CALVI Students LIEGE 2011 6 PDF

Spacecraft Loads Analysis
An Overview
Adriano Calvi, PhD

ESA / ESTEC, Noordwijk, The Netherlands
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 Analysis - A. Calvi
Spacecraft Loads Analysis - Contents

1.
2.
3.
4.
5.
6.
7.
Introduction and general aspects

Mechanical environment
Requirements for spacecraft structures
Mathematical models and structural analyses
Spacecraft mechanical testing
Mathematical models validation
Conclusions
Example of satellite structural design concept
What 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 for
designed or built hardware.
Spacecraft loads analysis process disciplines

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
It is 3 years that I work in this

company. Now, finally I have
understood what I do, but
still I have to understand why.
Organizations and Levels of Assembly an example

Launcher Authority
Spacecraft + launcher
Spacecraft Authority
Spacecraft
Spacecraft Prime Contractor
Spacecraft
Payload Contractor
Instruments/sub-systems
Other Contactors
Units/components/parts
Levels of Assembly
RPM
RFFE
Design 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 significant
ground, 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
Design loads cycle process
Loads and Factors

ECSS E-ST-32-10
Satellites
Test Logic
Common Design Logic
Expendable launch vehicles,

pressurized hardware and
manned system Test Logic
Limit Loads - LL
Increasing Load Level
x KQ
QL
x KA
x Coef. A
AL
Design Limit Loads
DLL
x Coef. B
x Coef. C
x KQ
x KA
AL
DYL
DUL
QL
Protoflight
Prototype
10
2. Mechanical environment
11
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
12
Accelerations some remarks

The parameter most commonly used (in the industry) to define the
motion of a mechanical system is the acceleration
Good reasons: accelerations are directly related to forces/stresses and
easy to specify and measure
In practice accelerations are used as a measure of the severity of the
mechanical 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 they
are 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)
13
Launch 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
14
Steady-state and low-frequency transient accelerations
15
Acoustic Loads
During the lift off and the early phases of
the 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
16
Broadband and high frequency vibrations

Broad band random vibrations are produce by:
Engines functioning
Structural response to broad-band acoustic loads
Aerodynamic turbulent boundary layer
17
Shocks
Mainly caused by the actuation of pyrotechnic devices:
Release mechanisms for stage and satellite separation
Deployable mechanisms for solar arrays etc.
18
Shocks
[g]
[g]
Time [t]
Frequency [Hz]
19
Static and dynamic environment specification (typical ranges)
20
Static and dynamic environment specification (typical ranges)
21
Quasi-Static Loads (accelerations)

Loads independent of time or which vary slowly, so that the dynamic
response 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 RP1403, NASA-HDBK-7004). Note: this definition is fully adequate for the
design 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-state
accelerations (unchanging applied force balanced by inertia loads). For
design purposes (e.g. derivation of design limit loads, selection of the
fasteners, etc.), the quasi-static loads are normally calculated by
combining both static and dynamic load contributions. In this context the
quasi static loads are equivalent to (or interpreted by the designer as)
static loads, typically expressed as equivalent accelerations at the C.o.G.
22
3. Requirements for spacecraft structures
23
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 and

spacecraft subsystems with enough strength and stiffness to
preclude any failure (rupture, collapse, or detrimental deformation)
that may keep them from working successfully.
24
Requirements evolution
25
Some definitions
Design:
The process used to generate the set information describing the
essential characteristics of a product (ECSS-P-001A)
Design means developing requirements, identifying options, doing
analyses and trade studies, and defining a product in enough detail so
it can be built (T. P. Sarafin)
Verification:
Confirmation by examination and provision of objective evidence that
specified requirements have been fulfilled (ISO 8402:1994)
Verification means providing confidence through disciplined steps that
a product will do what it is supposed to do (T. P. Sarafin)
Note: we can prove that the spacecraft satisfies the measurable criteria
we have defined, but we cannot prove a space mission will be successful
26
Design requirements and verification
27
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 or
exhibiting 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.
28
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 be
delivered 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.
29
Spacecraft 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 between
low frequency dynamics of the launch vehicle and payload modes
30
4. Mathematical models and structural analyses
4.1 Dynamic analysis types - Overview

4.2 Effective mass concept
4.3 Launcher/Spacecraft coupled loads analysis
31
Dynamic analysis types

Real eigenvalue analysis (undamped free vibrations)
Modal parameter identification, etc.
Linear frequency response analysis (steady-state response of

linear 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
32
Reasons to compute normal modes (real eigenvalue analysis)

