicle manuals specify minimum values for the payload natural(fundamental) frequency of vibration in order to avoid dynamic coupling betweenlow frequency dynamics of the launch vehicle and payload modes

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    31/126

    Spacecraft Loads Analysis - A. Calvi 31

    4. Mathematical models and structural analyses4. Mathematical models and structural analyses

    4.1 Dynamic analysis types - Overview

    4.2 Effective mass concept

    4.3 Launcher/Spacecraft coupled loads analysis

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    32/126

    Spacecraft Loads Analysis - A. Calvi 32

    Dynamic analysis typesDynamic analysis types

    Real eigenvalue analysis (undamped free vibrations)

    Modal parameter identification, etc.

    Linear frequency response analysis (steady-state response oflinear structures to loads that vary as a function of frequency)

    Sine test prediction, transfer functions calculation, LV/SC CLA etc.

    Linear transient response analysis (response of linear structures to

    loads that vary as a function of time). LV/SC CLA, base drive analysis, jitter analysis, etc.

    Shock response spectrum analysis

    Specification of equivalent environments (e.g. equivalent sine input),

    Shock test specifications, etc. Vibro-acoustics (FEM/BEM, SEA) & Random vibration analysis

    Vibro-acoustic test prediction & random vibration environment definition

    Loads analysis for base-driven random vibration

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    33/126

    Spacecraft Loads Analysis - A. Calvi 33

    Reasons to compute normal modes (real eigenvalue analysis)Reasons to compute normal modes (real eigenvalue analysis)

    To verify stiffness requirements

    To assess the dynamic interaction between a component and itssupporting structure

    To guide experiments (e.g. modal survey test)

    To validate computational models (e.g. test/analysis correlation)

    As pre-requisite for subsequent dynamic analyses To evaluate design changes

    Mathematical model quality check (model verification)

    Numerical methods: Lanczos,

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    34/126

    Spacecraft Loads Analysis - A. Calvi 34

    Real eigenvalue analysisReal eigenvalue analysis

    Note: mode shape normalizationScaling is arbitraryConvention: Mass, Max or Point

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    35/126

    Spacecraft Loads Analysis - A. Calvi 35

    Mode shapesMode shapes

    Cantilever beam Simply supported beam

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    36/126

    Spacecraft Loads Analysis - A. Calvi 36

    Satellite Normal Modes Analysis

    Mode 1: 16.2 Hz Mode 2: 18.3 Hz

    INTEGRAL Satellite (FEM size 120000 DOFs)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    37/126

    Spacecraft Loads Analysis - A. Calvi 37

    Frequency Response AnalysisFrequency Response Analysis

    Used to compute structural response to steady-state harmonic

    excitation

    The excitation is explicitly defined in the frequency domain

    Forces can be in the form of applied forces and/or enforced

    motions

    Two different numerical methods: direct and modal Damped forced vibration equation of motion with harmonic

    excitation:

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    38/126

    Spacecraft Loads Analysis - A. Calvi 38

    Frequency response considerationsFrequency response considerations

    If the maximum excitation frequency is much less than the lowestresonant frequency of the system, a static analysis is probably

    sufficient

    Undamped or very lightly damped structures exhibit large dynamic

    responses for excitation frequencies near natural frequencies

    (resonant frequencies)

    Use a fine enough frequency step size (f) to adequately predict

    peak response.

    Smaller frequency spacing should be used in regions near resonant

    frequencies, and larger frequency step sizes should be used in

    regions away from resonant frequencies

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    39/126

    Spacecraft Loads Analysis - A. Calvi 39

    Harmonic forced response with dampingHarmonic forced response with damping

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    40/126

    Spacecraft Loads Analysis - A. Calvi 40

    Transient Response AnalysisTransient Response Analysis

    Purpose is to compute the behaviour of a structure subjected to time-

    varying excitation

    The transient excitation is explicitly defined in the time domain

    Forces can be in the form of applied forces and/or enforced motions

    The important results obtained from a transient analysis are typically

    displacements, velocities, and accelerations of grid points, andforces and stresses in elements

    Two different numerical methods: direct (e.g. Newmark) and modal

    (e.g. Lanczos + Duhamels integral or Newmark)

    Dynamic equation of motion:

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    41/126

    Spacecraft Loads Analysis - A. Calvi 41

    Modal Transient Response AnalysisModal Transient Response Analysis

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    42/126

    Spacecraft Loads Analysis - A. Calvi 42

    Transient response considerationsTransient response considerations

    The integration time step must be small enough to represent accuratelythe variation in the loading

    The integration time step must also be small enough to represent the

    maximum frequency of interest (cut-off frequency)

    The cost of integration is directly proportional to the number of time steps Very sharp spikes in a loading function induce a high-frequency transient

    response. If the high-frequency transient response is of primary

    importance in an analysis, a very small integration time step must be

    used

    The loading function must accurately describe the spatial and temporal

    distribution of the dynamic load

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    43/126

    Spacecraft Loads Analysis - A. Calvi 43

    ShockShock response spectrum (and analysis)response spectrum (and analysis)

