Vol. 4 N.1 - Journal of Aerospace Technology and Management by Journal of Aerospace Technology and Management

General Information JOURNAL OF AEROSPACE TECHNOLOGY AND MANAGEMENT (JATM) is a techno-scientific publication serialized, published by Departamento de Ciência e TecnologiaAeroespacial (DCTA) and aims to serve the international aerospace community. It contains articles that have been selected by an Editorial Committee composed of researchers and technologists from the scientific community. The journal is quarterly published, and its main objective is to provide an archival form of presenting scientific and technological research results related to the aerospace field, as well as promote an additional source of diffusion and interaction, providing public access to all of its contents, following the principle of making free access to research and generate a greater global exchange of knowledge. JATM is added/indexed in the following databases: SCOPUS-Elsevier; CAS – Chemical Abstracts Service; DOAJ – Directory of Open Access Journals; J-GATE – The e-journal gateway from global literature; LIVRE – Portal to Free Access Journals; GOOGLE SCHOLAR; SUMÁRIOS.ORG – Summaries of Brazilian Journals; EZB – Electronic Journals Library; ULRICHSWEB – Ulrich´s Periodicals Directory and PERIÓDICOS CAPES. In WEB QUALIS System, JATM is classified as B4 in the Geosciences and Engineering III areas. JATM is affiliated to ABEC – Brazilian Association of Scientific Editors and all published articles contain DOI numbers attributed by CROSSREF.

JATM is supported by:

Journal of Aerospace Technology and Management Vol. 4, n.1 (Jan./Mar. 2012) – São José dos Campos: Zeppelini Editorial, 2012 Quartely issued

1. Aerospace sciences 2. Technologies 3. Aerospace engineering CDU:629.73

Journal of Aerospace Technology and Management J. Aerosp. Technol. Manag. Vol. 4, No.1, Jan. - Mar. 2012

EDITOR IN CHIEF Ana Cristina Avelar – Instituto de Aeronáutica e Espaço – (IAE) – São José dos Campos/SP – Brazil editor@jatm.com.br EXECUTIVE EDITOR Ana Marlene F. Morais – Instituto de Aeronáutica e Espaço – (IAE) – São José dos Campos/SP – Brazil secretary@jatm.com.br SCIENTIFIC COUNCIL Angelo Passaro – IEAv – São José dos Campos/SP– Brazil Antonio Pascoal Del'A rco Jr. – IAE – São José dos Campos/SP– Brazil Carlos Antonio M. Kasemodel – IAE – São José dos Campos/SP– Brazil Carlos de Moura Neto – ITA – São José dos Campos/SP– Brazil Eduardo Morgado Belo – EESC/USP – São Carlos/SP – Brazil Francisco Carlos M. Pantoja – IAE – São José dos Campos/SP– Brazil Francisco Cristóvão L. Melo – IAE – São José dos Campos/SP– Brazil João Marcos T. Romano – UNICAMP – Campinas/SP – Brazil Marco A. Sala Minucci – IEAv – São José dos Campos/SP– Brazil Mischel Carmen N. Belderrain – ITA – São José dos Campos – Brazil Paulo Tadeu de Melo Lourenção – EMBRAER– São José dos Campos/SP– Brazil Rita de Cássia L. Dutra – IAE – São José dos Campos/SP– Brazil ASSOCIATE EDITORS Acir Mércio Loredo Souza – UFRGS – Porto Alegre/RS – Brazil Adam S. Cumming – DSTL – Salisbury/Wiltshire–England Adiel Teixeira de Almeida – UFPE – Recife/PE – Brazil Alain Azoulay – SUPELEC– Gif–Sur–Yvette – France Alexandre Queiroz Bracarense – UFMG – Belo Horizonte/MG – Brazil Altamiro Susin – UFRGS – Porto Alegre/RS – Brazil Álvaro Damião – IEAv– São José dos Campos/SP– Brazil André Fenili – UFABC – Santo André/SP – Brazil Antonio F. Bertachini – INPE – São José dos Campos/SP–Brazil Antonio Henriques de Araújo Jr – UniFOA – Volta Redonda/RJ – Brazil Antonio Sergio Bezerra Sombra – UFC – Fortaleza/CE – Brazil Bert Pluymers – KU – Leuven – Belgium Carlos Henrique Marchi – UFPR – Curitiba/PR – Brazil Carlos Henrique Netto Lahoz – IAE – São José dos Campos/SP – Brazil Cosme Roberto Moreira da Silva – UnB – Brasília/DF – Brazil Cynthia Junqueira – IAE – São José dos Campos/SP– Brazil Daniel Alazard – ISAE – Toulouse – France David Murray–Smith – University of Glasgow – Glasgow – Scotland Edson Aparecido de A. Querido Oliveira – UNITAU – Taubaté/SP – Brazil Edson Cocchieri Botelho – FEG/UNESP – Guaratinguetá/SP – Brazil Elizabeth da Costa Mattos – IAE – São José dos Campos/SP– Brazil Fabiano Fruett – UNICAMP – Campinas/SP – Brazil Fabrice Burel – INSA – Lion – France

Fernando Luiz Bastian – UFRJ – Rio de Janeiro/RJ – Brazil Flamínio Levy Neto – UnB – Brasília/DF – Brazil Francisco Carlos P. Bizzaria – IAE – São José dos Campos/SP– Brazil Francisco José de Souza – UFU – Uberlândia/MG – Brazil Gilberto Fisch – IAE – São José dos Campos/SP– Brazil Gilson da Silva – INPI – Rio de Janeiro/RJ – Brazil Hugo H. Figueroa – UNICAMP – Campinas/SP – Brazil João Luiz F. Azevedo – IAE – São José dos Campos/SP – Brazil José Alberto Cuminato – ICMC/USP – São Carlos/SP– Brazil Jose Atilio Fritz F. Rocco – ITA – São José dos Campos/SP – Brazil José Leandro Andrade Campos – UC – Coimbra – Portugal José Rubens G. Carneiro – PUC Minas – Belo Horizonte – Brazil José Márcio Machado – Ibilce/UNESP – São José do Rio Preto/SP – Brazil José Maria Fonte Ferreira – UA – Aveiro – Portugal José Pissolato Filho – UNICAMP – Campinas/SP – Brazil José Roberto de França Arruda – UNICAMP – Campinas/SP – Brazil Luís Carlos de Castro Santos – EMBRAER– São José dos Campos/SP– Brazil Luiz Claudio Pardini – IAE – São José dos Campos/SP– Brazil Marcello de Medeiros Faraco – EESC/USP – São Carlos/SP – Brazil Márcia B. H. Mantelli – UFSC– Florianópolis/SC – B Brazil Marc Lesturgie – ONERA– Palaiseau–France Marcos Pinotti Barbosa – UFMG– Belo Horizonte/MG – Brazil Maurizio Ferrante –UFSCar – São Carlos/SP – Brazil Michael Gaster – Queen Mary University of London – London – England Michele Leali Costa – FEG/UNESP – Guaratinguetá/SP – Brazil Mirabel Cerqueira Rezende – IAE – São José dos Campos/SP– Brazil Othon Cabo Winter – FEG/UNESP – Guaratinguetá/SP – Brazil Paulo Celso Greco – EESC/USP – São Carlos/SP – Brazil Paulo Sérgio Varoto – EESC/USP – São Carlos/SP – Brazil Raimundo Freire –UFCG– Campina Grande/PB–Brazil Renato Machado Cotta – UFRJ – Rio de Janeiro/RJ – Brasil Roberto Costa Lima – IPqM – Rio de Janeiro/RJ – Brazil Romis R. F. Attux – UNICAMP – Campinas/SP– Brasil Samuel Machado Leal da Silva – CTEx – Rio de Janeiro /RJ– Brazil Sandro Haddad – UnB– Brasília/DF–Brazil Selma Shin Shimizu Melnikoff – EP/USP – São Paulo/SP – Brazil Sérgio Frascino M. Almeida – ITA – São José dos Campos/SP – Brazil Ulrich Teipel – Georg Simon OHM – Nürnberg – Germany Valder Steffen Junior – UFU – Uberlândia/MG – Brazil Vassilis Theofilis – UPM – Madrid– Spain Waldemar de Castro Leite Filho – IAE – São José dos Campos/SP – Brazil Willian Roberto Wolf – IAE– São José dos Campos/SP – Brazil Wim P. C. de Klerk – TNO – Rijswijk/SH – The Netherlands EDITORIAL PRODUCTION Glauco da Silva – IAE – São José dos Campos/SP– Brazil Helena Prado A.Silva – IAE – São José dos Campos/SP– Brazil Janaina Pardi Moreira – IAE – São José dos Campos/SP– Brazil Lucia Helena de Oliveira – DCTA – São José dos Campos/SP– Brazil Mônica E. Rocha de Oliveira – INPE – São José dos Campos/SP–Brazil

ISSN 1984-9648 ISSN 2175-9146 (online)

Vol. 4, Nº1, Jan.-Mar. 2012

CONTENTS 05

EDITORIAL JATM: Achievements, Challenges and Targets Ana Cristina Avelar

ORIGINAL PAPERS Establishment of Satellite Formation with Initial Uncertainty by Control Lyapunov Function Approach M. Navabi, M. Barati, Hossein Bonyan Khamseh

5 1

Attitude Control of a Satellite by Using Digital Signal Processing Adrielle C. Santana, Luiz S. Martins-Filho, Ricardo O. Duarte, Gilberto Arantes Jr, Ivan S. Casella

Indoor Radar Cross Section Measurements of Simple Targets Marcelo Alexandre Souza Miacci, Evandro Luís Nohara, Inácio Malmonge Martin, Guilherme Gomes Peixoto, Mirabel Cerqueira Rezende

Ammonium Perchlorate and Ammonium Perchlorate-Hydroxyl Terminated Polybutadiene Simulated Combustion Rene Francisco Boschi Gonçalves, Koshun Iha, Francisco Bolivar Correto Machado, José Atílio Fritz Fidel Rocco

Water Influence in Poly(epichlorohydrin) Synthesis: An Intermediate to Energetic Propellants Jairo Sciamareli, Silvana Navarro Cassu, Koshun Iha

An Analysis of the Initiation Process of Electro-explosive Devices Paulo Cesar de Carvalho Faria, Koshun Iha, José Atílio Fritz Fidel Rocco

Methodology for Structural Integrity Analysis of Gas Turbine Blades Tiago de Oliveira Vale, Gustavo da Costa Villar, João Carlos Menezes 3

Experimental Determination of Temperature During Rotary Friction Welding of AA1050 Aluminum with AISI 304 Stainless Steel Eder Paduan Alves, Francisco Piorino Neto, Chen YingAn, Euclides Castorino da Silva

Characterization of Surface Level Wind in the Centro de Lançamento de Alcântara for Use in Rocket Structure Loading and Dispersion Studies Edson R. Marciotto, Gilberto Fisch, Luiz E. Medeiros

8 1

Study of a Lower-deck Galley for Airliners Marcelo Vieira Abritta, Jürgen Thorbeck, Bento Silva de Mattos

9 5

Proposal of a Methodology of Stakeholder Analysis for the Brazilian Satellite Space Program Mônica Elizabeth Rocha de Oliveira, Leonel Fernando Perondi

1 05

An Overview of the Certification of VSB-30 with Emphasis on Technological Innovation Antonio Ramalho de Souza Carvalho, José Henrique Damiani, Andrea de Oliveira Netto Follador, Marcelo Guido de Oliveira Guimarães

THESIS ABSTRACT

117

Development of a Novel RANS-based Method for the Computational Aeroacoustics of High-speed Jets Carlos Roberto Ilário da Silva

1 17

Study on the Accelerated Aging of the Composite Solid Propellant Júlio de Barros Magalhães

118

Evaluation of Static Strength and Fatigue Life of Carbon Fiber/Epoxy and Glass Fiber/Epoxy Composites under Shear Stress by Using the IosipescuMethod Vanderlei de Oliveira Gonçalves

1 18

Reduction of Autonomy-level Requirement in LEO Sun-synchronous Satellites via Optimization of Longitude Distance Between two High-latitude Ground Segments Hossein Bonyan Khamseh

1 19

Adaptive Performance Optimization for a Transport Aircraft Seyed Hossein Mortazavi

119

Fault Detection and Isolation on Inertial Measurement Units with Minimal Redundancy of Fiber Optic Gyros Élcio Jeronimo de Oliveira

120

INSTRUCTIONS TO AUTHORS

Editorial JATM Achievements, Challenges and Targets Dr. Ana Cristina Avelar* Editor in Chief editor@jatm.com.br

As announced in the previous JATM issue launch event at EMBRAER, Prof. Francisco Cristovão L. Mello stepped down as JATM Editor-In-Chief in December, 2011 due to personal projects. To replace him, the Editorial Committee wanted someone from Instituto de Aeronáutica e Espaço (IAE), which is where JATM was invented, committed with the journal and that used to be integrant of the work group that created it. Therefore, I have been invited and accepted the challenge together with the editorial board of making that JATM improves and achieves its purpose of being a prestigious international source of knowledge dissemination in its area of expertise. I am conscious that we have a long way on, but I really believe that this purpose is worth of being pursued and it can be achieved through personal effort, dedication, and commitment. I take this opportunity to thank the Editorial Committee for the invitation and confidence, and also in behalf of the Editorial Board, I gratefully thank Francisco for all his contributions to our journal. In consonance with its mission of “increasing the knowledge and developing technological-scientific solutions for strengthening the aerospace power, contributing for the national sovereignty and for the progress of the Brazilian society, through education, development, innovation, and technological specialized services in the aerospace field”, IAE has offered the initial financial support for the production of JATM. Although it is still very recent, important achievements have been obtained since the first circulated issue in June, 2009. In fact, JATM has been created because of an initiative of the IAE director, Brig. Francisco Carlos Melo Pantoja, and his commitment with our journal has been determinant for all JATM realizations. Our journal has been publishing papers originating from post-graduate programs, universities, and Research and Development Institutes, both in Brazil and overseas, and from aerospace companies. The articles deal with a large variety of aerospace technology and management areas, which, besides being a relevant topic in the aerospace world, makes the journal even more comprehensive. JATM is already included in several important indexing systems and database of scientific publication: SCOPUS; CAS; DOAJ; J-GATE; LIVRE; Google Scholar; Sumários.ORG; EZB; ULRICHSWEB; SOCOL@R; Periodica; LATINDEX; Periódicos-CAPES, and QUALIS, from where we have just received the classification B4 in Geosciences and Engineering III. The criterions for JATM to remain in these databases are continuously monitored, ensuring the quality of the journal. An increasing interest in our journal, since its creation, has been observed. Last year, the number of papers submitted to JATM increased 35% compared to 2010. The number of papers that has been received in 2012 is also 50% higher than the one received in the same period of 2011.