To verify stiffness requirements
To assess the dynamic interaction between a component and its
supporting 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,
33
Real eigenvalue analysis
Note: mode shape normalization

Scaling is arbitrary
Convention: Mass, Max or Point
34
Mode shapes
Cantilever beam
Simply supported beam

35
Satellite Normal Modes Analysis
Mode 1: 16.2 Hz
Mode 2: 18.3 Hz
INTEGRAL Satellite (FEM size 120000 DOFs)

36
Frequency 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:
37
Frequency response considerations

If the maximum excitation frequency is much less than the lowest
resonant 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
38
Harmonic forced response with damping
39
Transient Response Analysis

Purpose is to compute the behaviour of a structure subjected to timevarying 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, and
forces 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:
40
Modal Transient Response Analysis
41
Transient response considerations

The integration time step must be small enough to represent accurately
the 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
42
Shock response spectrum (and analysis)

Response spectrum analysis is an approximate method of
computing 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 response
such as stress analysis
Note 1: the part (2) of the response spectrum analysis has a
limited use in structural dynamics of spacecraft (e.g. preliminary
design) since the accuracy of the method may be questionable
Note 2: the term shock can be misleading (not always a physical
shock, i.e. an environment of a short duration, is involved. It
would be better to use response spectrum)
43
Generation of a response spectrum (1)
44
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 is
computed
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
45
Shock Response Spectrum. Some remarks

The 1-DOF system is used as reference structure (since the
simplest) for the characterization of environments (i.e.
quantification of the severity equivalent environments can be
specified)
In practice, the criterion used for the severity is the maximum
response which occurs on the structure (note: another criterion
relates to the concept of fatigue damage)
A risk in comparing two excitations of different nature is in the
influence of damping on the results (e.g. maxima are proportional
to 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)
46
Shock Response Spectrum
(A) is the shock spectrum

of a terminal peak
sawtooth (B) of 500 G
peak amplitude and 0.4
millisecond duration
47
Random vibration (analysis)

Random vibration is vibration that can be described only in a
statistical sense
The instantaneous magnitude is not known at any given time;
rather, the magnitude is expressed in terms of its statistical
properties (such as mean value, standard deviation, and probability
of exceeding a certain value)
Examples of random vibration include earthquake ground motion,
wind pressure fluctuations on aircraft, and acoustic excitation due
to rocket and jet engine noise
These random excitations are usually described in terms of a
power spectral density (PSD) function
Note: in structural dynamics of spacecraft, the random vibration
analysis is often performed with simplified techniques (e.g. based
on Miles equation + effective modal mass models)
48
Random noise with normal amplitude distribution
49
Power Spectral Density (conceptual model)
50
Sound Pressure Level (conceptual model)
51
Vibro acoustic 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
52

53
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, in
terms of the load it will cause in the structure, when excited
Note: avoid using: it is the mass which participates to the mode!
Modal reaction forces
Dynamic amplification factor

Base (junction) excitation

54

Effective mass
Gen. mass
Resultant of modal interface forces

Modal participation factors
i-th mode
Rigid body modes

Eigenvector max value
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 percentage
of 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
55

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)
56
Example of Effective Mass table

(MPLM test and FE model)
57

58
A5 Typical Sequence of events
59
A5 Typical Longitudinal Static Acceleration
60
Sources of Structural Loadings (Launch)

80
70
60
acceleration [m/s2]
50
40
30
20
10
0
-10
-20
0
0.1
0.2
0.3
0.4
0.5
t [s]
0.6
0.7
0.8
0.9
4
3
2
1
0
50
100
150
200
250
300
t, s
Axial-Acceleration Profile for the Rockot Launch Vehicle

61
80
70
60
acceleration [m/s2]
50
40
30
20
10
0
-10
-20
0
0.1
0.2
0.3
0.4
0.5
t [s]
0.6
0.7
0.8
0.9
Axial Acceleration at Launcher/Satellite Interface (Engines Cut-off)

62
Load Factors for Preliminary Design (Ariane 5)

63
Quasi-Static Flight Limit loads for Dnepr and Soyuz
64
Launcher / Satellite C.L.A.
A5 / Satellite Recovered System Mode shapes
CLA: simulation of the structural response to

low frequency mechanical environment
Main Objective: to calculate the loads on the
satellite caused by the launch transients (liftoff, transonic, aerodynamic gust, separation of
SRBs)
Loads (in this context): set of internal forces,
displacements and accelerations that
characterise structural response to the applied
forces
Effects included in the forcing functions :
thrust built-up, engine shut-down/burnout,
gravity, aerodynamic loads (gust), separation
of boosters, etc.
Mode 18: 2.93 Hz
Mode 53: 16.9 Hz