    Response spectrum analysis is an approximate method ofcomputing the peak response of a transient excitation applied to a

    structure or component There are two parts to response spectrum analysis: (1) generation

    of the spectrum and (2) use of the spectrum for dynamic responsesuch as stress analysis

    Note 1: the part (2) of the response spectrum analysis has alimited use in structural dynamics of spacecraft (e.g. preliminarydesign) since the accuracy of the method may be questionable

    Note 2: the term shock can be misleading (not always a physicalshock, i.e. an environment of a short duration, is involved. Itwould be better to use response spectrum)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    44/126

    Spacecraft Loads Analysis - A. Calvi 44

    Generation of a response spectrum (1)Generation of a response spectrum (1)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    45/126

    Spacecraft Loads Analysis - A. Calvi 45

    Generation of a response spectrum (2)Generation of a response spectrum (2)

    the peak response for one oscillator does not necessarily occur at the

    same time as the peak response for another oscillator

    there is no phase information since only the magnitude of peak response iscomputed

    It is assumed in this process that each oscillator mass is very small relative

    to the base structural mass so that the oscillator does not influence the

    dynamic behaviour of the base structure

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    46/126

    Spacecraft Loads Analysis - A. Calvi 46

    Shock Response Spectrum. Some remarksShock Response Spectrum. Some remarks

    The 1-DOF system is used as reference structure (since thesimplest) for the characterization of environments (i.e.

    quantification of the severity equivalent environments can bespecified)

    In practice, the criterion used for the severity is the maximumresponse which occurs on the structure (note: another criterionrelates to the concept of fatigue damage)

    A risk in comparing two excitations of different nature is in theinfluence of damping on the results (e.g. maxima are proportionalto Q for sine excitation and variable for transient excitation!)

    The absolute acceleration spectrum is used, which provides

    information about the maximum internal forces and stresses The shock spectrum is a transformation of the time history which is

    not reversible (contrary to Fourier transform)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    47/126

    Spacecraft Loads Analysis - A. Calvi 47

    Shock Response SpectrumShock Response Spectrum

    (A) is the shock spectrum

    of a terminal peaksawtooth (B) of 500 Gpeak amplitude and 0.4millisecond duration

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    48/126

    Spacecraft Loads Analysis - A. Calvi 48

    Random vibration (analysis)Random vibration (analysis)

    Random vibration is vibration that can be described only in astatistical sense

    The instantaneous magnitude is not known at any given time;rather, the magnitude is expressed in terms of its statisticalproperties (such as mean value, standard deviation, and probabilityof exceeding a certain value)

    Examples of random vibration include earthquake ground motion,wind pressure fluctuations on aircraft, and acoustic excitation dueto rocket and jet engine noise

    These random excitations are usually described in terms of apower spectral density (PSD) function

    Note: in structural dynamics of spacecraft, the random vibrationanalysis is often performed with simplified techniques (e.g. basedon Miles equation + effective modal mass models)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    49/126

    Spacecraft Loads Analysis - A. Calvi 49

    Random noise with normal amplitude distributionRandom noise with normal amplitude distribution

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    50/126

    Spacecraft Loads Analysis - A. Calvi 50

    Power Spectral Density (conceptual model)Power Spectral Density (conceptual model)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    51/126

    Spacecraft Loads Analysis - A. Calvi 51

    Sound Pressure Level (conceptual model)Sound Pressure Level (conceptual model)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    52/126

    Spacecraft Loads Analysis - A. Calvi 52

    VibroVibro acoustic analysis at spacecraft levelacoustic analysis at spacecraft level

    Detailed analysis using Finite Elements (FE), Boundary Elements

    (BEM) and Statistical Energy Analysis (SEA)

    Random levels on units and instruments can be compared to

    specifications or qualification levels

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    53/126

    Spacecraft Loads Analysis - A. Calvi 53

    4. Mathematical models and structural analyses4. Mathematical models and structural analyses

    4.1 Dynamic analysis types - Overview

    4.2 Effective mass concept

    4.3 Launcher/Spacecraft coupled loads analysis

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    54/126

    Spacecraft Loads Analysis - A. Calvi 54

    Modal effective mass (1)Modal effective mass (1)

    It may be defined as the mass terms in a modal expansion of the

    drive point apparent mass of a kinematically supported system

    Note: driving-point FRF: the DOF response is the same as the excitation

    This concept applies to structure with base excitation

    Important particular case: rigid or statically determinate junction

    It provides an estimate of the participation of a vibration mode, interms of the load it will cause in the structure, when excited

    Note: avoid using: it is the mass which participates to the mode!