*Ana Cristina Avelar received a Mechanical Engineering degree from the Universidade Federal de Itajubá, UNIFEI, in 1994. She received the M.S.c and Doctor degree in Mechanical Engineering in the area of Thermal and Fluids Sciences in 1997 and 2001, respectively, by the Universidade Estadual de Campinas, UNICAMP. She starded her carrier in the Instituto de Aeronáutica e Espaço, IAE, as a researcher in 2002. She has been actuating in the area of experimental methods in wind tunnel, mainly with the optical methods of Particle Image Velocimetry (PIV) and Pressure Sensitive Paint (PSP). She participated in courses in these subjects in the German Aerospace Center, DLR, in 2005 and 2009, respectively, and from September 2010 to September 2011 she carried out and a post-doc stage in the Laboratoire de Mécanique de Lille, Lille, France, working with Tomographic PIV. Today, besides researches activities using PIV and PSP she participates in the recently created post-graduation program in Sciences and Space Technologies, PG-CTE, in the Instituto Tecnológico de Aeronáutica, ITA, and coordinates a portfolio of projects of Research and Development in Space Vehicles Associated Technologies.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 5-6, Jan. - Mar., 2012

Since September, 2011, JATM became an official publication from Departamento de Ciência e Tecnologia Aerospacial (DCTA), and it has been financially supported by the prestigious Fundação Conrado Wessel through a partnership with DCTA. This foundation supports other important vehicles of knowledge dissemination in Brazil, such as Revista da Fapesp, Anais da Academia Brasileira de Ciências, among others. It also distributes annually awards to the Brazilian best scientists, artists, and writers in cooperation with important Brazilian research founding agencies. For JATM it is a great honor to be supported by Fundação Conrado Wessel, which is thankfully acknowledged. JATM has also a role in the process of motivating students and professors to start or enhance their scientific production. Free tools for improvement of writing aspects as language and reference revisions are offered, and lectures are promoted for disclosure of the best practices in scientific writings, increasing awareness of ethical responsibility in scientific publications, especially regarding the issue of fraudulent authorship, originality, and novelty. This journal is edited and published in São José dos Campos, São Paulo State, which is a pole of aerospace development. Thus, in spite of publishing papers from several Brazilian states and also from international institutions, predominantly, the papers received by JATM are from São José dos Campos and region. Efforts have been done for divulging JATM through the country and internationally. Our electronic version, which is worldwide freely accessed, through the website www.jatm.com.br, has been a very important tool for showing our journal. Aimed at increasing the journal international visibility as well as at enhancing its quality, starting with this issue, some modifications have been carried out. From now on JATM is going to publish papers only in English. In addition, our editorial board has been reformulated in order to have well-known national and international specialists from different institutions and geographical regions, and a scientific council has been created. In order to more appropriately characterize the field of Aerospace Technology and Management, some specialties, like Materials, Mechanical Engineering and Electronics, which are too generic, have been replaced for more specific ones, to allow allocating the associated editors according to their specializations. The JATM specialties are: Acoustics, Aerodynamics, Aerospace Circuitry, Aerospace Systems, Aerospace Meteorology, Air and Space Defense Systems, Applied Computation, Astrodynamics, Ceramic Materials, Composites, Eletromagnetic Compatibility, Energetics, Fluid Dynamics and Turbulence, Guidance Navigation and Control, Management Systems, Metallic Materials, Photonics, Polymeric Materials, Processing of Aerospace Materials, Propulsion and Combustion, Radars and Tracking Systems, Robotics and Automation, Structures, Synthesis and Characterization of Aerospace Materials, Thermal Sciences, and Vibration and Structural Dynamics. In order to make easier the administration of the papers, since our scope is really broad, every specialty has one to three associated editors. The JATM growing has been carefully planned since its creation in a way that its first volume would have two issues, one publication every six months, and from the second volume on, three issues, at every four months, maintaining its periodicity and punctuality. The journal would become a quarterly publication when prepared for this accomplishment. In fact, the journal is now strong enough, and in the present issue JATM becomes a quarterly periodical. Regarding minor changes, our paper template has also experimented some improvements, concerning to authors’ affiliation and format. In this new version, it is easier to consider all affiliations of the authors, and the correspondence authors are more easily identified. Our cover has been improved as well. In addition, the instructions to the authors are more comprehensive, and an online submission system is being tested and will be shortly available. In summary, JATM has been successful in achieving short-time targets as disseminating knowledge to the scientific community, while maintaining its periodicity and improving its quality. Actually, all JATM accomplishments have been reached due to a great commitment of colleagues from the editorial board, who have been doing a hard work since the creation of the journal. Some of them are Ana Marlene F. Morais, who has been the leader of the work group that created the journal, Cynthia Junqueira, Elizabeth C. Mattos and João Luis F. Azevedo, among others. Finish by assuring our community that a great endeavor is going to be made for a continuous enhancement of JATM quality, representativeness, and international visibility, pursuing in this way our major purpose of becoming a very successful international journal.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 5-6, Jan. - Mar, 2012

doi: 10.5028/jatm.2012.04014511

Establishment of Satellite Formation with Initial Uncertainty by Control Lyapunov Function Approach M. Navabi*, M. Barati, Hossein Bonyan Khamseh Shahid Beheshti University, GC - Tehran - Iran Abstract: In recent years, dynamics and control of satellite formation flying have been active areas of research. From the mission planning perspective, three main areas namely formation establishment, maintenance and reconfiguration have been discussed. In this paper, a study of formation establishment under initial uncertainty is presented. In this regard, dynamics of low Earth orbit satellite formation is discussed. Control Lyapunov function approach is adopted to bring a deputy satellite, with perturbed initial conditions into formation with a chief satellite. In order to take account of the initial orbit insertion error, uncertainty in initial conditions of the deputy satellite is considered. For a case study, a relatively small formation is adopted, with air-launched Pegasus as the launch vehicle. For several initial conditions, control function and required time to achieve a given mission accuracy are determined, and results are provided as illustration. Keywords: Satellite Formation, Lyapunov Function, Dynamics, Control.

INTRODUCTION In recent years, there has been an increasing tendency to replace single large satellites with several small ones in formation flying (Baoyin et al., 2002; Kapila et al., 1999; Navabi et al., 2011; Orr et al., 2007; Carpenter et al., 2003; Schaub, 2004; Sabolet et al., 2001). For that purpose, dynamics of formation flying must be comprehensively studied, and effective control schemes must be devised. Formation control is carried out either by continuous or impulsive methods (Alfriend et al., 2009). Two of the most popular continuous methods are the Control Lyapunov Function (CLF) and the Linear Quadratic Regulator (LQR). Also, from the mission planning perspective, control methods can be studied in the context of formation establishment, maintenance, and reconfiguration (Alfriend et al., 2009). Satellite formation missions with Projected Circular Orbit (PCO) in the local horizon plane are amongst promising missions and thus are considered here (No et al., 2009). Applications of these formations in missions, such as interferometric measurements, have been the theme study in papers such as in Peterson and Zee (2008). Yet, in the establishment phase, launch insertion errors frequently impose uncertainties on initial conditions of the formation and _____________________ Received: 05/10/11. Accepted: 29/01/12 *author for correspondence: navabi.edu@gmail.com/1983963113, Daneshju Blvd.,Yaman st., Chamran Highway, Tehran, Iran.

hence PCO conditions are violated. In this paper, applicability and effectiveness of CLF approach to bring a satellite formation into a specific PCO configuration are studied. In order to take account of the initial insertion errors, uncertainty in initial conditions of the deputy satellite is considered. From the previous launch logs, insertion error can be statistically computed. In most related literature, such as that of Pegasus XL from Orbital Sciences Corporation (2000), insertion error is given in terms of semi-major axis and inclination deltas from the intended values, and the same approach is taken here. Control laws to bring the satellites into a formation with a given position error budget are derived. Then, a case study of satellite formation with PCO is considered. It is assumed that satellites are placed in orbit by Pegasus XL. From the Pegasus launch log, three-sigma deltas of semi-major axis and inclination are derived. With initial conditions under uncertainty, the required time to bring the satellites into given bounded limits is determined and discussed. The rest of the paper is as follows. In the next section, mathematical modeling of dynamics of satellite formation with PCO configuration is presented. Also, CLF to establish a formation with given bounded limits is discussed. Furthermore, a case study is developed and uncertainties in initial conditions due to insertion errors of Pegasus are considered. In the remaining chapters, results are presented and discussed, and possible future works are highlighted.

J. Aerosp. Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.4, No 1, pp. 7-13, Jan. - Mar., 2012

Navabi, M. et al.

MATHEMATICAL MODELING OF DYNAMICS AND CONTROL METHODOLOGY Ordinary Differential Equations (ODE) can be employed to model dynamics of satellite formation flying (Hang et al., 2008). Several ODE models were discussed in a previous work (Navabi et al., 2011), and their sensitivity to J2 perturbation and eccentricity was investigated. It was found that the perturbed nonlinear model exhibits accurate results in all studied scenarios. Thus, the same model is used in this paper. Control methodology pursued here is the popular CLF approach. Mathematical foundations of dynamics and control employed in this paper are further discussed.

where Γ(r), taking account of J2 effects, is defined as Eq. 1.3:

J Z 2 XN K c1 - 5` r j m r O K O n Re K - Z 2 Y O r -3 j c1 5` j m (1.3) J 2 c 2 m` C^ r h = n 3 2 r K r r O r r K O 2 K c3 - 5` Z j m Z O r r L P In Eq. 1.3, μ is Earth gravitational parameter, J2 is a coefficient of Earth gravitational harmonics, and Re is equatorial radius of the Earth. Also, X,Y,Z are position coordinates in the ECI frame. If the desired relative orbit of deputy satellite is designated rd, position error of the deputy shown by δr is as Eq. 1.4. The control method for this purpose is discussed in the next subsection.

DYNAMICS OF SATELLITE FORMATION FLYING

dr = rd - rdd (1.4) The ODE approaches undertaken in dynamics modeling of satellite formation flying are often described in two main Cartesian coordinates. The first coordinate employed in this context is the earth-centered inertial (ECI) frame. In this frame, the fundamental plane is the equator, the unit vector Xt is directed from the Earth center towards the Vernal Equinox, Zt is normal to the fundamental plane and points in the Northward direction, Yt completes the triad. This inertial coordinate frame is shown in Fig. 1

It is assumed that the chief satellite is in free motion, and only the deputy satellite can be controlled to maintain a given desired relative orbit. If position vector of the chief satellite is denoted with rc and that of the deputy with rd, one may obtain the perturbed nonlinear model as Eq. 1.1 and 1.2:

rp c = C^rch (1.1)

2 m = ^ ^ Xd- Xddh + ^Yd- Yddh + ^ Zd- Zddh hm (1.5) 2

where, m is the index of simulation time steps. In order to quantify the expression in Eq.1.4, orbits of the chief and deputy satellite and also the desired relative orbit must be determined. For that purpose, initial conditions for solving Eq. 1.1 can be either given by inertial states of the chief satellite or by its orbital elements. For the desired relative orbit, PCO conditions are considered. Therefore, desired relative orbit is given by chief satellite orbital elements added by certain delta values derived from PCO conditions, which are given by Eq. 1.6 (Alfriend et al., 2009):

Figure 1. Earth-centered inertial frame.

rp d = C^rdh + u

It must be mentioned that the goal of CLF approach in this paper is to eliminate this position error. Yet, if one considers Eq. 1.4 as an error index, positive and negative errors may cancel out each other. Thus, l2 norm of Eq. 1.4 is adopted as the error index, as shown by Eq. 1.5:

(1.2)

dm (0) + dX (0) * cos (ic) =

t (0) ec cos (~ c (0) + a (0)) 2ac

t (0) cos (~ c (0) + a (0)) 2ac t (0) (1.6) sin (~ c (0) + a (0)) ed cos ^dM^0hh = ec 2ac t^0h dX^0h =sin a^0h ac sin ic t^0h di = cos ^a (0h ac ed sin ^dM^0hh =

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 7-13, Jan. - Mar, 2012

Establishment of Satellite Formation with Initial Uncertainty by Control Lyapunov Function Approach

where indexes “ c ” and “ d ” correspond to chief and deputy satellite, respectively. In Eq. 1.6, a, e, i, ω, Ω, M are the classical orbital elements and λ = M + ω. ρ(0) and α(0) are initial PCO radius and phase angle in the magnitude-phase form. δa is obtained from no along-track drift condition or the so-called J2-invariant condition, given by Eq. 1.7 (Alfriend et al., 2009):

R 2 3h c + 4 m6^ 1 3 co 2 h dh da = 0. J2 ac c e m c ac h5c (1.7)

In Eq. 1.7, h = 1 - e Initial conditions of the deputy satellite are considered as those of the desired relative orbit with some initial uncertainty due to insertion error. In a compact form, one may write Eq. 1.8:

N d = N d + dNinsertion error = N c + dN PCO + dNinsertion error d

(1.8)

where, N d and N d are the actual and desired orbital elements sets of the deputy satellite, and Ξc is the orbit al elements set of the chief satellite. δΞPCO are the expressions given by Eqs. 1.6 and 1.7. Also, the term δΞinsertion error is considered to accommodate uncertainties in initial conditions of the deputy satellite orbit and it is quantified based on previous insertion error history of the launch vehicle of the formation. In the following sections, insertion error of Pegasus will be adopted as a case study. With all the initial conditions given, dynamics modeling in ECI frame is accomplished. However, in most formation flying applications, it is desirable to visualize the formation in the rotating so-called Hill frame, shown in Fig. 2. d

and orbit momentum direction, respectively. The transformation from ECI reference frame to Hill frame is given by Eq. 1.9.

cX ci - sX si ci sX si + cX si ci si si = + T ( X , i , i ) c s s c c s s c c c ci si H > Xi Xi i X i X i i (1.9) - cX si sX si ci where cξ and sξ denote cosine and sine of a given argument ξ. For illustration of our results regarding desired relative orbit and actual position of the deputy satellite, Hill frame will be used. As it was mentioned, in our context, it is desired to eliminate position error given by Eq. 1.4. The control method for this purpose is discussed in the next subsection.

CONTROL LYAPUNOV FUNCTION FOR SATELLITE FORMATION FLYING To establish a formation within a bounded limit, in Schaub and Junkins (2002), the following form of Lyapunov function V was defined (eq. 1.10):

1 1 V = ^dr, dro h dro T dro + dr T 6 K1 @ dr (1.10) 2 2 where the 3x3 matrix [K1] is a positive definite position feedback gain and (.)T denotes transpose. It can be seen that Eq. 1.10 is positive definite. If one considers an equivalent dot product form of Eq. 1.10 as V^dr, dro h = 1 dro .dro + 1 dr.^6 K1 @drh 2 2 and takes its derivative, one may find Eq. 1.11:

Vo = dro T ^rpd - rpd + 6 K1 @ drh (1.11) d

Substituting Eq. 1.2 for the 1.11 and for a J2-invariant desired relative orbit, the following relationship is obtained:

Vo = dro T ^C (rd) + u - C (rd ) + 6 K1 @ drh (1.12) d

where Γ(r) is given by Eq. 1.3. To obtain a negative definite expression for Vo , the following form is adopted for control law (Eq. 1.13): Figure 2. A general satellite formation in the Hill frame.

u =-^C (rd) - C (rd )h - 6 K1 @ dr - 6 K2 @ dro (1.13)

As it can be seen from Fig. 2, Ôr ,Ôθ and Ôh are Hill’s unit vectors. These unit vectors are in radial, tangential,

where the 3x3 [K2] matrix is a positive definite velocity feedback gain. From Eqs. 1.12 and 1.13, one can obtain the

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 7-13, Jan. - Mar., 2012

Navabi, M. et al.

following relationship for

Vo (Eq. 1.14):

Vo =- dro 6 K2 @dro (1.14) In order to ensure asymptotic stability, from Eq. 1.14 one may obtain dro T Vp =- 2dro T drp. 6 K2 @ dro . Also, from T T Eq.1.11, dro drp = Vo - dro 6 K1 @ dr . After simplification, one obtains Eq. 1.15: T

for such missions. As a result, relative insertion error is taken as a percentage of the absolute insertion error. For Pegasus XL, the absolute insertion errors are given in Table 3. Table 1. Initial conditions of the chief satellite. Orbit Element

Vp = 2^6 K2 @dro + 6 K1 @drhT 6 K2 @ dro (1.15)

One may note that Vp = ^dro = 0h = 0 and:

Vq = 2^ 6 K2 @drp + 6 K1 @dro

(1.16)

Value

Dimension

7032.9998

0.00998

97.9980

Deg

9.9964

Deg

30.1762

Deg

-0.1803

Deg

Table 2. Desired initial conditions of the deputy satellite.