65
UPPER
COMPOSITE
Ariane-5 Dynamic Mathematical Model

PAYLOAD
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
EAP-
EAP+
EPC
66
Sizing flight events (CLA with VEGA Launcher)
6. Z9 Ignition
5. Z23 Pressure Oscillations

4. Z23 Ignition
3. P80 Pressure Oscillations

2. Mach1/QMAX Gust
1. Lift-off (P80 Ignition and Blastwave)
67
CLA Output
LV-SC interface accelerations
Equivalent sine spectrum
LV-SC interface forces

Equivalent accelerations at CoG
Internal responses
Accelerations,
Displacements
Forces
Stresses
68
Payload / STS CLA
Lift-off
Main Fitting
I/F Force
X Dir. [N]
Lift-off
Main Fitting
I/F Force
Z Dir. [N]
Lift-off
Keel Fitting
I/F Force
Y Dir. [N]
Lift-off Force Resultant in X [lbf]
69
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.
70
SRS/ESI of the following transient acceleration:
ESI
SRS
Q
ESI
ESI
71
ESI for Spacecraft

CLA (Coupled Load Analysis)
Difference is negligible for small damping ratios

250
SRS
ESI
SRS [m/s2]
200
SRS
Q
ESI
150
100
SRS
Q2 1
SRS
50
ESI
0
0
10
20
30
40
50
frequency [Hz]
60
70
80
72
[m / s ]
SDOF
0.01
natural frequency 23 Hz
[m / s
2.41
Transient response
2DOF
Transient response
[s ]
[s ]
[m / s ]
1.97
0.01
natural frequency 23 Hz
Frequency response at ESI level
[m / s ]
2.46
Frequency response at ESI level
73
[Hz
[Hz
5. Spacecraft testing
5.1
5.1
Mechanical tests
Modal survey test
Sinusoidal vibration test
Acoustic noise test
Shock test
Random vibration test
5.2
Over-testing and under-testing
74
Testing techniques 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 structure
identification (the objective is to determine the dynamic characteristics of
the 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 technically
possible to perform them using the same test facility
75
Testing techniques 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
76
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
77
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

Verify strength and structural life by introducing random vibration
through the mechanical interface (typically up to 2000 Hz )
78
Tests for verifying mechanical requirements (purposes)

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 are
not exceeded.
79
Marketed by: Eurockot
Manufactured by: Khrunichev
Actually flight qualified
Capability: 950 kg @ 500 km
Environment
Sine vibration
Acoustic
Shock
Launch site: Plesetsk
Level
Longitudinal= 1 g on [5-10] Hz
Lateral = 0.625 g on [5-100] Hz
1.5 g at 20 Hz
1 g on [40-100] Hz
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
100 Hz = 50 g
700 Hz = 800 g
1000 Hz 1500 Hz = 2000 g
4000 Hz 5000 Hz = 4000 g

10000 Hz = 2000 g
Rockot Dynamic Specification

80
5.1
5.2
Mechanical tests
Modal survey test
Acoustic noise test
Shock test
5.3
81
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 whose
parameters are then determined by processing the measurements
82
Data consistency
Increasing effort
Different ways to get modal data from tests
Hammer test
Vibration test data analysis
Dedicated FRF measurement & modal analysis
Full scale modal survey with mode tuning
83
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)
84
Ariane 5 - Sine excitation at spacecraft base (sine-equivalent

dynamics)
(
85
Test Set-up for Satellite Vibration Tests
86
Herschel on Hydra
87
Acoustic test (objectives)

Demonstrate the ability of a specimen to
withstand the acoustic environment during
launch
Validation of analytical models
System level tests verify equipment
qualification loads
Acceptance test for S/C flight models
88
Ariane 5 Acoustic noise spectrum under the fairing

89
Shock test. Objectives and remarks

Demonstrate the ability of a specimen to
withstand the shock loads during launch
and operation
Verify equipment qualification loads
during system level tests
System level shock tests are generally
performed with the actual shock
generating equipment (e.g. clamp band
release)
or by using of a sophisticated pyroshock generating system (SHOGUN for
ARIANE 5 payloads)
90
Shock machine (metal-metal pendulum impact machine)
91
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
92
Random vibration test with slide table
93
5.1
5.2
Mechanical tests
Modal survey test
Acoustic noise test
Shock test
5.3
94
Overtesting:
an introduction
(vibration absorber effect)
95
Introduction to overtesting and notching

