Page 1

MDI France
58 Rue Pottier
F - 78150 LE CHESNAY
EMail : info@mdi.fr
Tel. : <33> 01 39 66 04 00
Fax : <33> 01 39 66 94 74
ARIANE 4
PAYLOAD FAIRING
SEPARATION ASSESMENT
R. LAGIER (Arianespace)
T. BERNARD (MDI France)
September 2000
ARIANE 4 payload fairing separation
Page 2
Summary
I Introduction / objectives
I Separation system description
I Available measurements
I Simulation methodology
I Nastran model correlation
I Linear approximation effects
I Adams non-linear correction
I Pyrotechnics model
I Adams model correlation
I Simulation of flight 126
I Conclusion / perspectives
Introduction / objectives
I Background of existing non-linear FEA based models :
N complex modeling : many assumptions on component physics
N CPU expensive : limited use for sensitivity studies
I Setup an efficient simulation methodology for ARIANE
4 payload fairing separation :
N ease of modeling to take in account all components physics
N extended parametrization (components physics, system configurations)
N CPU costless to allow extensive use of DOE / optimization
N good correlation / predictivity
I Quite short delay starting from scratch in May 2000 : 1
month for modeling, 1 month for investigations
Page 3
Separation system description ()
I VEB : vehicle equipements bay
I 2 half fairing structures (T & Tn)
I HSS : horizontal separation
system
I VSS : vertical separation
system
I Separation sequence :
N 2G constant acceleration of VEB
N ignition of HSS to untie fairing from
VEB / payload assembly
N ignition of VSS to separate the 2
half fairing =>focus of the
assesment
Roll axis
Yaw axis
Pitch axis
T
Sn
Tn
S
VEB
VSS
HSS
Tn half
fairing
T half
fairing
a
a'
Separation system description
I VSS pyrotechnic system :
N ignition : cord explosion (1)
N vessel expanding in cavity (2)
N pressure ramping
N shear loading of rivets untill
resistance limit (3)
N separation pulse
(1) (2)
(3)
(3)
Tn T
aa cross section
Page 4
Available measurements
I Ground measurements :
N mass, inertia, CoG position
N free-free frequencies
N deflection under opening / closing
static loads
N separation sensitivity (100, 150,
200 % rivets density)
I Flight measurements :
N radial displacement
N pitch, roll angular velocities
Simulation methodology
I Build Nastran models of 2 half
fairings :
N correlate mass report
N correlate free-free frequencies
N correlate static deflection
I Assemble Adams model of
fairing together with VSS rivets
and pyrotechnic cord :
N check (and correct if possible)
linear approximation effects
N identify pyrotechnic pulse
parameters to correlate separation
sensitivity to rivets density
N correlate flight measurements
Build
NASTRAN
T & Tn models
Assemble
ADAMS
model
(T+Tn+VSS)
mass report
+ free-free freq.
separation
sensitivity to
rivets density
check / correct
linear approx.
Flight prediction
structure
parameters
parameters
pyrotechnic
Page 5
Nastran model correlation
I Use Adams DOE for Nastran
model parameters tuning to
correlate measured mass and
frequencies :
N mass, inertia and CoG position
correlated within 0.1 %
N first 6 free-free frequencies
correlated within 5 %
I Deflection under static loads
showed unpredicted non-
linearity (composite material
behavior ?) :
N Nastran non-linear -380 / +380
N measurements -395 / +300
2.39 Hz
6.02 Hz
13.1 Hz
Linear approximation effects
I Estimate the effects of linear approximation by comparing Adams
beams w.r.t. Nastran equivalent flexible body :
Page 6
Adams non-linear correction
I Use new Adams modal loads to add non-linear correction :
ML(7) =IF( Q(7) : 0 , 0 , - DK * Q(7) )
Pyrotechnics model
I Function-of-time-only pyro force
is not sufficient to reproduce
separation sensitivity to rivets
density
I How to guess a realistic
parametric pyro law ?
Pressure
rampup
law
Expansion
law
F(t,d)
Fmax(d)
Coef *
Fmax(d)
Fmax(d)
Tf
Df
time
distance
Fmax(0)
Page 7
Adams model correlation
I Only the first 3 symmetric
modes are significantly involved
in separation dynamics ...
I Use Adams sensitivity study for
pyro parameters tuning to
correlate separation sensitivity
to rivets density
Simulation of flight 126 ()
I Flight 126 - Ariane 4 on J anuary
2000 : successfull injection of
GALAXY 10-R satellite
I Use of Adams optimization to
guess pyrotechnic levels (S / Sn
balance) out of flight 126
measurements
Page 8
Simulation of flight 126
Conclusion / perspectives ()
I Adams with modal flexibility was the good choice to setup
an efficient simulation methodology for fairing separation
studies :
N sensitivity / DOE / optimization to correlate parametric components
N powerfull environment to take in account any component physics
N modal loads to add non-linear correction
N 1 mn CPU / simulation
I ARIANE 4 payload fairing separation :
N better understanding of rivet / pyrotechnic coupling
N limited number of significant modes
N collision clearance (safety margin) prediction
Page 9
Conclusion / perspectives
I Payload fairing separation :
N validated tool and methodology to specify components tolerance
requirements (VSS S / Sn balance)
N easy extend to other configurations (ARIANE 5, etc ...)
N ability to investigate non conformances consequences
I Use of Adams modal flexibility :
N a successfull step in the domain of non-linear structures behavior