    Dynamic amplification factor

    Modal reaction forces

    Base (junction) excitation

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    55/126

    Spacecraft Loads Analysis - A. Calvi 55

    Modal effective mass (2)Modal effective mass (2)

    The effective mass matrix can be calculated either by the modal

    participation factors or by using the modal interface forces Normally only the values on the leading diagonal of the modal

    effective mass matrix are considered and expressed in percentageof the structure rigid body properties (total mass and second

    moments of inertia) The effective mass characterises the mode and it is independent

    from the eigenvector normalisation

    Gen. mass

    Resultant of modal interface forces

    i-th mode Rigid body modes

    Modal participation factorsEffective mass

    Eigenvector max value

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    56/126

    Spacecraft Loads Analysis - A. Calvi 56

    Modal effective mass (3)Modal effective mass (3)

    For the complete set of modes the summation of the modal

    effective mass is equal to the rigid body mass

    Contributions of each individual mode to the total effective mass

    can be used as a criterion to classify the modes (global or local)

    and an indicator of the importance of that mode, i.e. an indication of

    the magnitude of participation in the loads analysis

    It can be used to construct a list of important modes for the

    test/analysis correlation and it is a significant correlation parameter

    It can be used to create simplified mathematical models (equivalent

    models with respect to the junction)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    57/126

    Spacecraft Loads Analysis - A. Calvi 57

    Example of Effective Mass table

    (MPLM test and FE model)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    58/126

    Spacecraft Loads Analysis - A. Calvi 58

    4. Mathematical models and structural analyses4. Mathematical models and structural analyses

    4.1 Dynamic analysis types - Overview

    4.2 Effective mass concept

    4.3 Launcher/Spacecraft coupled loads analysis

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    59/126

    Spacecraft Loads Analysis - A. Calvi 59

    A5 Typical Sequence of eventsA5 Typical Sequence of events

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    60/126

    Spacecraft Loads Analysis - A. Calvi 60

    A5 Typical Longitudinal Static AccelerationA5 Typical Longitudinal Static Acceleration

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    61/126

    Spacecraft Loads Analysis - A. Calvi 61

    Sources of Structural Loadings (Launch)Sources of Structural Loadings (Launch)

    Axial-Acceleration Profile for the Rockot Launch Vehicle

    g

    8

    7

    6

    5

    4

    3

    2

    1

    0t,150 200 250 300100500

    0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1-20

    -10

    0

    10

    20

    30

    40

    50

    60

    70

    80

    acceleration[m/s2]

    t [s]

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    62/126

    Spacecraft Loads Analysis - A. Calvi 62

    Axial Acceleration at Launcher/Satellite Interface (Engines Cut-off)

    0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1-20

    -10

    0

    10

    20

    30

    40

    50

    60

    70

    80

    acceleration[m/s2]

    t [s]

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    63/126

    Spacecraft Loads Analysis - A. Calvi 63

    Load Factors for Preliminary Design (Ariane 5)Load Factors for Preliminary Design (Ariane 5)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    64/126

    Spacecraft Loads Analysis - A. Calvi 64

    QuasiQuasi--Static Flight Limit loads for Dnepr and SoyuzStatic Flight Limit loads for Dnepr and Soyuz

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    65/126

    Spacecraft Loads Analysis - A. Calvi 65

    Launcher / Satellite C.L.A.

    Mode 18: 2.93 Hz Mode 53: 16.9 Hz

    A5 / Satellite Recovered System Mode shapes

    CLA: simulation of the structural response tolow frequency mechanical environment

    Main Objective: to calculate the loads on thesatellite caused by the launch transients (lift-off, transonic, aerodynamic gust, separation ofSRBs)

    Loads (in this context): set of internal forces,displacements and accelerations thatcharacterise structural response to the appliedforces

    Effects included in the forcing functions :thrust built-up, engine shut-down/burnout,gravity, aerodynamic loads (gust), separationof boosters, etc.

    UPPER

    COMPOSITE

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    66/126

    Spacecraft Loads Analysis - A. Calvi 66

    Ariane-5 Dynamic Mathematical Model

    PAYLOAD

    EAP+EAP-

    EPC

    Dynamic effects up to about 100 Hz

    3D FE models of EPC, EAP, UC

    Dynamic Reduction using Craig-Bampton formulation

    Incompressible or compressible fluids models for liquid

    propellants

    Structure/fluid interaction Nearly incompressible SRB solid propellant modeling

    Pressure and stress effects on launcher stiffness

    SRB propellant and DIAS structural damping

    Non-linear launch table effects

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    67/126

    Spacecraft Loads Analysis - A. Calvi 67

    Sizing flight events (CLA with VEGA Launcher)Sizing flight events (CLA with VEGA Launcher)

    1. Lift-off (P80 Ignition and Blastwave)

    2. Mach1/QMAX Gust

    3. P80 Pressure Oscillations

    4. Z23 Ignition

    5. Z23 Pressure Oscillations

    6. Z9 Ignition

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    68/126

    Spacecraft Loads Analysis - A. Calvi 68

    CLA OutputCLA Output

    LV-SC interface accelerations

    Equivalent sine spectrum

    LV-SC interface forces

    Equivalent accelerations at CoG

    Internal responses Accelerations,

    Displacements

    Forces

    Stresses

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    69/126

    Spacecraft Loads Analysis - A. Calvi 69

    Payload / STS CLAPayload / STS CLA

    Lift-off Force Resultant in X [lbf]

    Lift-offMain Fitting

    I/F ForceX Dir. [N]

    Lift-offMain Fitting

    I/F Force

    Z Dir. [N]

    Lift-off

    Keel FittingI/F ForceY Dir. [N]

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    70/126

    Spacecraft Loads Analysis - A. Calvi 70

    Shock Response Spectrum and Equivalent Sine Input

    A shock response spectrum is a

    plot of maximum response (e.g.displacement, stress, acceleration)

    of single degree-of-freedom

    (SDOF) systems to a given input

    versus some system parameter,generally the undamped natural

    frequency.