The qualification of the satellite to low
frequency transient is normally
achieved by a base-shake test
The input spectrum specifies the
acceleration input that should excite
the satellite, for each axis
This input is definitively different from
the mission loads, which are transient
Notching: Reduction of acceleration
input spectrum in narrow frequency
bands, usually where test item has
resonances (NASA-HDBK-7004)
96
GOCE on ESTEC Large Slip Table

Herschel on ESTEC Large Slip Table

97
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 and
neither 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. In
practice the notching simulates the antiresonances in the coupled
configuration
98
Notching
ESI
(equivalent sine)
99
100
101
Random vibration test: notching of test specification
Illustration of notching of random vibration test specification,

at the frequencies of strong test item resonances
102
6. Mathematical models validation
103
Validation of Finite Element Models

(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
104
Verification and Validation Definitions

(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 of
the intended uses of the model
Physics issue: Solving the right equations
Note: objective of the validation is to maximise confidence in

the predictive capability of the model
105
Terminology: Correlation, Updating and Validation

Correlation:
the process of quantifying the degree of similarity and dissimilarity between
two 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 subject
structure with an acceptable degree of accuracy, or correctness
106
Some remarks on the validation of critical analyses

Loads analysis is probably the single most influential task in
designing a space structure
Loads analysis is doubly important because it is the basis for static
test loads as well as the basis for identifying the target responses
and notching criteria in sine tests
A single mistake in the loads analysis can mean that we design and
test the structure to the wrong loads
We must be very confident in our loads analysis, which means we
must check the sensitivity of our assumptions and validate the
loads analysis that will be the basis of strength analysis and static
testing
Note: Vibro-acoustic, random and shock analyses are usually not
critical in the sense that we normally use environmental tests to
verify mechanical requirements
107
Targets of the correlation

(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
108
Correlation of mode shapes
Spacehab FEM coupled to the test rig model & Silhouette

109
GOCE modal analysis and survey test
110
Cross-Orthogonality Check (COC) and Modal Assurance Criterion (MAC)

The cross-orthogonality between the analysis and test mode
shapes with respect to the mass matrix is given by:
T
mM
The MAC between a measured mode and an analytical mode is

defined as:
2
T
mr as
T
T
mr mr as as
MACrs

Note: COC and MAC do not give a useful measure of the error!
111
Columbus: Cross-Orthogonality Check up to 35 Hz (target modes)

TEST
1
FEM Err.% [Hz]
10
11
12
13
-2.94 13.37
-0.95 15.65
-1.73 16.90
-3.26 23.03
0.93
0.35
-1.16 23.95
0.34
0.93
-2.00 24.16
-1.98 24.86
-0.12 25.56
0.86
-0.95 26.34
0.22
10
-2.65 26.47
11
-0.40 27.42
12
-3.65 27.82
0.82
0.27
13
-6.00 28.38
0.46
0.89
15
1.19 30.91
17
1.63 33.26
-
-4.72 34.45
20
1.21 34.99
16
17
18
19
20
1.00
1.00
0.99
33.71
19
15
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
18
14
0.95
0.95
0.27
0.90
0.95
0.26
0.96
0.26
0.95
0.94
0.21
0.32
0.64
0.34
0.62
0.95
0.57
0.81
112
Soho SVM Cross-Orthogonality Check

TEST
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Err. % Freq. Hz
35.86
2.87
37.24
0.00
45.99
4.17
4.78
47.46
-3.39
49.82
0.96
53.19
56.65
2.10
58.67
60.24
3.30
64.30
66.40
67.50
68.73
69.68
71.69
72.71
3.30
73.34
74.78
75.63
78.77
82.12
7.72
84.51
5.54
86.31
88.64
7.21
89.29
5.45
94.44
97.15
99.56
F.E.M.
1
2
34.83 37.24
0.87
0.46
0.87
0.47
3
44.07
4
45.19
8
51.51
0.77
-0.33
0.24
0.76
9
52.68
10
55.46
15
62.18
21
70.92
26
77.99
29
81.53
30
82.25
31
84.42
0.87
0.75
0.22
0.79
0.21
-0.35
0.46
-0.28
-0.30
0.21
0.45
-0.22
-0.22
0.46
0.61
0.32
-0.43
-0.38
0.37
0.21
-0.33
0.85
-0.24
0.76
0.87
-0.23
0.64
-0.33
0.63
113
GOCE - MAC and Effective Mass
114
Lack of Matching between F.E. Model and Test

Modelling uncertainties and errors (model is not completely physically
representative)
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.)
115
Test-Analysis Correlation Criteria

The degree of similarity or dissimilarity establishing that the correlation
between measured and predicted values is acceptable
ECSS-E-ST-32-11 Proposed Test / Analysis Correlation Criteria

116
7. Conclusions
117
Spacecraft Loads Analysis - Contents

1.
2.
3.
4.
5.
6.
7.