Evaluating Eq. 1.11 at dro = 0 and noting that Vo < 0 , one obtains drp =- [K1] dr . Substituting the

Orbit Element

previous expression in Eq.1.16, it is easy to show that Vq (dro = 0) =- 2dr T 6 K1 @T 6 K2 @6 K1 @dr 1 0 and thus, asymptotic stability of CLF approach is ensured (Mukherjee and Chen, 1993). In the next section, a case study will be discussed for illustration.

Value

Dimension

7033

0.0100

Deg

Orbital period: 5869.8 seconds

CASE STUDY Table 3. Three-sigma absolute insertion errors of Pegasus XL.

In this section, a case study was developed to investigate effectiveness of the methodology already presented. In accordance with a previous work (Navabi et al., 2011), the following initial conditions are considered for the chief satellite, given in Table 1. Desired orbital elements of the deputy satellite are those of the chief satellite given in Table 1 added by delta values given by Eqs. 1.6 and 1.7. In Orr et al. (2007) and No et al., (2009), two case studies were discussed in which PCO radius had the value of 1 and 0.5km, respectively. Here, it is assumed that radius of PCO is 0.5km and also phase angle is 120º. Table 2 provides the desired orbital elements of deputy satellite. To fully describe input data for this case study, initial uncertainty in deputy satellite initial conditions is discussed. As it was mentioned, this uncertainty is adopted as a mechanism to accommodate insertion error of the launch vehicle in formation establishment phase. To this date, few practical formation flying missions have been realized and thus there are not sufficient data regarding relative insertion error 10

Delta in semi-major axis

Delta in inclination

Pegasus XL with HAPS

±15 km

±0.08º

configuration

Relative insertion errors for formation establishment are assumed to be from 1 to 5% of the three-sigma absolute insertion errors. Assuming 1% step, five scenarios for relative insertion error of semi-major axis and inclination are obtained. For semi-major axis, the five insertion error scenarios are: δainsertion error=150, 300, 450, 600, 700m. And for inclination delta due to insertion error: δainsertion error= 0.0008, 0.0016, 0.0024, 0.0032, 0.0040deg. Thus, 25 scenarios must be studied to accommodate initial uncertainty in deputy orbital initial conditions, due to initial insertion error. A maximum acceptable error index of is considered so that response time of CLF approach for various initial conditions can be compared. A MATLAB® code was

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 7-13, Jan. - Mar, 2012

Establishment of Satellite Formation with Initial Uncertainty by Control Lyapunov Function Approach

developed to fully study this case study. Results are given in the next section.

Fig.5, controlled actual orbit of the deputy satellite is given. In Fig. 6, time history of error index (Eq. 1.5) is presented.

RESULTS AND DISCUSSION For the case study presented in the previous section, simulations were carried out and results are given in this section. Initial conditions of the chief satellite orbit and those of desired relative orbit are constant in all scenarios, but initial conditions of the deputy satellite vary in each scenario. For the first scenario, initial errors in insertion are:

dNinsertion error = )dainsertion error = 150m diinsertion error = 0.0008 deg Figure 4. Control accelerations in the earth-centered inertial frame.

For illustration, in this scenario, Fig. 3 depicts the desired relative orbit along with the actual formation, without any control effort.

Figure 5. Controlled actual and desired relative orbits in the Hill

frame.

Figure 3. Actual (without control) and desired relative orbits in

Hill frame.

As it can be seen from Fig. 3, actual and desired relative orbits do not initially coincide, due to the orbit insertion error. If not controlled, the deputy satellite will gradually drift away. In order to control the formation, control law given in Eq. 1.13 must be applied. In Fig. 4, control accelerations for this scenario are shown in the ECI frame. With the above control accelerations, simulations resulted in a controlled orbit with prescribed accuracy, as shown in Fig. 5. In Fig. 5, the desired relative orbit and its projections in three orthogonal, xy, xz, yz, planes are shown. As it can be seen, the yz projection is a circle of radius 0.5km. Also, in

Figure 6. Error index time history.

From Fig. 6, initial error of the formation is 156.3m, which corresponds to δainsertion error= 150m and δainsertion error= 0.0008deg. At t = 2,114 seconds, i.e., in less than

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 7-13, Jan. - Mar., 2012

Navabi, M. et al.

half a period of the chief satellite, formation is controlled in a manner that maximum acceptable error criterion is satisfied. Including similar figures for the remaining, 24 scenarios would make this paper prohibitively long and, thus, results are summarized in Table 4. Data presented in Table 4 are organized as follows. In each five successive scenarios, i.e., scenario numbers 1-5, 6-10, 11-15, 16-20, and 21-25, inclination error is propagated from 0.0008 to 0.0040 with steps of 0.0008. In the first five scenarios, δa = 150m and in the next five scenarios (scenario no. 6-10), δa = 300m. In the third, fourth and fifth five scenarios, i.e. scenarios numbers 11-15, 16-20 and 21-25, δa = 450, 600, 750m, respectively. The scenarios discussed in Table 4 have been chosen in a manner so that initial error follows a rising tendency with scenario number. From Table 4, in the first scenario which has minimum insertion error (д0 = 156.3m), it takes 2,114 seconds for CLF approach to bring and keep the deputy and chief

satellites to proximity of less than 50m (Fig. 6). In the fifth scenario, inclination error is largest and initial position error is approximately 285m. For this scenario, it takes 3,177 seconds (slightly more than half a period of chief satellite) to establish the formation, with maximum acceptable error of 50m. In the second five scenarios where δa = 300m, minimum and maximum insertion error is 301m (scenario number 6) and 384m (scenario number 10). For these scenarios, it takes CLF control law 3,272 and 3,753 seconds to establish the formation, respectively. Initial insertion error in scenario numbers 11 and 15 are 448m and 507m. For these scenarios, required time for formation establishment by CLF approach is 4,041 and 4,322 seconds, respectively. In the 16 th and 20 th scenarios, initial insertion error is 596m and 642m and required time for formation establishment with these initial errors is 4,611 and 4,711 seconds, respectively. Finally, for

Table 4. Response time of formation control based on Control Lyapunov Function approach for several initial conditions.

Scenario #

Initial error (m)

Time to maximum acceptable error (seconds)

δa = 150, δi = 0.0008 → д0 = 156.3m

2,114

δa = 150, δi = 0.0016 → д0 = 177.5m

2,311

δa = 150, δi = 0.0024 → д0 = 208.1m

2,600

δa = 150, δi = 0.0032 → д0 = 244.6m

2,889

δa = 150, δi = 0.0040 → д0 = 284.8m

3,177

δa = 300, δi = 0.0008 → д0 = 301.0m

3,272

δa = 300, δi = 0.0016 → д0 = 312.5m

3,366

δa = 300, δi = 0.0024 → д0 = 330.9m

3,463

δa = 300, δi = 0.0032 → д0 = 355.0m

3,653

δa = 300, δi = 0.0040 → д0 = 383.8m

3,753

δa = 450, δi = 0.0008 → д0 = 448.2m

4,041

δa = 450, δi = 0.0016 → д0 = 456.0m

4,041

δa = 450, δi = 0.0024 → д0 = 468.8m

4,134

δa = 450, δi = 0.0032 → д0 = 486.1m

4,229

δa = 450, δi = 0.0040 → д0 = 507.5m

4,322

δa = 600, δi = 0.0008 → д0 = 596.0m

4,611

δa = 600, δi = 0.0016 → д0 = 601.9m

4,614

δa = 600, δi = 0.0024 → д0 = 611.6m

4,616

δa = 600, δi = 0.0032 → д0 = 625.0m

4,708

δa = 600, δi = 0.0040 → д0 = 641.8m

4,711

δa = 750, δi = 0.0008 → д0 = 744.1m

4,999

δa = 750, δi = 0.0016 → д0 = 748.8m

5,001

δa = 750, δi = 0.0024 → д0 = 756.7m

5,001

δa = 750, δi = 0.0032 → д0 = 767.5m

5,091

δa = 750, δi = 0.0040 → д0 = 781.2m

5,095

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 7-13, Jan. - Mar, 2012

Establishment of Satellite Formation with Initial Uncertainty by Control Lyapunov Function Approach

Carpenter, J. R. et al., 2003, “Benchmark problems for spacecraft formation flying missions,” in Proceedings of the AIAA Guidance Navigation and Control Conference.

scenario numbers 21 and 25, initial error is 744 and 781m and required time for formation establishment is 4,999 and 5,095 seconds. It must be mentioned that even for the 25th scenario with worst-case insertion error, CLF can establish the formation in less than a chief satellite period. This finding proves effectiveness of CLF approach for initial formation establishment taking into account uncertainty due to initial insertion error. For any specific mission, one may adopt a similar approach, and considering various possible initial insertion errors, employ CLF for initial formation establishment

Hang, Y. et al., 2008, “Comparison study of relative dynamic models for satellite formation flying,” Proceedings of 2nd IEEE ISSCAA.

CONCLUSIONS

Mukherjee, R., and Chen, D., 1993, “Asymptotic Stability Theorem for autonomous Systems,” Journal of Guidance, Control, and Dynamics, Vol. 16, No. 5, p. 961-963.

Dynamics of satellite formation flying was discussed. Initial conditions of a desired relative orbit, based on J2-invariant and PCO configurations, were adopted. Uncertainty in initial condition was employed as a mechanism to take account of initial insertion error. CLF approach was adopted to establish satellite formation under these uncertainties. For a sunsynchronous chief satellite at 655km of altitude, it was shown that initial uncertainties in deputy satellite orbit is high 750m in semi-major axis and 0.004 degree in inclination can be corrected in less than a chief satellite period.

FUTURE WORK

Kapila, V. et al., 1999, “Spacecraft formation flying: dynamics and control,” Proceedings of the American Control Conference, San Diego, CA, pp. 4137-4141.

Navabi, M. et al., 2011, “A comparative study of dynamics models for satellite formation flying – Ordinary Differential Equations approach,” IEEE RAST Conference, Istanbul, Turkey, pp. 829-832. No, T. S. et al., 2009, “Spacecraft formation-keeping using a closed-form orbit propagator and optimization technique,” Acta Astronautica, Vol. 65, Issue 4, pp. 537-548. Orr, N. G. et al., 2007, “Precision formation flight: the CanX4 and Can-X 5 dual nanosatellite mission,” 21st AIAA/USU Conference on Small Satellites, Salt Lake City, UT.

In this study, required time for initial formation establishment by the means of CLF approach was discussed. In future works, this method will be extended by considering simultaneously required time and total control effort for formation establishment.

“Pegasus User’s Guide”, 2000, Orbital Science Inc, Verison 5.0.

REFERENCES

Sabol, C. et al., 2001,“Satellite formation flying design and evolution,” Journal of Spacecraft and Rockets, Vol. 38, No. 2, pp. 270-278.

Alfriend, K. T. et al., 2009, “Spacecraft formation flying, dynamics, control and navigation”, Published by ButterworthHeinemann. Baoyin, H. et al., 2002, “Dynamical behaviors and relative trajectories of the spacecraft formation flying,” Journal of Aerospace Science and Technology, Vol. 6, Issue 4, pp. 295-301.

Peterson, E., and Zee, R. E., 2008, “Possible Orbit Scenarios for an InSAR Formation Flying Microsatellite Mission,” 22nd AIAA/USU Conference on Small Satellites, Logan, Utah.

Schaub, H., 2004, “Relative orbit geometry through classical orbit element differences,” Journal of Guidance, Control, and Dynamic, Vol. 27, No. 5, pp. 839-848. Schaub, H., and Junkins, J. L., 2002, “Analytical Mechanics of Space Systems”, AIAA education series, p. 539-40.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 7-13, Jan. - Mar., 2012

doi: 10.5028/jatm.2012.04014611

Attitude Control of a Satellite by Using Digital Signal Processing Adrielle C. Santana1, Luiz S. Martins-Filho1,*, Ricardo O. Duarte2, Gilberto Arantes Jr. 3, Ivan S. Casella1 Universidade Federal do ABC - Santo André/SP - Brazil Universidade Federal de Minas Gerais - Belo Horizonte/MG - Brazil 3 Universität Bremen - Bremen - Germany 1 2

Abstract: This article has discussed the development of a three-axis attitude digital controller for an artificial satellite using a digital signal processor. The main motivation of this study is the attitude control system of the satellite Multi-Mission Platform, developed by the Brazilian National Institute for Space Research for application in different sort of missions. The controller design was based on the theory of the Linear Quadratic Gaussian Regulator, synthesized from the linearized model of the motion of the satellite, i.e., the kinematics and dynamics of attitude. The attitude actuators considered in this study are pairs of cold gas jets powered by a pulse width/pulse frequency modulator. In the first stage of the project development, a system controller for continuous time was studied with the aim of testing the adequacy of the adopted control. The next steps had included an analysis of discretization techniques, the setting time of sampling rate, and the testing of the digital version of the Linear Quadratic Gaussian Regulator controller in the MATLAB/SIMULINK. To fulfill the study, the controller was implemented in a digital signal processor, specifically the Blackfin BF537 from Analog Devices, along with the pulse width/pulse frequency modulator. The validation tests used a scheme of co-simulation, where the model of the satellite was simulated in MATLAB/SIMULINK, while the controller and modulator were processed in the digital signal processor with a tool called Processor-In-the-Loop, which acted as a data communication link between both environments.function and required time to achieve a given mission accuracy are determined, and results are provided as illustration. Keywords: Satellite Attitude Control, Linear Quadratic Gaussian Control, Digital Signal Processor, Pulse Width/ Pulse Frequency Modulation.