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    71/126

    Spacecraft Loads Analysis - A. Calvi 71

    SRS/ESI of the following transient acceleration:

    Q

    SRS

    ESI

    ESI

    ESI

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    72/126

    Spacecraft Loads Analysis - A. Calvi 72

    ESI for Spacecraft

    CLA (Coupled Load Analysis)

    0 10 20 30 40 50 60 70 800

    50

    100

    150

    200

    250

    SRS[m

    /s2]

    frequency [Hz]

    SRS

    Q

    SRSESI

    ESI

    12

    Q

    SRSESI

    SRS

    Difference is negligible for small damping ratios

    2DOF

    SDOF

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    73/126

    Spacecraft Loads Analysis - A. Calvi 73

    /[ sm

    ][s

    2.41

    [Hz

    ]/[ sm2.46

    2DOF2DOF]/[ sm

    ][s

    01.0Hz23frequencynatural

    [Hz

    Hz23frequencynatural

    1.97 01.0]/[ sm

    Transient responseTransient response

    Frequency response at ESI levelFrequency response at ESI level

    SDOFSDOF

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    74/126

    Spacecraft Loads Analysis - A. Calvi 74

    5. Spacecraft testing5. Spacecraft testing

    5.1 Introduction and general aspects

    5.1 Mechanical tests

    Modal survey test

    Sinusoidal vibration testAcoustic noise test

    Shock test

    Random vibration test

    5.2 Over-testing and under-testing

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    75/126

    Spacecraft Loads Analysis - A. Calvi 75

    Testing techniquesTesting techniques Introduction (1)Introduction (1)

    Without testing, an analysis can give completely incorrect results

    Without the analysis, the tests can represent only a very limited reality

    Two types of tests according to the objectives to be reached:

    Simulation tests for structure qualification or acceptance

    Identification tests (a.k.a. analysis-validation tests) for structureidentification (the objective is to determine the dynamic characteristics ofthe tested structure in order to update the mathematical model)

    Note: identification and simulation tests are generally completely

    dissociated. In certain cases (e.g. spacecraft sine test) it is technicallypossible to perform them using the same test facility

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    76/126

    Spacecraft Loads Analysis - A. Calvi 76

    Testing techniquesTesting techniques Introduction (2)Introduction (2)

    Generation of mechanical environment

    Small shakers (with flexible rod; electrodynamic)

    Large shakers (generally used to impose motion at the base)

    Electrodynamic shaker

    Hydraulic jack shaker

    Shock machines (pyrotechnic generators and impact machines) Noise generators + reverberant acoustic chamber (homogeneous and

    diffuse field)

    Measurements

    Force sensors, calibrated strain gauges Accelerometers, velocity or displacement sensors

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    77/126

    Spacecraft Loads Analysis - A. Calvi 77

    Classes of tests used to verify requirements (purposes)Classes of tests used to verify requirements (purposes)

    Development test

    Demonstrate design concepts and acquire necessary information for

    design

    Qualification test

    Show a design is adequate by testing a single article

    Acceptance test

    Show a product is adequate (test each flight article)

    Analysis validation test

    Provide data which enable to confirm critical analyses or to change

    (update/validate) mathematical models and redo analyses

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    78/126

    Spacecraft Loads Analysis - A. Calvi 78

    Tests for verifying mechanical requirements (purposes)Tests for verifying mechanical requirements (purposes)

    Acoustic test

    Verify strength and structural life by introducing random vibration

    through acoustic pressure (vibrating air molecules) Note: acoustic tests at spacecraft level are used to verify adequacy of

    electrical connections and validate the random vibration environments

    used to qualify components

    (Pyrotechnic) shock test Verify resistance to high-frequency shock waves caused by separation

    explosives (introduction of high-energy vibration up to 10,000 Hz)

    System-level tests are used to verify levels used for component testing

    Random vibration test Verify strength and structural life by introducing random vibration

    through the mechanical interface (typically up to 2000 Hz )

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    79/126

    Spacecraft Loads Analysis - A. Calvi 79

    Tests for verifying mechanical requirements (purposes)Tests for verifying mechanical requirements (purposes)

    Sinusoidal vibration test

    Verify strength for structures that would not be adequately tested in

    random vibration or acoustic testing Note 1: cyclic loads at varying frequencies are applied to excite the

    structure modes of vibration

    Note 2: sinusoidal vibration testing at low levels are performed to verify

    natural frequencies Note 3: the acquired data can be used for further processing (e.g.

    experimental modal analysis)

    Note 4: this may seem like an environmental test, but it is not.

    Responses are monitored and input forces are reduced as necessary

    (notching) to make sure the target responses or member loads arenot exceeded.