Mechanical environment
Requirements for spacecraft structures
Mathematical models and structural analyses
Spacecraft mechanical testing
Mathematical models validation
Conclusions
118
Final Verification
Consist of:
Making sure all requirements are satisfied (compliance)
Validating the methods and assumptions used to satisfy
requirements
Assessing risks
119
Criteria for Assessing Verification Loads (strength)

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
120
Final Verification (crucial points)

To perform a Verification Loads Cycle for structures designed
and tested to predicted loads
Finite element models correlation with the results of modal and
static 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
121
Technical inconsistencies
Uni-axial vibration and shock test facilities
Low frequency transient often simulated at the subsystem and
system 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!)
122
NEW TRENDS
Protoflight vs prototype
Testing (force limited vibration testing)
Computational mechanics
Vibro-acoustic analysis
Random vibration analysis
Complexity of the Industrial organization
123
Bibliography
Sarafin T.P. Spacecraft Structures and Mechanisms, Kluwer, 1995
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/
124
Bibliography - ECSS Documents
ECSS-E-HB-32-26
ECSS-E-ST-32
ECSS-E-ST-32-03
ECSS-E-ST-32-10
ECSS-E-ST-32-11
ECSS-E-ST-32-01
Spacecraft Loads Analysis (to be published)

Space Project Engineering - Structural
Structural finite element models
Structural factors of safety for spaceflight
hardware
Structural design and verification of
pressurized hardware
Modal survey assessment
Fracture control
ECSS-E-10-02
ECSS-E-10-03
Space Engineering - Verification

Space Engineering - Testing
ECSS-E-ST-32-02
125
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
126

CALVI Students LIEGE 2011 6 PDF

Uploaded by

Copyright:

Available Formats

You might also like

CALVI Students LIEGE 2011 6 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CALVI Students LIEGE 2011 6 PDF

Uploaded by

Copyright:

Available Formats

Spacecraft Loads Analysis

Adriano Calvi, PhD

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

Spacecraft Loads Analysis - A. Calvi

Spacecraft Loads Analysis - Contents

Introduction and general aspects

Spacecraft Loads Analysis - A. Calvi

Example of satellite structural design concept

Spacecraft Loads Analysis - A. Calvi

What is Loads Analysis

The goal or purpose of loads analysis

Spacecraft Loads Analysis - A. Calvi

Spacecraft loads analysis process disciplines

Spacecraft Loads Analysis - A. Calvi

It is 3 years that I work in this

Organizations and Levels of Assembly an example

Spacecraft Prime Contractor

Spacecraft Loads Analysis - A. Calvi

Spacecraft Loads Analysis - A. Calvi

Design Loads Cycles

Spacecraft Loads Analysis - A. Calvi

Design loads cycle process

Spacecraft Loads Analysis - A. Calvi

Loads and Factors

Common Design Logic

Expendable launch vehicles,

Increasing Load Level

Spacecraft Loads Analysis - A. Calvi

Spacecraft Loads Analysis - A. Calvi

Mechanical loads are caused by:

Spacecraft Loads Analysis - A. Calvi

Accelerations some remarks

Launch mechanical environment

Spacecraft Loads Analysis - A. Calvi

Steady-state and low-frequency transient accelerations

Spacecraft Loads Analysis - A. Calvi

Acoustic noise (as pressure waves)

Spacecraft Loads Analysis - A. Calvi

Broadband and high frequency vibrations

Spacecraft Loads Analysis - A. Calvi

Spacecraft Loads Analysis - A. Calvi

Spacecraft Loads Analysis - A. Calvi

Static and dynamic environment specification (typical ranges)

Spacecraft Loads Analysis - A. Calvi

Static and dynamic environment specification (typical ranges)

Spacecraft Loads Analysis - A. Calvi

Quasi-Static Loads (accelerations)

3. Requirements for spacecraft structures

Spacecraft Loads Analysis - A. Calvi

Typical Requirements for Spacecraft Structures

Basic requirement: the structure shall support the payload and

Spacecraft Loads Analysis - A. Calvi

Spacecraft Loads Analysis - A. Calvi

Design requirements and verification

Spacecraft Loads Analysis - A. Calvi

Examples of (Mechanical) Requirements (1)

Spacecraft Loads Analysis - A. Calvi

Examples of (Mechanical) Requirements (2)