INTRODUCTION The development of satellites in Brazil has become, in the last decades, a subject of extreme importance. The economic relevance and strategic role for the nation are appealing. There are scientific and technological efforts from the government in order to stimulate the space research. There are also in research centers and in the private sector. Nowadays, many activities are associated with the achievements and technological breakthrough in space. The applicability of satellites can be found for a large number of purposes, e.g., terrestrial, maritime and aerial navigation by using GPS, communication, weather forecasting, etc. ___________________ Received: 07/10/11. Accepted: 22/12/11 *author for correspondence: luiz.martins@ufabc.edu.br - Av. dos Estados, 5001- Bangú - Santo André/SP - Brazil

The frequent manifestations of authorities from the Brazilian Space Program confirm their intention to dedicate financial and human resources in activities that can place the country in an autonomous level. Those activities include the creation of an infrastructure for launch, operation, and development of their own satellites. It is worth to mention that the costs of images from weather satellite, or the use of communication satellites, represent relevant sums and boost investments in the global economy sector. Moreover, the space research provides important achievements for several areas into science and technology. To illustrate the economic relevance, one can mention the 47-million-dollar contract, which was signed between Brazil and an Argentinean company (INVAP) for the development of attitude control and data handling (ACDH) system for the Multi-mission Platform (MMP) mission satellite. The MMP

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 15-24, Jan. - Mar., 2012

Santana, A.C. et al.

consists of a carrier module that can incorporate different payloads for different applications, such as the Amazon-1 satellite for Earth observation and surveillance. The MMP satellite is being developed at the National Institute for Space Research (INPE, acronyms in Portuguese). According to the director of INPE, Gilberto Camara, the Brazilian space industry is characterized by the specialization in different areas, such as cameras for remote sensing, structures, etc. However, there is no company in the country able to develop or that holds the technology of an ACDH system. This is a result of no specialized manpower. The actual educational formation is focused on the development of hardware, but not software (Mileski, 2009). In this context, the objective of this paper was to contribute to overcome this lack of technological capability. This paper studies an attitude control system (ACS), more specifically, a digital system for attitude control for three-axis stabilized satellites. The development on an embedded system is addressed. In this case, the information obtained by the sensors is performed by a digital signal processor (DSP). The controller design is based upon a linear-quadratic Gaussian (LQG) optimal control, which is a well-known control theory in terms of space applications. The design combines a Kalman filter and a linear quadratic regulator (LQR) control law. In addition, within the synthesis of the control law, we adopted a pulse-width pulse frequency (PWPF) modulator, which is responsible for modulating the signal control sent to the reaction thrusters that are considered in this work. The quality of an ACS using propulsion controllers is strongly influenced by the modulation of the control command. Therefore, the practical issue of translating the continuous desired signal to an on/off signal is done.

THE MMP SATELLITE The MMP satellite is a project based on an advanced concept in terms of architecture of a satellite. It intends to bring together in one versatile platform all equipment that are essential to the operation of a satellite, such as power generation and distribution, attitude control, and propulsion, etc. It can accomplish a variety of applications with different orbits and mission requirements. In this architecture, there is a physical separation between the platform and the payload module. Thereafter, both modules can be design, built and tested separately before the integration and final assembly and testing of the satellite. The advantage 16

associated with the MMP concept is the cost reduction in manufacture of the satellite, i.e., the navigation module would be the same unless any updates and upgrades are implemented, regardless of the payload hosting the MMP. Figure 1 illustrates the MMP (INPE, 2010).

Figure 1. Illustration of the Multi-Mission Platform (INPE, 2010).

Some applications are already planned for the MMP, including the first Remote Sensing Satellite (SSR-1), also named Amazonia-1. The Amazonia-1 will be equipped with instrumentation for imaging a wide field imager (AWFI) full developed in Brazil. It is also equipped with transponders for collecting data from data-collecting platforms (DCPs), which can be used for Weather Forecasting and Climate Studies in Brazil (INPE, 2004). Its applications include evaluation and estimation of productivity in cultivated areas, and monitoring of pollution in coastal areas. The main objective of the mission is the acquisition of data for monitoring the Amazon system. The Remote Sensing Satellite 2 (SSR2), also known as MultiApplication Purpose SAR (MAPSAR), is another candidate to a future mission based on MMP. MAPSAR is the result of a joint study conducted by INPE and the German Aerospace Center (DLR) targeting a mission for assessment, management and monitoring of natural resources, comprising cartography, forestry, geology, geomorphology, hydrology, agriculture, disaster management, oceanography, urban studies, and security (Schröder et al., 2005). Attitude model: kinematics and dynamics In order to model the attitude of the satellite, we have considered two reference systems: the orbital frame or the Local-Vertical-Local-Horizontal (LVLH) and the satellite frame. The coordinate frame LHLV (x0 , y0 , z0) moves with

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 15-24, Jan. - Mar., 2012

Attitude Control of a Satellite by Using Digital Signal Processing

the satellite. Its origin coincides with the center of the mass of the satellite. The axis z0 of the LVLH frame is defined by the satellite radius vector, the axis y0 by the orbital normal, and the axis x0 completes a right handed coordinate frame. The body or satellite coordinate frame (x ,y ,z) has its origin in the center of mass of the satellite. Their axes coincide with the principal axes of inertia of the body. The nominal or the desired attitude is one that the axes of the reference body are aligned with those of the reference frame LVLH (Wie and Arapostathis, 1989). The Euler’s angles are used to parameterize the attitude, the sequence of rotations 3-2-1 is chosen. The rotation matrix is given by Equation 1 (Shuster, 1993):

- cos d - sin d ur = url ur mc m Rd c m e o=c sin d - cos d vr vrl vr (1)

The kinematics equation is obtained by the time derivative of the rotation matrix equation (Shuster, 1993), then (Eq. 2): t^t0, g h =

vl (t0) vl (t0 + g) 2 vv

(2)

The kinematics equation can be simplified if one considers small angle maneuvers, in this case we can approximate: sin z, sin i, sin }, cos z, cos i e cos } to z, i, }, 1, 1 e 1 , respectively. Moreover, the non-linear terms o h are very small compared to the linear ones. The o , }z o , iz ^}i velocities }o and io are also small compared to the orbital velocity ω0 (Arantes Jr. et al., 2009). Considering those approximations, the kinematic equation is given by Eq. 3: t

()=

(3)

In this paper, the satellite is modeled as a rigid body. Therefore, the dynamic model can be obtained with Euler’s equation that describes the rotation of a rigid body. The dynamic equation is given by Eq. 4: t 2 2 2 t v y = 2v v (t) x L 8 - 1 + exp `- jB xL xL

where: b J is the inertia matrix of the satellite; x d represents the torques b from external perturbations acting on the satellite; and x p represents the control torques (τx, τy, τz). b All vectors described in the body frame, ~ib is the angular velocity of the body with respect to the inertial frame written in the body frame. The gravity gradient is the only external perturbation considered in this work. Its effect cannot be neglected in a low-orbit satellite, an asymmetric body subject to the Earth gravitational field will experience a torque tending to align the axis of minor inertia with the field direction. The gravity gradient torque is modeled as Eq. 6:

v xL = u b / (6) xE vv Substituting the applied torques and the kinematic equation (Eq. 3) into the dynamic equation, and representing the dynamic equation in the state space form, we have (Arantes Jr. et al., 2009) Eq. 7: xo = Ax + Bu

(7)

y = Cx + Du

CONTINUOUS-TIME LQG CONTROL The LQG is designed upon the linearization of the dynamic model. The theory is developed for linear systems. However, the simulation takes into consideration the complete non-linear model of the satellite. The optimal control approach consists of minimizing a quadratic cost function and computing a feedback gain matrix (Dorato et al., 1998). The optimization problem aims at obtaining a control law expressed by a linear relationship between the state variable x (expressing the Euler's angles and the angular velocities) and the control variable u ( applied torques) given by Eq. 8: u = - K (t)x

(8)

(4) with a gain K that minimizes a quadratic cost function formulated as Eq. 9

We rewrite the expression as Eq. 5: 2 2

v t 2 v y (t) . ) v 2 2v v x L t

for t % x L for t & x L

(5)

Jlqr

= 0

# 6xT Qc x + uT Rc u @dt

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 15-24, Jan. - Mar., 2012

(9)

Santana, A.C. et al.

where, Qc and Rc are the weight matrices of state and control vectors, respectively. The gain matrix K for the optimal control law is obtained by solving the algebraic matrix Riccati equation for a time-invariant system and considering an infinite horizon context (Eq. 10).

After obtaining L, it is possible to obtain the transfer function of open loop LQG controller, according to (Arantes Jr. et al., 2009).

T -1 T A P + PA - PBR c B P + Qc = 0

where, is the transfer function of the attitude dynamics. The matrices Qc and Rc are computed by adjustments from the values obtained by using the rule of Bryson (1994), and Qc and Rc are obtained by heuristic iterative adjustments. The complete control system implemented and tested in MATLAB/ SIMULINK environment is illustrated in Fig. 2.

(10)

where, A and B are the matrices of the linearized attitude model. The optimal control gain is given in terms of the solution P of the Riccati equation (Eq. 11):

-1 Klqg G (s) = K (sI - A + BK + LC) LG (s)

(16)

-1 T K = R B P (11)

Due to the presence of noise, a filter is needed to obtain more reliable information about the states measured by the sensors. For the inclusion of a sensors signals filtering, we adopted the controller design based on the theory of LQG. The LQG regulator problem can be described as the synthesis of a control law, which stabilizes the system and minimizes a quadratic error criterion (Dorato et al., 1998). We consider in this work the presence of white noise in the observations, and that the system is observable. In the LQG issue, we want to minimize the cost function as in Eq. 12:

Figure 2. The Linear-quadratic Gaussian control system developed in MATLAB/SIMULINK environment.

(12) Gas jets actuation where, Qf is the covariance matrix of the measurements noise, and Rf is the covariance matrix of the dynamics noise (to represent the inaccuracy of the model). The solution of the LQG problem will be given by dividing the main problem into two sub problems: setting the controller for the linear quadratic deterministic problem, previously defined; setting the Kalman-Bucy filter for optimum estimation of xt state x. The formulation of the Kalman-Bucy filter is given by Eq. 13: xto = Axt + Bu + L (y - Cxt )

(13)

where L is the Kalman filter gain that is obtained by solving the algebraic Riccati matrix equation (Eq. 14): T -1 T A S + SA - SCR f C S + Q f = 0

(14)

The gain of the optimal filter is given by Eq. 15:

-1 T L = R f C S (15)

The linear quadratic controller action can be implemented using actuators, such as reaction wheels and magnetic actuators for a continuous control command. However, during orbital operations, such as rapid detumbling maneuvers, the required torques are usually too high for reaction wheels. Therefore, on-off propulsion strategies are used for such operations (Arantes Jr. et al., 2009). The choice of cold gas jets actuation in the present study aims at testing the control in a most critical situation in terms of the small attitude adjustments difficulties. Future works should use other actuator types (e.g. reaction wheels, magnetic torquers), including combined use with gas jets. It is important to mention that gas jets are also used for reaction wheels desaturation. The firing of gas jets is controlled by a PWPF modulator (Buck, 1996). The PWPF is an interesting option for the thrusters control system due to its advantages over other types of pulse modulators. PWPF is designed to provide propulsion output proportional to the input command. The modulator optimizes the use of propellants; it provides a smoother

J. Aerosp. Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.4, No 1, pp. 15-24, Jan. - Mar., 2012

Attitude Control of a Satellite by Using Digital Signal Processing

control and increases the equipment life. The PWPF structure is shown in Fig. 3 (Arantes Jr. et al., 2009).

Figure 3. Scheme of the pulse width/frequency modulator.

When the positive input in the Schmitt trigger is greater than Uon, the trigger output is Um. If the input falls below Uoff , the trigger output becomes null. This response is also reflected for negative inputs. The error signal e(t) is the difference between the output of Schmitt trigger Uon and system input r(t). This error is sent to a pre-filter, whose input f(t) feeds the Schmitt trigger (Arantes Jr. et al., 2009). Figure 4. Simulation results of attitude control for three angles

Simulations results for continuous-time LQG controller The results obtained in the case of continuous time LQG controller in MATLAB/SIMULINK environment, considering an initial error of 10º in the three Euler's angles, are shown in Fig. 4. It can be seen that even for a fairly large deviation in the stabilization mode context, the controller meets the specification of 0.5º upper error limit. This error can be improved by using other types of actuators, like reaction wheels, which have higher resolution in action.

DIGITAL LQG CONTROLLER The LQG controller design in a digital version requires some study steps of major importance for a successful implementation of the algorithm in an embedded DSP. In this section, we present the main design analysis and decisions with respect to time of sampling, technique of system discretization, and adaption of the LQG controller, originally designed for continuous time systems, to discrete time application. Selection of the sampling time The appropriate selection of the sampling period T is a crucial factor in the digital controller design, since if this period

(roll z, pitch i, yall }) .

is too large there are problems in the signal reconstruction, and if it is too small, system instability and processing capacity problems can occur. In principle, one can believe that smaller sampling period is the best digital approximation. However, if the sampling period is too small, the controller poles approach the unit, causing instability. According to the equation mapping from s and z spaces, where z = esT , we can see that if T is very small, tending to zero, the poles of the controller in z tend to 1, which are those of a marginally unstable system, making the closed-loop system unstable. However, it is not necessary that T approaches zero to start the problems, if it is small enough the poles at z cannot be anymore distinguished by the computer unit. Furthermore, small sampling periods can introduce significant distortions in the system dynamics behavior (Soares, 1996). Very large sampling periods may result in violation of the rule established by the sampling theorem, which says that the sampling frequency ωs must be twice greater than the highest signal component frequency ωM (Ǻstrom and Witternmark, 1997). If the condition of the theorem is not satisfied, there are information losses in the signal reconstruction. A reasonable choice for the sampling rate is 10 to 30 times the bandwidth of the system ωB in closed loop (Ǻstrom and Witternmark, 1997). A suggestion of Franklin et al. (1998)

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 15-24, Jan. - Mar., 2012

Santana, A.C. et al.

is to adopt a sampling frequency greater than 20ωB, in order to have a fairly smooth control response. In Dorf and Bishop (2008), the indication is to adopt the sampling period T =1/(10 fb), where fb = ωB/2π and fb is the bandwidth of the closed loop continuous system. In this work, we adopted the sampling period corresponding to 20fb, which is an average value of the literature suggestions. In order to analyze the system frequencies, we consider the linearized equation of the dynamics of attitude, according to Euler’s angles (roll z, pitch i, yall })

For each dynamics equations, we calculated the open loop transfer function of LQG controller. The closed loop transfer function is obtained by considering a unity feedback. Next, we performed a frequency analysis of the system by the Bode plot of the three transfer functions, and the bandwidth was identified for each equation separately. For each transfer function, a different value of bandwidth was obtained. The obtained bandwidth values for roll, pitch, and yaw angles were respectively 0.4480, 0.0013 and 0.4200 rad/s. The highest value of bandwidth was chosen for this work, since the LQG control law approximates a continuoustime control when the sampling period approaches zero. Taking this into account, the choice of the higher bandwidth will provide a sampling period closer to zero. The obtained maximum sampling period was approximately 0.7 seconds. This reference value of sampling period serves as a starting point to a tuning and adjustment procedure, thereafter a better sampling time value is adopted in this work. It was observed that values higher than 0.7 seconds resulted in a bad dynamic behavior, with the appearance of large oscillations around the reference (zero). Values smaller than 0.7 seconds resulted in a considerable improvement of this dynamics. After several recursive adjustments, the value of 10 ms was considered as a quite satisfactory value, considering the desired pointing accuracy. Furthermore, since the maximum speed of the adopted DSP core is 600 MHz, a sampling period of 10 ms is a reasonable value; it can process the received signal and compute the control signal in the co-simulation in the interval between samples. System discretization The design of control systems in the continuous domain is mathematically simpler and allows the use of a large set of tools. In the case of control system design in discrete domain, the mathematical problem is quite more complicated. In addition, in the continuous time domain, the visualization 20

of the relationship between physical reality and mathematical representation of a control system is more evident. Therefore, the usual starting point for a discrete time control design is the continuous time control system study, followed by discretization procedures. There are several methods of discretization of a given continuous time system, and they are basically divided into open loop methods and closed loop methods. In the openloop methods, the discretization essentially consists of turning the transfer function G(s) in G(z). This transformation is performed by substituting the terms in s by z ones to satisfy some criterion. In the closed loop methods, the discretization of a controller function G(s) is obtained taking into account information from the operation of the closed-loop system, and also the knowledge of all the transfer functions involved in the system, including the plant that is intrinsically continuous. In Soares (1996), several methods of discretization are presented and compared. Among them, the open loop method of transformation of Tustin, also called bilinear transformation, giving satisfactory results even when compared to closed loop ones – which often have better results by taking into account the whole system –, when applied to systems of low order, and also compared with the system of Grenoble, which is of order 9. The Grenoble system consists of a plant considered as a benchmark, developed by the Automatic Control Laboratory of Grenoble. Another well-known discretization method is of retaining elements where the most used is the zero order holder (ZOH). One of the most popular methods is Euler’s, also known as forward difference (Soares, 1996). This method consists of approximating the mapping of the z plane z in the s plane as a truncated series expansion, like in Eq. 17: z-1 sT z = e . 1 + sT & s . T

(17)

In the stability analysis of this method, it could be observed a problem: it is possible that a stable system in s is mapped to an unstable system in z. Although simple, this method is very often adopted. The method of Tustin's bilinear transformation is not based on the approximation of s, as in the case of other methods. It is based on the approximation of the integral represented by the factor 1/s (Soares, 1996). Approaching the integral by trapezoidal integration method, we obtain Eq. 18: s.