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    80/126

    Spacecraft Loads Analysis - A. Calvi 80

    RockotRockot Dynamic SpecificationDynamic Specification

    Marketed by: Eurockot

    Actually flight qualified

    Manufactured by: Khrunichev

    Capability: 950 kg @ 500 km

    Launch site: Plesetsk

    Environment Level

    Sine vibration

    Longitudinal= 1 g on [5-10] Hz

    1.5 g at 20 Hz

    1 g on [40-100] Hz

    Lateral = 0.625 g on [5-100] Hz

    Acoustic

    31.5 Hz = 130.5 dB

    63 Hz = 133.5 dB

    125 Hz = 135.5 dB

    250 Hz = 135.7 dB

    500 Hz = 130.8 dB

    1000 Hz = 126.4 dB

    2000 Hz = 120.3 dB

    Shock

    100 Hz = 50 g

    700 Hz = 800 g

    1000 Hz 1500 Hz = 2000 g

    4000 Hz 5000 Hz = 4000 g

    10000 Hz = 2000 g

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    81/126

    Spacecraft Loads Analysis - A. Calvi 81

    5. Spacecraft testing5. Spacecraft testing

    5.1 Introduction and general aspects

    5.2 Mechanical tests

    Modal survey test

    Sinusoidal vibration testAcoustic noise test

    Shock test

    Random vibration test

    5.3 Over-testing and under-testing

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    82/126

    Spacecraft Loads Analysis - A. Calvi 82

    Modal survey test (identification test)Modal survey test (identification test)

    Purpose: provide data for dynamic mathematical model validation

    Note: the normal modes are the most appropriate dynamic

    characteristics for the identification of the structure

    Usually performed on structural models (SM or STM) in flight

    representative configurations

    Modal parameters (natural frequencies, mode shapes, damping,

    effective masses) can be determined in two ways:

    by a method with appropriation of modes, sometimes called phase

    resonance, which consists of successively isolating each mode by an

    appropriate excitation and measuring its parameters directly

    by a method without appropriation of modes, sometimes called phase

    separation, which consists of exciting a group of modes whoseparameters are then determined by processing the measurements

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    83/126

    Spacecraft Loads Analysis - A. Calvi 83

    Different ways to get modal data from testsDifferent ways to get modal data from tests

    Hammer test

    Vibration test data analysis

    Dedicated FRF measurement & modal analysis

    Full scale modal survey with mode tuningIncreasi

    ngeffort

    Dataconsistency

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    84/126

    Spacecraft Loads Analysis - A. Calvi 84

    Modal Survey Test vs. Modal Data extracted from the Sine Vibration Test

    Modal Survey:

    requires more effort (financial and time)

    provides results with higher quality

    Modal Data from Sine Vibration:

    easy access / no additional test necessary

    less quality due to negative effects from vibration

    fixtures / facility tables not indefinitely stiff

    higher sweep rate (brings along effects like beating or control instabilities)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    85/126

    Spacecraft Loads Analysis - A. Calvi 85

    Ariane 5Ariane 5 -- Sine excitation at spacecraft base (Sine excitation at spacecraft base (sine-equivalent dynamics)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    86/126

    Spacecraft Loads Analysis - A. Calvi 86

    Test Set-up for Satellite Vibration Tests

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    87/126

    Spacecraft Loads Analysis - A. Calvi 87

    Herschel on Hydra

    A ti t t ( bj ti )A ti t t ( bj ti )

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    88/126

    Spacecraft Loads Analysis - A. Calvi 88

    Acoustic test (objectives)Acoustic test (objectives)

    Demonstrate the ability of a specimen to

    withstand the acoustic environment duringlaunch

    Validation of analytical models

    System level tests verify equipment

    qualification loads

    Acceptance test for S/C flight models

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    89/126

    Spacecraft Loads Analysis - A. Calvi 89

    Ariane 5Ariane 5 Acoustic noise spectrum under the fairingAcoustic noise spectrum under the fairing

    Shock test Objecti es and remarksShock test Objecti es and remarks

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    90/126

    Spacecraft Loads Analysis - A. Calvi 90

    Shock test. Objectives and remarksShock test. Objectives and remarks

    Demonstrate the ability of a specimen to

    withstand the shock loads during launchand operation

    Verify equipment qualification loadsduring system level tests

    System level shock tests are generallyperformed with the actual shockgenerating equipment (e.g. clamp bandrelease)

    or by using of a sophisticated pyro-shock generating system (SHOGUN forARIANE 5 payloads)

    Shock machine (metalShock machine (metal metal pendulum impact machine)metal pendulum impact machine)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    91/126

    Spacecraft Loads Analysis - A. Calvi 91

    Shock machine (metalShock machine (metal--metal pendulum impact machine)metal pendulum impact machine)

    Random Vibration Test (vs Acoustic Test)Random Vibration Test (vs Acoustic Test)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    92/126

    Spacecraft Loads Analysis - A. Calvi 92

    Random Vibration Test (vs. Acoustic Test)Random Vibration Test (vs. Acoustic Test)