2z - 1 Tz + 1

(18)

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 15-24, Jan. - Mar., 2012

Attitude Control of a Satellite by Using Digital Signal Processing

Design of the digital controller The dynamics of the satellite plant in the case of discrete control remains the same as in the continuous time case. Figure 5 illustrates the LQG controller scheme after its discretization. The main change observed in this scheme is the addition of a ZOH in the output signal, which is sent to the satellite's block, as indicated by the output named “torque”. The sensors signals are received by the “measured states” gateway. Finally, the controller subsystem was changed by the discretization in terms of its integration function.

giving opportunity to study realistic problems related to communication and to exchange data between the embedded processor and the controlled system, such as time delays, reliability of transmitted and received data, processing time of the controller, among others.

Figure 5. Scheme of the discret linear-quadratic Gaussian controller.

Results of digital LQG control simulations Figure 6 shows the results of simulations carried out exclusively in MATLAB/SIMULINK environment, using the discrete LQG controller previously obtained by the method of Tustin's transformation, and the controlled plant modeling the non-linear artificial satellite. Figure 6 also shows the variation of the three attitude Euler's angles. This simulation considered the same initial deviation of 10º for each of the three angles.

VALIDATION TESTS The validation of the digital attitude controller was carried through a scheme of co-simulation, where a computer performs the simulation of the satellite's motion, in MATLAB/ SIMULINK, using models of attitude kinematics and dynamics, and a Blackfin 537 DSP device (installed on ADSPBF537 kit, Analog Devices) plays the digital LQG controller processing, and also the PWPF modulator. This type of validation scheme has been used as a way to move beyond on strictly computer simulations,

Figure 6. Simulation results of discrete time attitude control, in terms of the Euler's angles.

An example of this kind of application can be found in Seelaender (2009), where a field programmable gate array (FPGA) processor performs the attitude control of a satellite simulated in computer, combining the MATLAB/SIMULINK and LabVIEW-RT tools. This attitude control uses reaction wheels as actuators. The validation tests comprise two distinct scenarios. In the first one, the tests consider the same case studied in the precedent sections, i.e., three axes attitude stabilization. In the second scenario, the attitude control is aimed at performing a maneuver to achieve a new orientation, that is, a task of attitude tracking. Co-simulation scheme The co-simulation scheme is based on the computer communication with the DSP (Fig. 7), which is facilitated by

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 15-24, Jan. - Mar., 2012

Santana, A.C. et al.

MATLAB/SIMULINK tool, allowing integration of several external processors, including the BF537, by creating a block Processor-In-the-Loop (PIL) in the simulation diagram in SIMULINK. This communication is made possible through the interaction of the MATLAB/SIMULINK with the BF537 development environment, the Visual DSP ++.

Figure 7. Co-simulation scheme, with the digital signal processor processing linear-quadratic Gaussian controller and the pulse width/frequency modulator, and MATLAB/ SIMULINK simulating the motion of the satellite attitude.

The block PIL is inserted into the block diagram developed in the SIMULINK environment. It is responsible for the communication with the DSP, it is in charge of sending the information of the satellite attitude and of receiving the commands related to the control action (the gas jets driving from the PWPF modulator).

Figure 8. Co-simulation results of attitude control using a digital signal processor for the three angles.

Tests results for attitude stabilization Figure 8 shows the results for the three Euler’s angles of the validation tests for the case of three-axis stabilization scheme using co-simulation. The considered initial deviation is 10º for each three Euler’s angle, as in the case of the previous simulations. A comparative analysis of the obtained results in the case of continuous LQG controller, shown in Fig. 4, and those obtained in the scheme of co-simulation (Fig. 8), can be made from a plot of the differences of the results. This plot of differences is shown in Fig. 9. The differences are smaller than 0.1º for the three angles. However, we cannot conclude about the best precision scenario. Both simulations met the accuracy specifications, nevertheless the simulation of continuous-time system lacks of realism for an experimental application. In fact, the small difference between the two cases shows only that the co-simulation scheme works very similarly to the idealized 22

Figure 9. Difference between the simulation results of the continuous controller and co-simulation of digital control for the three angles, in degrees.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 15-24, Jan. - Mar., 2012

Attitude Control of a Satellite by Using Digital Signal Processing

one, suggesting an optimistic outcome in relation to the expectations of an experimental application. In terms of adopted sampling time, it was observed that the simulation in MATLAB/SIMULINK waits the DSP processing to continue the calculations, preventing it from some problem related to the interval between two controllers processing and actuation. However, there is a DSP development platform tool that allows the measurement of processing time and data traffic. Consequently, it is possible to verify that the DSP processing time remains inferior to the maximum period provided for sampling (10ms in this study). This tool is the Cycle Counter, which considers the frequency of the DSP core, 500MHz. The obtained result was processing and traffic time of 8.72×10-5, which is much smaller than the time available, due to the sampling period. It is possible to conclude that there are no problems in this application related to the aspect of the controller discretization and the use of a digital processor to perform the function of controlling and modulating the actuators. Tests results for attitude maneuver tracking In this second test case, the satellite is commanded to change its attitude, initially with three angles (roll, pitch, and yaw) at the same value of 10º, for a new attitude defined by the values 50, 60 and 45º, respectively. The results are shown in Fig. 10. It was observed that the controller performs satisfactorily its task, with tracking maneuver time of about 45 seconds. An important remark: the gap angles between initial and final attitudes are relatively large for the considered approximations in the synthesis of LQG control law. In the controller design, we adopted a linearized model for small angle values, whereas in the simulation of the motion of the satellite attitude we used nonlinear models of kinematics and dynamics. These models, in the case of large angles, exhibit very different behaviors of the linearized model, especially for higher angular velocities. It shows that the adopted controller achieves adequate performance even with this difference between models, and it presents interesting features of robustness, suggesting further studies on this issue.

CONCLUSIONS This paper presents different stages of designing a digital controller for artificial satellite attitude, based on the theory of LQG regulator, defined from the linearized model of the attitude dynamics. The satellite considered in this study was the

Figure 10. Co-simulation results of attitude control for the three Euler’s angles, considering a tracking maneuver.

MMP, developed at INPE. In order to deal with more realistic features, the controller was developed and implemented in a digital processor, and tested within a co-simulation scheme in MATLAB/SIMULINK (simulation of the satellite motion) and DSP (PWPF controller and modulator). The results of numerical simulations, for the three stages of the study development, show the suitability of the LQG controller, as well as the process of discretization of the designed controller and the implementation in a DSP. Continuity of the project can consider three main possibilities. The first will be the designing and test/validation of variant controllers, based on different approaches and theories. The second one involves the use of other types of actuators, such as reaction wheels and magnetic torquers, considered also in use combined with gas jets and other modes or phases of stabilization of the satellite. The latter alternative is to continue the study using experimental platforms, possibly in simplified experimental arrangements in relation to a real satellite in flight. This option allows the performance analysis of the controller in action in an experimental system. The co-simulation scheme adds aspects of realism in the simulations that could not be entirely done in the computing environment, but it does not

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 15-24, Jan. - Mar., 2012

Santana, A.C. et al.

ACKNOWLEDGEMENTS

Franklin, G.F. et al., 1998, “Digital Control of Dynamic Systems”, Manio Park, California, USA: Addison Wesley Longman. INPE - Instituto Nacional de Pesquisas Espaciais, 2010, São José dos Campos, São Paulo, Brazil. Retrieved in March 8, 2010, from http://www.inpe.br/.

The authors acknowledge the support of ABC Federal University (UFABC), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), and of the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP).

INPE – Instituto Nacional de Pesquisas Espaciais, 2004, “Programa Espacial Brasileiro de Observação da Terra”, Anais Seminário de Avaliação e Planejamento da OBT, Campos do Jordão, São Paulo, Brazil.

replace the opportunity to verify aspects of the typical physical experiments.

REFERENCES Arantes Jr.,G. et al., 2009, “Optimal on-off attitude control for the Brazilian multi-mission platform satellite”, Mathematical Problems in Engineering, v. 2009, ID. 750945. Ǻstrom, K.J., Witternmark, B., 1997, “Computer Controlled Systems: Theory and Design”, New Jersey, USA: Prentice Hall. Bryson Jr, A.E., 1994, “Control of Spacecraft and Aircraft”, New Jersey, USA: Princeton University Press. Buck, N.V., 1996, “Minimum Vibration Maneuvers Using Input Shaping and Pulse-Width, Pulse-Frequency Modulated Thruster Control”, MsC. Thesis, Naval Postgraduate School, Monterey, California, USA. Dorato, P. et al., 1998, “Linear Quadratic Control: an Introduction”, New Jersey, USA: Prentice Hall. Dorf, R.C., Bishop, R.H., 2008, “Modern Control Systems”, Upper Saddle River, New Jersey, EUA: Pearson Prentice Hall.

Mileski, A.M., 2009, “Entrevista com Gilberto Câmara - Parte II”, Portal Panorama Espacial, Retrieved in March 30, 2009, from http://panorama.blogspot.com/2009/03/entrevista-comgilberto-camara-parte-ii.html. Schröder, R. et al., 2005, “Mapsar: a small l-band sar mission for land observation”, Acta Astronautica, Vol. 56, p. 35-43. Shuster, M.D., 1993, “A survey of attitude representations”, The Journal of the Astronautical Sciences, Vol. 41, No. 4, p. 439-517. Seelaender, G., 2009, “Emulação e Co-Simulação do Sistema de Controle de Atitude da PMM e do Sistema Eletro-Hidráulico de umaAeronave Usando FPGAs”, MSc. Thesis, Instituto Nacional de Pesquisas Espaciais, São José dos Campos, SP, Brazil. Soares, P.M.O.R., 1996, “Discretização de Controladores Contínuos”, Dissertação de Mestrado, Faculdade de Engenharia da Universidade do Porto, Porto, Portugal. Wie, H.W.B., Arapostathis, A., 1989, “Quaternion feedback regulator for spacecraft eigenaxis rotations”, Journal of Guidance Control and Dynamics, Vol. 12, No. 3, p. 375-380.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 15-24, Jan. - Mar., 2012

doi: 10.5028/jatm.2012.04014711

Indoor Radar Cross Section Measurements of Simple Targets Marcelo Alexandre Souza Miacci1, Evandro Luís Nohara2, Inácio Malmonge Martin2, Guilherme Gomes Peixoto3, Mirabel Cerqueira Rezende3,* Instituto Tecnológico de Aeronáutica - São José dos Campos/SP - Brazil Universidade de Taubaté - Taubaté/SP - Brazil 3 Instituto de Aeronáutica e Espaço - São José dos Campos/SP - Brazil 1 2

Abstract: This paper has described the radar absorbing materials characterization and radar cross section measurements, in the frequency range of 8 to 12 GHz, using a very simple setup. Simple targets like sphere, cylinder, flat plate, and dihedral corner were characterized by measuring the backscattered radiation patterns when these targets were illuminated by monostatic microwave radiation. Measurements were carried out inside an anechoic chamber (9x5x4m3). Typical radar cross section patterns were obtained in different aspect angles, by rotating the targets around their vertical axes. The measured values and the theoretical previsions for each metallic target showed that the used setup guarantees a good precision of the obtained data. By recovering one of the targets, a cylinder, with a specific radar absorbing material developed at Materials Division of the Brazilian Aeronautics and Space Institute, it was possible to compare the obtained patterns and to discuss the influence of radar absorbing materials on the observed radar cross section reduction patterns. The used methodology showed to be useful for attending aeronautical and telecommunication applications. Keywords: Radar Cross Section, Radar Absorbing Materials, Microwave Measurements.

INTRODUCTION Radar cross section (RCS) is the equivalent effective area of a target when it is impinged by a radar wave. In other words, the RCS measures the target ability to reflect radar signals impinged on it into the radar receiver. Radio Detection and Ranging (Radar) systems were developed in the 1940's, mainly to attend military purposes (Skolnik, 1990). Radar is an active remote system that has its own source of energy to produce images. Thus, it does not require sunlight as the optical systems, and data can be acquired either by day or by night. Furthermore, due to the specific wavelength of radar (microwave spectrum), cloud cover or smog can be penetrating by this radiation without any effect on the imagery. Radar is based on the transmission and reception of pulses in a narrow beam. In this paper, this frequency range involved centimeter length of the electromagnetic spectrum. The returning echoes were then recorded, taking into consideration their power, time interval, and wave phases. The received power by the antenna from the transmitted radar pulse was ___________________ Received: 07/10/11. Accepted: 22/11/11 *author for correspondence: mirabelmcr@iae.cta.br /Pç. Mal. Eduardo

directly connected with the physical characteristics of the target through the backscattering coefficient. The value of this backscattering coefficient is basically dependent on three factors: the target surface roughness; the target material composition (electrical and/or magnetic properties), and the wavelength of the radar. The RCS of a target can be viewed as a comparison of the reflected signal power from it to the reflected one from a perfectly smooth sphere with a cross section area of 1m2. The conceptual definition of RCS includes the fact that only one part of the radiated energy reaches the target. The RCS of a target (σ) is most easily visualized as the product of three parameters (Knott et al.,1993). σ = Projected cross section x reflectivity x directivity. The RCS (σ) is used in the range radar equation representing exactly the re-radiated (scattered) power from the target (Eq. 1), that is (Hartman and Berlekamp, 1988; Knott et al.,1993): - cos d url e o=c

- sin d

ur ur mc m = Rd c m

rl r r (1) v

sin d

- cos d

Gomes, 50. CEP: 12.228-901 - São José dos Campos/SP - Brazil J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 25-32, Jan. - Mar., 2012

Miacci, M.A.S. et al.

where, R: distance between radar and target (m); Pt: radar transmitted power (W); Pr: radar received power (W); G: antenna gain; λ: wavelength (m). The reflectivity is defined as the intercepted radiated (scattered) power by the target. The directivity is given by the ratio of the backscattered power into the direction of the radar to the power, which would have been backscattered, considering a uniform scattering in all directions (isotropic). The sphere is the unique target that the RCS is independent of the frequency if operating at sufficiently high frequency, where λ<<range, and also λ<<target radius (R). Experimentally, the radar return reflected from a target is compared to the one reflected from a sphere, which has a frontal or projected area of one square meter, i.e., a diameter of 1.13m (Knott et al.,1993). In general, the RCS of a sphere is taken as a reference value. If calibrated, other targets like cylinders, flat plates, corner reflectors, and Luneberg lens can also be used for comparative RCS measurements. Considering the RCS reduction, these targets can be recovered using layers of radar absorbing materials (RAM). They can be classified in two broad categories, dielectric and magnetic absorbers (Rezende et al.,2003; Johnson, 1992; Lee, 1991). Dielectric absorbers depend on the ohmic loss of the

energy that can be achieved by loading lossy fillers, such as carbon black, graphite, conducting polymers, and metallic particles (Faez et al., 2000; Faez et al., 2005; Folgueras et al., 2007; Biscaro et al., 2008; Folgueiras et al., 2010). The molecules in this kind of lossy material are essentially small dipoles that try to orient themselves along the incident field. If the field changes too fast or if the dipoles lag the impressed field variations, the material absorbs the incident radiation (Knott et al.,1993). Magnetic absorbers depend on magnetic losses, which are obtained when magnetic particles like ferrites and carbonyl iron are filled into a polymeric matrix (Gama and Rezende, 2010; Gama et al.,2011; Silva et al., 2009). Figure 1 shows the calculated RCS values of simple targets commonly employed in experimental measurements. The RCS measurements can be carried out using different techniques, depending on the shape of the target. When RAM efficiency is evaluated by RCS measurements is common to use a double face panel, where one side is used as a reflector material (reference), and the other is coated with RAM. The panel is fixed on a rotating support, which is positioned in front of the receiving and transmitting horns. The advantage of this methodology is that it allows the evaluation of the reference and the absorbing material by rotating the device from 0 to 360º, characterizing both sides of the panel, one after the other. Figure 2 depicts a simplified scheme of a common device used in the RCS technique (Currie, 1989; Miacci, 2002). The use of this device allows to characterize the reference and the RAM using the same experiment, because the RCS diagram

Figure 1. Backscattering from simple shapes (targets) and the RCS equations for each target. r: radius of sphere or cylinder; h: cylinder, flat plate or dihedral length; w: flat plate and dihedral width; L: trihedral length. Tilted Plate reflects away from the plate and could be zero reflected to radar. Based on Johnson, 1992.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 25-32, Jan. - Mar., 2012

Indoor Radar Cross Section Measurements of Simple Targets

of flat plat recovered with RAM is obtained by rotating the device from 0º to 180º and the reference (metal flat plate) is get from 180º to 360º . Thus, this methodology is considered a self-calibrating measurement, once the perpendicularity of the device is guaranteed (Currie, 1989; Miacci, 2002).

Figure 3. RCS measurement setup (R is the distance between the antennas and the target) (Miacci, 2002).

Figure 2. Device scheme used in the RCS method. Based on Miacci, 2002.

In order to have good precision, it is necessary to take care with the alignment between the transmitter/receiver antennas and the target (Burgess and Berlekamp, 1988). The use of a laser beam helps in the alignment of the system, improving the precision of the measurements. Figure 2 also shows the support column for the target, called pylon, which needs to be recovered with RAM in order to avoid any possible contribution of the reflective wave, which prejudices the target characterization. Using this procedure for simple targets and placing the transmitter (TX) and receiver (RX) antennas in the scheme showed in Fig. 3, it is possible to measure the RCS in different frequency ranges. The distance between both antennas needs to be tested in order to eliminate the radiation coupling between them, in near space of the setup. The RCS patterns depend exactly on the target object shape and also on the incidence angle of waves on the same object. The sphere pattern is essentially the same in all directions. The flat plate RCS is strongly dependent on its position in relation toward the radar. The corner reflector has a RCS pattern as high as the flat plate dimensions, with the maximum peaks in the angle range of +60º and -60º. The returned waves from a corner reflector seem analogous to the ones reflected by a flat plate, maintaining the perpendicularity to the line of the transmitter and receiver antennas.

Targets such as ships and aircraft are composed with many effective simple targets as cylinder and corners. An aircraft is a very complex target in terms of RCS signature. It has many reflecting elements and shapes. The RCS of a real aircraft varies significantly depending upon the direction and the frequency band of the illuminating radar. This work showed a study involving RCS measurements of simple targets (flat plate, cylinder, sphere, and dihedral), by using a simple setup projected and built in an anechoic chamber belonged to Instituto de Fomento e Coordenação Industrial do Departamento de Ciência e Tecnologia Aeroespacial (IFI/ DCTA), in the frequency range of 8 to 12GHz. RCS reduction measurements were obtained by recovering the cylinder with a specific RAM processed at the Materials Division of Instituto de Aeronáutica e Espaço (IAE).

EXPERIMENTAL The RCS measurements were carried out at the frequencies of 8, 9, 10, 11, and 12GHz. In each frequency, the target was fixed on the rotating support (Fig. 2) and rotated from 0 to 360º, at a scanning rate of 0.080rad/s. The built setup (Miacci et al., 2001a) for RCS measurements is constituted of: anechoic chamber, matched at 2 to 18GHz; sweep model HP 83630B (Hewlett Packard); spectrum analyzer model HP8593E; PC computer with HPIB interface; low loss coaxial cables from Huber-Suhner Company, model Sucoform SM-141-PE (50Ω); antennas in the range of 8 to 12GHz; Luneberg lens with RCS of 45m2 at 9.375GHz, from Thomson CSF International Inc (Thomson, 1988); targets with different shapes - a flat plate with dimensions of (30x20)cm2, a dihedral corner reflector with flat plates of (17x17)cm2, a metallic cylinder of 32cm of high and 15cm of diameter, all in

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 25-32, Jan. - Mar., 2012

Miacci, M.A.S. et al.

aluminum 2024 T3, and a steel sphere of 60.96cm of diameter, and RAM coating developed at the Materials Division of the IAE. The RAM preparation involved the mixture of 60% (weigh/weigh – w/w) of a commercial polyurethane matrix loaded with fillers, being NiZn ferrite (35% w/w) and carbon black (5% w/w). Physico-chemical characteristics of the fillers and the polyurethane resin as well as the coating preparation procedures were previously described (Dias, 2000). The RAM was applied on the target (cylinder) surface by brushing.

RESULTS AND DISCUSSION

Figure 4. Measured RCS pattern of the Luneberg lens at 9.375GHz.

Luneberg lens

Dihedral

In order to validate the proposed RCS measurement setup, firstly, a certified reference target called Luneberg lens of 30cm of diameter and RCS of 45m2 was used. A Luneberg lens is a sphere constituted of dielectric massive shells. Figure 4 depicts a measured RCS diagram of this target using the built setup, in function of aspect angles between 0 and 180º, and frequency of 9.375GHz. The maximum power (Pmax=-35.0dBm) observed in this diagram corresponds to the RCS value of 45m2, according to the certified value given by the manufacturer (Thomson, 1988). Using the relationship given by Eq. 2, it is possible to determine the RCS values of simple targets in the same frequency (9.375GHz). The constant intensity level of the signal (-35.0dBm), between the aspect angles of -65º and +65º, corresponds to the region with the same scattering level of the lens, being a characteristic RCS pattern for this target. The obtained RCS pattern shows good agreement with that one furnished by Thomson CSF International Inc, validating the built setup. Using these data obtained with the Luneberg lens, it is possible to determine the RCS value for other simple targets based on Eq. 2 and given by Currie, 1989; Miacci et al., 2001b; Miacci et al., 2002:

Figure 5 shows the plotted diagrams of the dihedral corner reflector at 9.375GHz. The theoretical RCS can be calculated using the expression given in Fig. 1. The typical diagram in Fig. 5 depicts three maximum value regions of RCS. Between -20º and +20º the value of -38.4dBm is due to the contribution of the interactions of reflected waves between the two plates. The second and third peaks at 45º and +45º, respectively, present an intensity of -41.3dBm, which is attributed to the individual contribution of each plate of the dihedral, separately. The calculated RCS value for this target at 9.375GHz is equal to 20.5m2 and the experimental value is 20.6 m2. The difference between these two RCS values is equal to 0.1 m2 showing good agreement. Thus, the comparison of the RCS pattern of the Luneberg lens and the dihedral, both at 9.375GHz, allowed calibrating the dihedral in other frequencies.

t^t0, g h =

vl (t0) vl (t0 + g) 2 vv

(2) where, σt : RCS of the test target (m2); Pt: received power from test target (W); Pr: received power from reference target (Luneberg lens) (W), and σr: RCS of the reference target (Luneberg lens) (m2). 28

Figure 5. RCS diagram pattern of a dihedral corner reflector (17x17)cm2 at 9.375GHz.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 25-32, Jan. - Mar., 2012

Indoor Radar Cross Section Measurements of Simple Targets

The RCS diagrams of the dihedral corner reflector in function of the frequency (from 8 to 12GHz) are shown in Fig. 6. Figures 5 and 6 show that the RCS diagrams of the dihedral corner reflector target present nearly the same shapes in function of the frequency when the vertical axes rotates from -90º to +90º. Meanwhile, the amplitude (intensity in dBm) of the signals in the same aspect angles is slightly different with the frequency (Fig. 6). Table 1 shows the measured and calculated RCS values for the dihedral at different frequencies. The determined deviation shows good agreement with the calculated values, using equations presented in Fig. 1, which confirm the adequate RCS methodology proposed. Table 1. Theoretical and experimental measured RCS values of a

measured and calculated RCS values for this sphere (using equation presented in Fig. 1). The mean value of determined RCS from the sphere was equal to 0.31 m2, and the calculated value was 0.29 m2. The small deviation between the measured and calculated values is is assigned to non-uniformity of the spherical surface of the target. Table 2. RCS values of a metallic sphere (diameter of 60.96 cm) in function of the frequency. Frequency (GHz)

Calculated RCS (m2)

Measured RCS (m2)

8.0 9.0

0.29 0.29

0.31 0.30

10.0

0.29

0.32

11.0

0.29

0.28

12.0

0.29

0.31

dihedral of (17x17)cm2 in function of the frequency. Frequency (GHz)

Calculated RCS (m)

Measured RCS (m )

8.0

14.9

14.8

9.0

18.8

19.0

10.0

23.3

23.4

11.0

28.2

28.1

12.0

33.5

33.7

Metallic sphere A metallic sphere with diameter of 60.96 cm was characterized at 8, 9, 10, 11, and 12GHz. Table 2 reports the

Flat plate Figure 7 depicts a typical RCS diagram of an aluminum flat plate, at 8GHz, with a rotation of 180º. It is observed a peak at 0º corresponding to the value of -41.4dBm, attributed to the contribution of the normal incidence of the electromagnetic waves on the flat plate. The position (perpendicularity) of the plate is a very critical parameter for the success and precision of the measurements. The RCS values of a perfect flat rectangular reflector can have its theoretical RCS value calculated as a function of the incident radiation frequency, according to equation depicted

Figure 6. RCS patterns (dBm) of the dihedral of (17x17)cm2, at: (a) 8.0GHz, (b) 9.0GHz, (c) 10.0GHz, (d) 11.0GHz, and (e) 12.0GHz. J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 25-32, Jan. - Mar., 2012

Miacci, M.A.S. et al.

in Fig. 1 (Johnson, 1992). Table 3 shows the calculated RCS values of an aluminum flat plate (20x17)cm2, in function of frequency.

Figure 8. RCS diagram (m2) of an aluminum plate (20x17)cm2 at 8GHz.

Figure 7. RCS diagram in dBm of an aluminum plate (20x17cm2) at 8GHz.

Figure 8 shows the equivalent RCS diagram of flat plate, in square meters, obtained at 8 GHz in function of the aspect angles of the incident wave. The measured peak, equal to –41.4dBm, at 8GHz (Fig. 7), was expressed in square meters (Fig. 8) and correlated with the calculated RCS value of 10.3m2 (Table 3). This comparison shows good agreement confirming the proper adjustment of the used RCS setup.

the reflected waves from this apparatus. The obtained RCS is almost constant at 9.375 GHz (-49.7±1.0dBm) rotating the support in 360º around its vertical axe. This curve is typical for this kind of target and it is attributed to the cylinder shape, which contributes only as a line when the wave impinges on it.

Table 3. RCS values of a flat plate of (20x17)cm2 in different frequencies. Frequency (GHz)

Calculated RCS (m2)

Measured RCS (m2)

8.0 9.0

10.3 13.0

10.2 12.9

10.0

16.1

16.2

11.0

19.5

19.6

12.0

23.2

23.1

Metallic cylinder Figure 9 depicts the plotted RCS diagram of a metallic cylinder of 32cm of length and 15cm of diameter, rotating the cylinder from +90º to -90º on its axes and keeping the TX and TR antennas in the same position, at 9.375GHz. The determined RCS value is constant and omnidirectional, equals to 1.61m2. The cylinder RCS measurement need some tight adjusts on the dielectric support to avoid contributions of 30

Figure 9. RCS diagram of a cylinder, at 9.375GHz

Afterwards, using the same RCS device and the same cylinder, it was obtained the RCS diagram of this target coated with a processed RAM. Figure 10 shows the RCS diagram of the cylinder coated with the RAM presenting 1.2±0.1mm of thickness loaded with NiZn ferrite and carbon black. In this case, the diagram shows the reflectivity variation in function of the aspect angles, characterizing a RCS reduction of the target. This variation is attributed to both the bulk heterogeneity of the processed RAM, related to the ferrite and carbon black particles distribution, and to the absorber thickness variation. Using the curve (a) as reference, Fig. 10 shows that the absorbing propriety presents a reflectivity value of

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 25-32, Jan. - Mar., 2012

Indoor Radar Cross Section Measurements of Simple Targets

-7.8dBm near –60º and a maximum value of -22.8dBm at +30º. Therefore, in these angles, the absorber presents more efficient attenuation values of the incident radiation. Figure 11 depicts the RCS diagram at 12GHz of the same cylinder coated with the RAM. In this case, the absorption mean value is lower than -7.8dBm (in relation to the values of curve (a)), showing that this material behaves as a more efficient absorber in higher frequency range. Again, it is observed that the used RAM reveals similar absorption behavior at 12GHz, showing better attenuation results near –60º and +30º aspect angles.

RCS of simple targets such as sphere, flat plates, Luneberg lens, dihedral corner reflectors, and cylinders. The validation of this setup was made by using a certified Luneberg lens with 45m2, which showed a very good agreement with the data furnished by the Luneberg lens manufacturer. Simple metallic targets, flat plates, cylinder, dihedral and sphere were characterized by measuring the backscattered radiation patterns in different aspect angles, in the frequency range of 8 to 12GHz. The comparison of the RCS data with that one calculated by theoretical equations showed good agreement. RCS reduction measurements were also made recovering a cylinder with RAM developed at the IAE. The obtained RCS patterns showed that this methodology is adequate to characterize this kind of material revealing that the used RAM reduced the RCS of the tested target and changed its reflectivity behavior as a function function of the aspect angles.

ACKNOWLEDGEMENTS

Figure 10. RCS diagrams of cylinder (a) without a RAM and (b) coated with a RAM, at 10GHz

The authors thank Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) (project No. 98/158394), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) with projects 305478/20095 and 300228-87, and the Estado-Maior da Aeronáutica (EMAer) for the financial support.