    Purpose: verify strength and structural life by introducing random

    vibration through the mechanical interface

    Random Vibration

    base driven excitation

    better suited for Subsystem / Equipment tests

    limited for large shaker systems

    Acoustic

    air pressure excitation

    better suited for S/C and large Subsystems with low mass / area

    density

    Random vibration test with slide tableRandom vibration test with slide table

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    93/126

    Spacecraft Loads Analysis - A. Calvi 93

    Random vibration test with slide tableRandom vibration test with slide table

    5 Spacecraft testing5 Spacecraft testing

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    94/126

    Spacecraft Loads Analysis - A. Calvi 94

    5. Spacecraft testing5. Spacecraft testing

    5.1 Introduction and general aspects

    5.2 Mechanical tests

    Modal survey test

    Sinusoidal vibration test

    Acoustic noise test

    Shock test

    Random vibration test

    5.3 Over-testing and under-testing

    Overtesting:Overtesting:

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    95/126

    Spacecraft Loads Analysis - A. Calvi 95

    Overtesting:Overtesting:an introductionan introduction(vibration absorber effect)(vibration absorber effect)

    Introduction to overtesting and notching

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    96/126

    Spacecraft Loads Analysis - A. Calvi 96

    Introduction to overtesting and notching

    The qualification of the satellite to lowfrequency transient is normallyachieved by a base-shake test

    The input spectrum specifies theacceleration input that should excitethe satellite, for each axis

    This input is definitively different fromthe mission loads, which are transient

    Notching: Reduction of acceleration

    input spectrum in narrow frequencybands, usually where test item hasresonances (NASA-HDBK-7004)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    97/126

    Spacecraft Loads Analysis - A. Calvi 97

    GOCE on ESTEC Large Slip Table Herschel on ESTEC Large Slip Table

    The overtesting problem (causes)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    98/126

    Spacecraft Loads Analysis - A. Calvi 98

    The overtesting problem (causes)

    Difference in boundary conditions between test and flight

    configurations

    during a vibration test, the structure is excited with a specified input

    acceleration that is the envelope of the flight interface acceleration,

    despite the amplitude at certain frequencies drops in the flight

    configuration (there is a feedback from the launcher to the spacecraft in

    the main modes of the spacecraft)

    The excitation during the flight is not a steady-state sine function andneither a sine sweep but a transient excitation with some cycles in a

    few significant resonance frequencies

    The objective of notching of the specified input levels is to take into

    account the real dynamic response for the different flight events. Inpractice the notching simulates the antiresonances in the coupled

    configuration

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    99/126

    Spacecraft Loads Analysis - A. Calvi 99

    Notching

    ESI(equivalent sine)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    100/126

    Spacecraft Loads Analysis - A. Calvi 100

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    101/126

    Spacecraft Loads Analysis - A. Calvi 101

    Random vibration test: notching of test specificationRandom vibration test: notching of test specification

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    102/126

    Spacecraft Loads Analysis - A. Calvi 102

    g pg p

    Illustration of notching of random vibration test specification,at the frequencies of strong test item resonances

    6. Mathematical models validation6. Mathematical models validation

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    103/126

    Spacecraft Loads Analysis - A. Calvi 103

    Validation of Finite Element Models(with emphasis on Structural Dynamics)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    104/126

    Spacecraft Loads Analysis - A. Calvi 104

    (with emphasis on Structural Dynamics)

    Everyone believes the test data except for the

    experimentalist, and no one believes the finite

    element model except for the analyst

    All models are wrong, but some are still useful

    Verification and Validation Definitions(ASME Standards Committee: V & V in Computational Solid Mechanics)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    105/126

    Spacecraft Loads Analysis - A. Calvi 105

    (ASME Standards Committee: V & V in Computational Solid Mechanics )

    Verification (of codes, calculations): Process of determining that a

    model implementation accurately represents the developers

    conceptual description of the model and the solution to the model Math issue: Solving the equations right

    Validation: Process of determining the degree to which a model is

    an accurate representation of the real world from the perspective ofthe intended uses of the model

    Physics issue: Solving the right equations

    Note: objective of the validation is to maximise confidence inthe predictive capability of the model

    Terminology: Correlation, Updating and Validation

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    106/126

    Spacecraft Loads Analysis - A. Calvi 106

    Correlation:

    the process of quantifying the degree of similarity and dissimilarity betweentwo models (e.g. FE analysis vs. test)

    Error Localization:

    the process of determining which areas of the model need to be modified

    Updating: mathematical model improvement using data obtained from an associated

    experimental model (it can be consistent or inconsistent)

    Valid model:

    model which predicts the required dynamic behaviour of the subjectstructure with an acceptable degree of accuracy, or correctness

    Some remarks on the validation of critical analyses

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    107/126

    Spacecraft Loads Analysis - A. Calvi 107

    Loads analysis is probably the single most influential task indesigning a space structure

    Loads analysis is doubly important because it is the basis for statictest loads as well as the basis for identifying the target responsesand notching criteria in sine tests

    A single mistake in the loads analysis can mean that we design andtest the structure to the wrong loads

    We must be very confident in our loads analysis, which means wemust check the sensitivity of our assumptions and validate theloads analysis that will be the basis of strength analysis and statictesting

    Note: Vibro-acoustic, random and shock analyses are usually notcritical in the sense that we normally use environmental tests toverify mechanical requirements

    Targets of the correlation(features of interest for quantitative comparison)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    108/126

    Spacecraft Loads Analysis - A. Calvi 108

    (features of interest for quantitative comparison)