REFERENCES Biscaro, R.S. et al., 2008, “Influence of doped polyaniline on the interaction of PU/PAni blends and on its microwave absorption properties”, Polymers for Advanced Technologies, Vol. 19, pp 151-157.

Figure 11. RCS diagrams of cylinder (a) without a RAM and (b)

Burgess, L. R. and Berlekamp, J., 1988, “Understanding Radar Cross-Section Measurements”, MSN & CT – Microwaves Systems News & Communications Technology, USA, pp. 54-61.

coated with a RAM, at 12GHz.

CONCLUSIONS

Currie, N.C., 1989, “Radar Reflectivity Measurement: Techniques and Applications”, 1st Edition, Artech House, Norwood, USA.

This paper presented a simple setup built inside an anechoic chamber (indoor measurements), proper to measure

Dias, J. C., 2000, “Obtenção de Revestimentos Absorvedores de Radiação Eletromagnética (2-18) GHz Aplicados no

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 25-32, Jan. - Mar., 2012

Miacci, M.A.S. et al.

Setor Aeronáutico”, PhD Thesis, Instituto Tecnológico de Aeronáutica, São José dos Campos, SP, Brazil.

Lee, S. M., 1991, “International Encyclopedia of Composites”, Vol. 6, VHC Publishers, New York, USA.

Faez, R. et al., 2000, “Polímeros condutores intrínsecos e seu potencial em blindagem de radiações eletromagnéticas”, Polímeros: Ciência e Tecnologia, Vol. 10, pp. 130-137.

Miacci, M. A. S. et al., 2001a, “Método de Medida de Seção Reta Radar de Objetos Refletores de Ondas Eletromagnéticas na Faixa de 1 GHz a 100 GHz para Caracterização Eletromagnética de Materiais Absorvedores de Radiação Patente”, Brazilian Patent (in analysis), MU8102042-2.

Faez, R. et al., 2005, “Microwave Absorbing Coatings Based on a Blend of Nitrile Rubber, EPDM Rubber and Polyaniline”, Polymer Bulletin (Berlin), Vol. 55, pp. 299-307. Folgueras, L. C. et al., 2007, “Dielectric Microwave Absorbing Material Processed by Impregnation of Carbon Fiber Fabric with Polyaniline”, Materials Research, Vol. 10, No. 1, pp. 95-99. Folgueiras, L.C. et al., 2010, “Dielectric Properties of Microwave Absorbing Sheets Produced with Silicone and Polyaniline”, Materials Research, Vol. 13, No. 2, pp. 197-201. Gama, A.M. et al., 2010, “Complex permeability and permittivity variation of carbonyl iron rubber in the frequency range of 2 to 18 GHz”, Journal of Aerospace Technology and Management, Vol. 2, pp. 59-62. Gama, A.M. et al., 2011, “Dependence of microwave absorption properties on ferrite volume fraction in MnZn ferrite/rubber microwave absorbing materials”, Journal of Magnetism and Magnetic Materials, Vol. 323, pp. 2782 - 2785. Hartman, R. and Berlekamp, J., 1988, “Fundamentals of Antenna Test Evaluation”. MSN & CT – Microwaves Systems News & Communications Technology, USA, pp. 8-20. Johnson, R.N., 1992, “Radar Absorbing Material: A Passive Role in an Active Scenario”, International Countermeasures Handbook, 11th Edition, E.W. Communications, Palo Alto, CA., USA. Knott, E.F. et al., 1993, “Radar Cross Section”. 2nd Edition, Artech House, Inc., Norwood, USA.

Miacci, M. A. S. et al., 2001b, “Radar Cross Section Measurements (8-12GHz) of Flat Plates Painted with Microwave Absorbing Materials”, SBMO/IEEE, Belém, PA., Brazil, Vol. 1, pp. 263. Miacci, M. A. S., 2002, MSc. Dissertation, “Determinação Experimental do Espalhamento Monoestático de Microondas por Alvos de Geometria Simples”, Instituto Tecnológico de Aeronáutica, São José dos Campos, SP, Brazil. Miacci, M. A. S. et al., 2002, “Medidas de Refletividade, Transparência e Seção Reta Radar (RCS) de Compósitos Avançados na Faixa de 8 – 12 GHz”, Brazilian Microwave and Optoeletronics Symposium, Recife, Brazil. Rezende, M.C. et al., 2003, “Reflectivity in the microwave range of polyurethane coating loaded with NiZn ferrites”, Materials Research, Vol. 1, pp. 1-10. Silva, V. A. et al., 2009, “Comportamento eletromagnético de materiais absorvedores de micro-ondas baseados em hexaferrita de Ca modificada com íons CoTi e dopada com La”, Journal of Aerospace Technology and Management, Vol. 1, pp. 255-263. Skolnik, M., 1990, “Radar Handbook”, 2nd Edition, McGraw Hill, USA. Thomson Inc., 1988, “Luneberg Reflectors and Lenses”, Technical Bulletin, September, France.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 25-32, Jan. - Mar., 2012

doi: 10.5028/jatm.2012.04014811

Ammonium Perchlorate and Ammonium PerchlorateHydroxyl Terminated Polybutadiene Simulated Combustion Rene Francisco Boschi Gonçalves*, Koshun Iha, Francisco Bolivar Correto Machado, José Atílio Fritz Fidel Rocco Instituto Tecnológico de Aeronáutica - São José dos Campos/SP - Brazil Abstract: The combustion simulation of ammonium perchlorate was carried out with the software Chemkin, in two steps: the burning behavior of pure ammonium perchlorate and the one of formulated ammonium perchlorate with hydroxyl terminated polybutadiene binder. In both cases, the room pressure varied in order to verify its influence in the system. The burning environment conditions were diverse. During the combustion process, the data obtained from the kinetic chemistry simulation software were compiled. The flame structure can be described by the molar fraction of the burning products and the temperature evolution from the surface of the material. Keywords: Ammonium Perchlorate, Ammonium Perchlorate, Hydroxyl Terminated Polybutadiene, Hydroxyl Terminated Polybutadiene Binder, Chemkin.

INTRODUCTION The first solid propulsion rocket motors date from 13th century and they are assigned to the Chinese (Mark, 2003). The rocket red glare from Spangled Banner of Francis Scott Key was initially used in 1812 in a war involving the United States of America and England. Significant advances in solid propellants technology occurred during the World War II and continue until today. Nowadays, solid propellants are used in satellite launcher vehicles, boosters and sustainer motors in military artifacts propulsion, alteration, and correction engines of geo-stationary satellites orbits, among others. Solid rockets motors are traditionally identified as solid engines. The solid propellant rocket motor is a container filled with solid propellant grain, which, when entering in permanent flow, expels hot gases from the nozzle, generating enough thrust to the system displacement. The solid propellant grain must burn in a pre-determined project speed (burning rate) to maintain the projected pressure in the combustion chamber and it must have enough structural rigidity (mechanical properties) to support the mechanical efforts, during the ignition stages and the burning period. Solid propellants combine next to stoichiometric quantities, both in fuel and in ___________________ Received: 10/10/11. Accepted: 15/12/11 *author for correspondence: renefbg@gmail.com - Pç. Mal. Eduardo Gomes, 50. CEP: 12.228-901 - São José dos Campos/SP - Brazil

oxidizing chemical species, in a solid block called solid propellant grain. These materials are classified as simple-based, doublebased, composite or double-based composite propellants (Pisacane, 2005). Simple-based propellants are active energetic materials separately considered, such as nitrocellulose or nitroglycerin. They are unstable materials and rarely employed in modern solid engines. Double-based propellants are the result of a homogenous mixture of two active ingredients, typically the nitrocellulose and a nitrated plasticizer (energetic), like the nitroglycerin, which dissolves and hardens inside of the uniform solid (grain). Composite propellants have a heterogeneous mixture of fuel species and an oxidizer one like distinct compounds suspended in a polymeric binder, which provides support to the combustion process in ballistic and mechanical behavior terms. Modern solid rocket motors are mostly charged by composite propellants. The mass distribution of their constituents is from 20 to 40% of metallic fuel to increase specific impulse of the propellant, usually aluminum, from 50 to 70% of oxidizer (like ammonium perchlorate – AP) and from 10 to 20% of the binder matrix, main source of carbon, which acts as fuel. Eventually, from 0 to 5% of plasticizer can be added to improve the flexibility and to help during the propellant processing when it is still in the liquid phase (Pisacane, 2005).

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 33-39, Jan. - Mar., 2012

GonĂ§alves, R.F.B. et al.

Examples of typical constituents of solid propellants are presented in Table 1. It is possible to make the synthesis of many solid propellant formulations from the combination of these compounds. The polymeric matrix, named binder, plays the role of binder of all the raw materials that compose the composite solid propellant grain. Molded in the rocket motor case, it passes through a curing process where mechanical and ballistic properties are developed, which are needed to the projected performance of the motor, rocket-motor, in the handling, transport, flight, and storage phases (Monteiro et al., 2007). The metallic additives have the function of stabilizing the burning process against instabilities and of increasing the involved energy during solid grain combustion. AP (NH4ClO4) is a powerful salt oxidizer largely used in solid-propellant formulations for application in airspace and for defense of materials industries. It is obtained by the reaction between ammonia and perchloric acid, or by double decomposition between an ammonium salt and sodium perchlorate, and it is crystallized with romboedric structure in room temperature and pressure, with relative density of 1.95 (Beckstead et al., 2007). Similarly to most ammonium salts, AP thermal decomposition occurs before its fusion. When submitted to a low-heating rate, it decomposes by releasing gases chlorine, nitrogen and oxygen and water in the vapor state; while with a high-heating rate stimulus, there are instant reactions with high energy release. The thermal decomposition mechanism of this

kind of material has not been totally elucidated yet, because of its elevated physical-chemical complexity. However, several studies have been published (Korobeinichev et al., 1990; Boggs, 1970; Beckstead and Puduppakkam, 2004; Jeppson et al., 1997). During the combustion process of AP crystals at high pressures, it is possible to observe the formation of a tiny layer of AP in the liquid phase at the grain surface (Boggs, 1970), followed by a region where gaseous AP is presented. According to Beckstead and Puduppakkam (2004), the combustion of a monopropellant can be divided into three regions (condensed, liquid-gas two-phase, and gas). The two-phase region consists of liquid and gaseous species resulting from the melting and/or decomposition of the solid phase. The precise division between the two-phase and gas-phase region (i.e. the "burning surface") is not welldefined due to chemical reactions, bubbles, and condensed material being convected away from the surface. In the gas phase region of a monopropellant, the flame is essentially premixed. The species emanating from the surface react with each other and/or decompose to form other species. A wide variety of reactions, involving many species, occurs in the gas flame until equilibrium is reached in the final flame zone. A schematic profile of the temperatures developed from the grain surface during the combustion is presented in Fig. 1. This consideration is based on a reactor model, which can be found in the computational package named Chemkin Chemical Kinetics, developed by Sandia Laboratories.

Table 1. Examples of typical constituents of solid propellant formulations. Metallic/ non-metallic fuels

Oxidizers

Aluminum

Ammonium Perchlorate (NH4ClO4)

Beryllium

Lithium Perchlorate (LiClO4)

Magnesium

Potassium Perchlorate (KClO4)

Sodium

Ammonium Nitrate (NH4NO3)

Hydrocarbons

Potassium Nitrate (KNO3)

Binders

Curing Agents

Plasticizers

Carboxyl Terminated

Isocyanates

DOA/DBT

Polybutadiene (CTPB) Epoxydes

Amines/Amides

Triethyleneglicol dinitrate

Hydroxyl terminated

Trimethylethane

polybutadiene (HTPB)

trinitrate

Nitrocellulose Polybutadiene acrylic acid(PBAC) Polybutadiene

Polymers

Sodium Nitrate (NaNO3)

Plastics

Cyclotetramethylenetetranitramine-HMX

Asphalt

Rubbers

Cyclotetramethylenetrinitramine-RDX

PVC

acrylonitrile (PBAN)

J. Aerosp. Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.4, No 1, pp. 33-39, Jan. - Mar, 2012

Ammonium perchlorate and ammonium perchlorate-hydroxyl terminated polybutadiene simulated combustion Table 2 . Decomposition reactions of ammonium perchloratea.

Figure 1. Scheme of the ammonium perchlorate crystal combustion process with evolution of the temperatures from the material surface.

Due to the difficulty of performing experiments with these materials in real combustion conditions, it is common to employ computational modeling techniques to estimate the burning behavior of solid propellants. A widely used software for the study of the flame formed during energetic materials combustion is Chemkin. This software considers the equations of mass and energy conservation, chemical species involved and linear momentum and it enables the analysis through temperature and chemical species molar fractions variation in function of the distance from the material surface, among others. The present paper treats the combustion study of pure AP and formulated in a solid composite propellant grain, focusing in the structure of the flame formed in this process.

Reaction

HClO4=ClO3+OH HClO4+HNO=ClO3+H2O+NO

1,00E+14 1,50E+13

0.0 0.0

39100.0 6000.0

ClO3=ClO+O2

1,70E+13

0.5

0.0

Cl2+O2+M=ClO2+Cl+M

6,00E+08

0.0

11200.0

ClO+NO=Cl+NO2

6,78E+12

0.0

311.0

ClO+ClOH=Cl2+HO2

1,00E+11

0.0

10000.0

ClOH+OH=ClO+H2O

1,80E+13

0.0

HCl+OH=Cl+H2O

5,00E+11

0.0

750.0

Cl2+H=HCl+Cl

8,40E+13

0.0

1150.0

ClO+NH3=ClOH+NH2

6,00E+11

0.5

6400.0

NH3+Cl=NH2+HCl

4,50E+11

0.5

100.0

NH3+OH=NH2+H2O

5,00E+07

1.6

955.0

NH2+O2=HNO+OH

3,00E+09

0.0

NH2+NO=H2O+N2

6,20E+15

-1.3

0.0

HNO+OH=NO+H2O

1,30E+07

1.9

-950.0

HNO+O2=NO2+OH

1,50E+13

0.0

10000.0

HNO+H=H2+NO

4,50E+11

0.7

660.0

NO+H+M=HNO+M

8,90E+19

-1.3

740.0

HO2+N2=HNO+NO

2,70E+10

0.5

41800.0

NO+HO2=NO2+OH

2,11E+12

0.0

480.0

H+NO2=NO+OH

3,47E+14

0.0

1480.0

H2+OH=H2O+H

2,16E+08

1.5

3430.0

k=A Tb exp(-E/RT). Units: A (mol-cm-s-K), E (J/mol); a

kinetic data composed from Korobeinichev et al. (1990).

EXPERIMENTAL PART The elementary reactions that compose the burning mechanism of AP are listed on Table 2. This mechanism was proposed by Gross (2007), according to data obtained from the literature. In the used simulation module, it is necessary the insertion of a temperature profile. This profile lists the developed temperature in the material burning in function of the distance from the solid surface. The data can be observed in Fig. 2. The present profile was obtained from the software’s database, and it is a good approach to the solid fuel used in the study. As this profile was developed for 1 atm, temperatures from 2,000 to 3,200K are expected for the simulations.

Figure 2. Temperature profile used in the simulations.