    Characteristics that most affect the structure response to applied forces

    Natural frequencies

    Mode shapes

    Modal effective masses

    Modal damping

    Total mass, mass distribution

    Centre of Gravity, inertia

    Static stiffness Interface forces

    Correlation of mode shapesCorrelation of mode shapes

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    109/126

    Spacecraft Loads Analysis - A. Calvi 109

    Spacehab FEM coupled to the test rig model & Silhouette

    GOCE modal analysis and survey testGOCE modal analysis and survey test

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    110/126

    Spacecraft Loads Analysis - A. Calvi 110

    CrossCross--Orthogonality Check (COC) and Modal Assurance Criterion (MAC)Orthogonality Check (COC) and Modal Assurance Criterion (MAC)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    111/126

    Spacecraft Loads Analysis - A. Calvi 111

    The cross-orthogonality between the analysis and test mode

    shapes with respect to the mass matrix is given by:

    The MAC between a measured mode and an analytical mode is

    defined as:

    aTm MC

    as

    Tasmr

    Tmr

    asTmr

    rsMAC 2

    Note: COC and MAC do not give a useful measure of the error!

    Columbus: Cross-Orthogonality Check up to 35 Hz (target modes)TEST

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    112/126

    Spacecraft Loads Analysis - A. Calvi 112

    1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

    FEM Err.% [Hz] 13.78 15.80 17.20 23.81 24.23 24.65 25.36 25.59 26.59 27.19 27.53 28.87 30.19 30.55 32.73 33.15 33.86 34.57 35.21 36.16

    1 -2.94 13.37 1.00

    2 -0.95 15.65 1.00

    3 -1.73 16.90 0.99

    4 -3.26 23.03 0.93 0.35

    5 -1.16 23.95 0.34 0.93

    6 -2.00 24.16 0.95

    7 -1.98 24.86 0.95 0.27

    8 -0.12 25.56 0.86

    9 -0.95 26.34 0.22 0.90

    10 -2.65 26.47 0.95

    11 -0.40 27.42 0.26 0.96

    12 -3.65 27.82 0.82 0.27

    13 -6.00 28.38 0.46 0.89

    15 1.19 30.91 0.26 0.95

    17 1.63 33.26 0.94 0.21 0.32

    18 - 33.71 0.64 0.34 0.62

    19 -4.72 34.45 0.95

    20 1.21 34.99 0.57 0.81

    Soho SVM Cross-Orthogonality CheckF E M

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    113/126

    Spacecraft Loads Analysis - A. Calvi 113

    F.E.M.

    1 2 3 4 8 9 10 15 21 26 29 30 31

    TEST Err. % Freq. Hz 34.83 37.24 44.07 45.19 51.51 52.68 55.46 62.18 70.92 77.99 81.53 82.25 84.42

    1 2.87 35.86 0.87 0.46

    2 0.00 37.24 0.47 0.87

    3 4.17 45.99 0.87

    4 4.78 47.46 0.77 0.245 -3.39 49.82 -0.33 0.76

    6 0.96 53.19 0.75 0.22

    7 2.10 56.65 0.79 0.21

    8 58.67 -0.28 -0.35 -0.22

    9 60.24 -0.30 -0.22 0.46 0.46

    10 3.30 64.30 0.61

    11 66.40 0.21 0.32

    12 67.50 0.45 -0.43

    13 68.73-0.38

    14 69.68

    15 71.69

    16 72.71 0.37 - 0.33

    17 3.30 73.34 0.21 0.85

    18 74.78

    19 75.63

    20 78.77 -0.24

    21 82.12

    22 7.72 84.51 0.76

    23 5.54 86.31 0.87 -0.23

    24 7.21 88.64 0.64

    25 5.45 89.29 -0.33 0.63

    26 94.44

    27 97.15

    28 99.56

    GOCEGOCE -- MAC and Effective MassMAC and Effective Mass

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    114/126

    Spacecraft Loads Analysis - A. Calvi 114

    Lack of Matching between F.E. Model and Test

    (

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    115/126

    Spacecraft Loads Analysis - A. Calvi 115

    Modelling uncertainties and errors (model is not completely physicallyrepresentative)

    Approximation of boundary conditions

    Inadequate modelling of joints and couplings Lack or inappropriate damping representation

    The linear assumption of the model versus test non-linearities

    Mistakes (input errors, oversights, etc.)

    Scatter in manufacturing Uncertainties in physical properties (geometry, tolerances, material properties)

    Uncertainties and errors in testing

    Measured data or parameters contain levels of errors

    Uncertainties in the test set-up, input loads, boundary conditions etc.

    Mistakes (oversights, cabling errors, etc.)

    Test-Analysis Correlation CriteriaThe degree of similarity or dissimilarity establishing that the correlationbetween measured and predicted values is acceptable

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    116/126

    Spacecraft Loads Analysis - A. Calvi 116

    between measured and predicted values is acceptable

    ECSS-E-ST-32-11 Proposed Test / Analysis Correlation Criteria

    7. Conclusions7. Conclusions

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    117/126

    Spacecraft Loads Analysis - A. Calvi 117

    Spacecraft Loads Analysis - Contents

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    118/126

    Spacecraft Loads Analysis - A. Calvi 118

    1. Introduction and general aspects

    2. Mechanical environment

    3. Requirements for spacecraft structures

    4. Mathematical models and structural analyses

    5. Spacecraft mechanical testing

    6. Mathematical models validation

    7. Conclusions

    Final VerificationFinal Verification

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    119/126

    Spacecraft Loads Analysis - A. Calvi 119

    Consist of:

    Making sure all requirements are satisfied (compliance)

    Validating the methods and assumptions used to satisfy

    requirements

    Assessing risks

    Criteria for Assessing Verification Loads (strength)Criteria for Assessing Verification Loads (strength)

    Analysis: margins of safety must me greater that or equal to zero

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    120/126

    Spacecraft Loads Analysis - A. Calvi 120

    Analysis: margins of safety must me greater that or equal to zero

    Test: Structures qualified by static or sinusoidal testing Test loads or stresses as predicted (test-verified math model and test

    conditions) are compared with the total predicted loads during the

    mission (including flying transients, acoustics, random vibration,

    pressure, thermal effects and preloads)

    Test: Structures qualified by acoustic or random vibration testing

    Test environments are compared with random-vibration environments

    derived from system-level acoustic testing

    Final Verification (crucial points)Final Verification (crucial points)

    To perform a Verification Loads Cycle for structures designed

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    121/126

    Spacecraft Loads Analysis - A. Calvi 121

    To perform a Verification Loads Cycle for structures designed

    and tested to predicted loads

    Finite element models correlation with the results of modal andstatic testing

    Loads prediction with the current forcing functions

    Compliance with analysis criteria (e.g. MOS>0)

    To make sure the random-vibration environments used to

    qualify components were high enough (based on data

    collected during the spacecraft acoustic test)

    Note: in the verification loads cycle instead of identifying required

    design changes (design loads cycle) the adequacy of the structure

    that has already been built and tested is assessed

    Technical inconsistenciesTechnical inconsistencies

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    122/126

    Spacecraft Loads Analysis - A. Calvi 122

    Uni-axial vibration and shock test facilities

    Low frequency transient often simulated at the subsystem andsystem assembly level using a swept-sine vibration test

    Infinite mechanical impedance of the shaker (and the standard

    practice of controlling the input acceleration to the frequency

    envelope of the flight data) Vibro-acoustic environment often simulated at the subsystem

    and units assembly level using a random vibration test

    Test levels largely based on computational analyses (we must

    validate critical load analyses!)

    NEW TRENDSNEW TRENDS

    Protoflight vs prototype

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    123/126

    Spacecraft Loads Analysis - A. Calvi 123

    Protoflight vs prototype

    Testing (force limited vibration testing)

    Computational mechanics

    Vibro-acoustic analysis

    Random vibration analysis

    Complexity of the Industrial organization

    BibliographyBibliography

    Sarafin T.P. Spacecraft Structures and Mechanisms, Kluwer, 1995

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    124/126

    Spacecraft Loads Analysis - A. Calvi 124

    p , ,

    Craig R.R., Structural Dynamics An introduction to computer methods, J. Wiley

    and Sons, 1981

    Clough R.W., Penzien J., Dynamics of Structures, McGraw-Hill, 1993

    Ewins D.J., Modal Testing Theory, practice and applications, Research Studies

    Press, Second Edition, 2000

    Wijker J., Mechanical Vibrations in Spacecraft Design, Springer, 2004

    Girard A., Roy N., Structural Dynamics in Industry, J. Wiley and Sons, 2008

    Steinberg D.S., Vibration Analysis for Electronic Equipment, J. Wiley and Sons,

    2000

    Friswell M.I., Mottershead J.E., Finite Element Model Updating in Structural

    Dynamics, Kluwer 1995 Ariane 5 Users Manual, Arianespace, http://www.arianespace.com/

    BibliographyBibliography -- ECSS DocumentsECSS Documents

    ECSS-E-HB-32-26 Spacecraft Loads Analysis (to be published)

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    125/126

    Spacecraft Loads Analysis - A. Calvi 125

    p y ( p )

    ECSS-E-ST-32 Space Project Engineering - Structural

    ECSS-E-ST-32-03 Structural finite element models ECSS-E-ST-32-10 Structural factors of safety for spaceflight

    hardware

    ECSS-E-ST-32-02 Structural design and verification of

    pressurized hardware ECSS-E-ST-32-11 Modal survey assessment

    ECSS-E-ST-32-01 Fracture control

    ECSS-E-10-02 Space Engineering - Verification

    ECSS-E-10-03 Space Engineering - Testing

    THE END!

  • 7/30/2019 Calvi Students Liege 2011 6(1)

    126/126

    Spacecraft Loads Analysis - A. Calvi 126

    THE END!

    Acknowledgements:

    TAS France for the data concerning the project SENTINEL-3

    ALENIA SPAZIO, Italy, for the data concerning the projects GOCE, COLUMBUS,MPLM and SOHO

    EADS ASTRIUM, UK, for the data concerning the project AEOLUS and

    EarthCARE

    ESA/ESTEC, Structures Section, NL, for the data concerning ARIANE 5 FE

    model and LV/SC CLA