For the experiments, the system pressure of the combustion chamber varied, in order to study its influence in the flame structure; it was simulated the combustion in atmosphere composed by the initial products of AP decomposition (pure), with composition and

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 33-39, Jan. - Mar., 2012

Gonçalves, R.F.B. et al.

concentration obtained from literature (Beckstead and Puduppakkam, 2004). A study of AP formulated with hydroxyl terminated polybutadiene (HTPB) binder was also carried out, whose mechanism can be observed in Table 3. This mechanism was based on Gross' work (Korobeinichev et al., 1990), according to data obtained from the literature. The initial concentration of the species inside the chamber was set according to an AP-HTPB rate of 30/70. Table 3 . Decomposition reactions of ammonium perchloratea.

Reaction

Cl+HO2=ClO+OH ClO+O=Cl+O2

2,47E+13 6,60E+13

0 0

894 440

H+HCl=Cl+H2

7,94E+12

3400

HCl+O=Cl+OH

2,30E+11

0.6

900

Cl2+O=Cl+ClO

2,51E+12

2720

N2O+M=N2+O+M

6,20E+14

56100

N2O+OH=N2+HO2

2,00E+12

21060

N2O+O=NO+NO

2,90E+13

23150

N2O+O=N2+O2

1,40E+12

10810

N2O+H=N2+OH

4,40E+14

18880

2H+M<=>H2+M

1,00E+18

-1

Reaction

-0.6

6,00E+08 6,78E+12

0 0

11200 311

2H+H2<=>2H2

9,00E+16

Cl2+O2+M=ClO2+Cl+M ClO+NO=Cl+NO2

2H+H2O<=>H2+H2O

6,00E+19

-1.3

HCl+OH=Cl+H2O

5,00E+11

750

2H+CO2<=>H2+CO2

5,50E+20

-2

Cl2+H=HCl+Cl

8,40E+13

1150

ClO2+NO=ClO+NO2

1,00E+11

NH3+Cl=NH2+HCl

4,50E+11

0.5

100

Cl+ClO2=ClO+ClO

5,00E+13

6000

NH3+OH=NH2+H2O

5,00E+07

1.6

955

ClO+ClO=Cl2+O2

1,00E+11

NH2+O2=HNO+OH

3,00E+09

NH2+NO=H2O+N2

6,20E+15

-1.3

HNO+OH=NO+H2O

1,30E+07

1.9

-950

HNO+O2=NO2+OH

1,50E+13

10000

HNO+H=H2+NO

4,50E+11

0.7

660

NO+H+M=HNO+M

8,90E+19

-1.3

740

HO2+N2=HNO+NO

2,70E+10

0.5

41800

NO+HO2=NO2+OH

2,11E+12

480

H+NO2=NO+OH

3,47E+14

1480

H2+OH=H2O+H

2,16E+08

1.5

3430

CH4+Cl=CH3+HCl

2,50E+13

3830

CH4+H=CH3+H2

6,60E+08

1.6

10840

CH4+OH=CH3+H2O

1,00E+08

1.6

3120

CH3+H+M=CH4+M

1,27E+16

-0.6

383

CO+OH=CO2+H

4,76E+07

1.2

CO+ClO=CO2+Cl

3,00E+12

1000

CO+ClO2=CO2+ClO

1,00E+10

H+O2=O+OH

8,30E+13

14413

CH2+H2=CH3+H

5,00E+05

7230

CH2+H+M=CH3+M

2,50E+16

-0.8

CH4+O=CH3+OH

1,02E+09

1.5

600

OH+CH3=CH2+H2O

5,60E+07

1.6

5420

C2H4+O2=2CO+2H2

1,80E+14

35500

NH2+NO2=2HNO

1,40E+12

NH2+ClO=HNO+HCl

2,50E+12

O2+HNO=NO+HO2

1,00E+13

13000

H+Cl+M=HCl+M

5,30E+21

-2

-2000

Cl+Cl+M=Cl2+M

3,34E+14

-1800

Cl+HO2=HCl+O2

1,80E+13

Cl+O2+M=ClO2+M

8,00E+06

5200

NO2+O=NO+O2

1,00E+13

600

HNO+HNO=H2O+N2O

3,95E+12

5000

NO2+NO2=NO+NO+O2

1,00E+14

25000

Cl+N2O=ClO+N2

1,20E+14

33500

OH+OH=H2O+O

6,00E+08

1.3

NH2+NO2=H2O+N2O

4,50E+11

HNO+NH2=NH3+NO

5,00E+11

0.5

1000

ClO+HNO=HCl+NO2

3,00E+12

HCl+HO2=ClO+H2O

3,00E+12

NH2+NO=H+N2+OH

6,30E+19

-2.5

1900

NH2+OH=H2O+NH

4,00E+06

1000

NH2+NH2=NH+NH3

5,00E+13

10000

NH+NO=N2+OH

1,00E+13

NH+NO=H+N2+O

2,30E+13

Cl+NH2=HCl+NH

5,00E+10

0.5

ClO2+NH=ClO+HNO

1,00E+14

N+NO2=NO+NO

1,00E+14

N+N2O=N2+NO

5,00E+13

NH+OH=H2O+N

5,00E+11

0.5

2000

NH+OH=H2+NO

1,60E+12

0.6

1500

NH+NH2=N+NH3

1,00E+13

2000

HO2+CH3<=>O2+CH4

1,00E+12

CH2+CH4<=>2CH3

2,46E+06

8270

k=A Tb exp(-E/RT). Units: A (mol-cm-s-K), E (J/mol); M: any metal surface or metallic additive used only as support or catalyst; a

kinetic data composed by Korobeinichev et al., (1990).

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 33-39, Jan. - Mar, 2012

Ammonium perchlorate and ammonium perchlorate-hydroxyl terminated polybutadiene simulated combustion

Similarly, in AP-HTPB simulation, the pressure of the combustion chamber was varied in order to study its influence in the flame structure. The atmosphere of the chamber was composed by the initial products of AP and HTPB binder decomposition.

noticed. Figure 4 shows the combustion simulation of the material at 5 atm pressure.

RESULTS AND DISCUSSION In AP combustion simulation process at pressure of 1 atm (Fig. 3), it can be noticed that the curves have exponential profiles, according to the equation proposed by Arrhenius, and also the flame must be presented in the region from 2 to 3mm of the burning material surface. There is a significant decay of the molar fractions of water and chlorine, while there is an increase on the mole fractions of chloridric acid, nitrogen, and oxygen. From this result, it was possible to observe that the combustion of AP in a combustion chamber is not complete, due to the reduced permanence time on it; therefore, there is the formation of a great quantity of chloridric acid and there is no complete consume of the oxygen. According to Cai et al., (2008), during the decomposition of AP, the formation of nitrogen is expected.

Figure 3. Ammonium perchlorate decomposition under pressure of 1 atmosphere.

The nitrogen usually presents as an inert gas, although in the system temperature (~1,700K), it participates as the mechanism, in reactions with AP decomposition products. However, the decomposition of ammonia in gaseous nitrogen is faster than the reactions of N2 with other species; therefore, an increase on the molar fraction is observed. As the pressure on the combustion chamber rises, significant differences on the combustion process of AP were

Figure 4. Ammonium perchlorate decomposition under pressure of 5 atmospheres.

The curves are presented with exponential profiles, although there was a displacement of the flame region, now presented at 1 to 2mm from the burning surface. As the pressure increases, the speed of the elementary reactions presented in AP decomposition mechanism rises, therefore, this may be the main cause of the approximation of the flame to the material. It was also observed behavior to the species in the flame region (presence of "elbows"), in which there is decrease and posterior increase of the molar fraction (for HCl) and increase and posterior decrease of the molar fraction (for water and Cl2). As the AP acts as an oxidizer, the increase of oxygen concentration is expected. The mass balance of the species is maintained from the beginning to the end of the simulation. The “elbows” appear due to the increase of the occurrence of intermediate reactions in the flame zone. This phenomenon generates a great variation of intermediates mole fractions, which modify the concentration of the main species (especially in the flame zone), so the different slope is observed. Figures 5 and 6 present new AP decomposition simulations for pressures of 30 and 60 atmospheres, respectively. It is observed that, as the pressure in the combustion chamber increases, there is an approximation of the flame to the surface of the material and an accentuation of the "elbows" presented on the flame region, indicating the influence of the speed increase of elementary reactions in the decomposition process of the material in study. Such gain in chemical speed reactions may be converted in thrust of rocket motor and specific impulse of solid propellant grain. So, this kind of simulation is interesting in order to know the behavior of one or more

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 33-39, Jan. - Mar., 2012

GonĂ§alves, R.F.B. et al.

materials, internally to a rocket motor combustion chamber, to develop the optimum propellant formulation, i.e., the one that generates higher thrust, specific impulse, and fewer residues.

Figure 5. Ammonium perchlorate decomposition under pressure of

50atm, in which a separation between the model calculations and experimental data becomes increasingly apparent. At low pressures (1 to 34atm), the propellant is more likely to burn as a premixed flame with the oxidizer and binder decomposition products from the condensed phase, mixing completely before the gas phase reactions occur. As the pressure increases, the diffusion of the fuel products into the reacting stream above the oxidizer is disrupted and the gas phase flame becomes increasingly diffusion-limited. Thus, at pressures above 40 to 50atm, the kinetic model, which assumes completely premixed combustion, would be expected to begin to yield calculations in excess of the measured burn rate. Independently of the pressure in the chamber, the packing and the molecular distribution of the material are not going to suffer alteration and are not going to lead to pressure variations in their behavior, proving the limitations of the model.

30 atmospheres.

Figure 7. Ammonium perchlorate HTPB binder decomposition. Figure 6. Ammonium perchlorate decomposition under pressure of 60 atmospheres.

Beyond the simulations of pure AP, simulations were carried out with AP formulated with HTPB, to approximate the maximum possible to a real-solid propellant formulation, because HTPB polymer matrix is a binder largely used in formulations. Figure 7 presents AP-HTPB decomposition. The combustion process of AP-HTPB has been invariable with pressure. This behavior should be attributed to the homogeneous dispersion of AP amidst the binder, in the solid phase, and to the lack of this species in relation to the binder (generating lower concentrations of O2 than necessary). Also, in the gas phases, it is assumed that all of the liquid AP and HTPB present on the condensed phase decompose to form gaseous species; evaporation is not included. According to Jeppson, Beckstead and Jing (1997), the kinetic model accurately calculates the flame structure up to about 40 to 38

In this simulation, the oxygen molar fraction suffers a decrease (and cancels), according to the reactions with HTPB decomposition products, for the formation of carbon monoxide and dioxide. Furthermore, it is interesting to highlight that in this case the carbon monoxide molar fraction suffers a decrease, because the restriction of oxidizer species makes the oxygen presented in CO to be also used as oxidizing source, viewing the reactive behavior of this species. In this simulation, the molar fractions of CO and CO2 are not null initially, given the system temperature, HTPB suffers an initial decomposition that should not be discarded, generating both carbon oxides.

CONCLUSIONS The AP decomposition process was simulated in different situations using the computational package Chemkin,

J. Aerosp. Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.4, No 1, pp. 33-39, Jan. - Mar, 2012

Ammonium perchlorate and ammonium perchlorate-hydroxyl terminated polybutadiene simulated combustion

Premix module. With the reactions involved in the proposed mechanisms, the variation of the AP combustion behavior was observed as the pressure of the combustion chamber of a solid propellant rocket motor was varied. Based on the results, the profiles of the flames originated from the AP and AP-HTPB combustion were modeled, presenting the different temperature regions and the molar fraction variation of chemical species, generated by the decomposition of the materials. This study contributes to understand the decomposition process, i.e. how the species involved behaves during the combustion and with variations of different parameters. The procedure adopted may be of use with new formulations of propellants, avoiding costly and long tests.

REFERENCES Beckstead, M.W. and Puduppakkam, K.V., 2004, “Modeling and Simulation of Combustion of Solid Propellant Ingredients Using Detailed Chemical Kinetics”, 40th AIAA/ASME/ SAE/ASEE Joint Propulsion Conference and Exhibit. Beckstead, M.W. et al., 2007, “Progress in Energy and Combustion Science”, Vol. 33, pp. 497-551. Boggs, T.L., 1970, “Deflagration Rate, Surface Structure and Subsurface Profile of Self-Deflagrating Single Crystals of Ammonium Perchlorate”, AIAA Journal, Vol. 8, No. 5, pp. 867.

Cai, W. et al., 2008, “A Model of AP/HTPB Composite Propellant Combustion in Rocket Motor Environments”, Combustion Science and Technology, Vol. 180, pp. 21432169. Gross, M.L., 2007, “Two-dimensional modeling of AP/HTPB utilizing a vorticity formulation and one-dimensional modeling of AP and ADN", Thesis presented at Brigham University. Jeppson, M. B. et al., 1997, “A Kinetic Model for the Premixed Combustion of a Fine AP/HTPB Composite Propellant”, American Institute of Aeronautics and Astronautics. Korobeinichev, O. et al., 1990, AIAA/SAE/ASME/ASEE 26th Joint Propulsion Conference, Orlando-USA. Mark, H., 2003, “Encyclopedia of Space Science and Technology”, Vol. 2, Wiley Interscience, 531p. Monteiro, R.R. et al., 2007, “Vulnerabilidade de motorfoguete a propelente sólido em relação ao impacto balístico (Arma de Fogo)” – IX Simpósio Interno de Guerra Eletrônica (SIGE), São José dos Campos, São Paulo, Brazil. Pisacane, V.L., 2005, “Fundamentals of Space Systems”, 2th ed., Oxford University Press, 206p.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 33-39, Jan. - Mar., 2012

doi: 10.5028/jatm.2012.04016011

Water Influence in Poly(epichlorohydrin) Synthesis: An Intermediate to Energetic Propellants Jairo Sciamareli*,1, Silvana Navarro Cassu1, Koshun Iha2 Instituto de Aeronáutica e Espaço - São José dos Campos/SP - Brazil Instituto Tecnológico da Aeronáutica - São José dos Campos/SP - Brazil

1 2

Abstract: Poly(epichlorohydrin) was synthesized using anhydrous and hydrate stannic chloride, separately, as initiator. Reactions were conducted in the presence of a diol at 70º C. Polymers were characterized by Fourier Transform Infrared Spectroscopy (FT-IR), differential scanning calorimetry (DSC), gel permeation chromatography (GPC), and hydroxyl value determination. It was observed that water has influence in growing chain, but not in its structural backbone. Besides, the monomer/initiator relationship has a higher influence in overall poly(epichlorohydrin) properties. Keywords: Synthesis, Poly(epichlorohydrin), Hydrate Initiators.

INTRODUCTION Poly(epichlorohydrin) (PECH) has been used to prepare glycidyl azide polymer (GAP), a worthy energetic material that can be used alone or associated with other polymers in copolymers, as energetic binder, to improve mechanical and ballistics properties in rocket propellants (Eroglu and Guven, 1996; Frankel et al., 1992; Mojan et al., 2004; Sciamareli, 2009a). In a general way, better mechanical properties in propellants are obtained when the polymer has a molecular weight between 2000 and 3000g/mol and hydroxyl value is up to 1.0mmol/g. Important GAP properties, as molecular weight, hydroxyl value and functionality, are obtained in PECH synthesis, since the molecular structure is preserved during the PECH to GAP transition (Sciamareli, 2009b). Polymerization of epichlorohydrin (ECH) proceeds by cationic mechanism in presence of Lewis acid like boron trifluoride etherate and stannic chloride. Diols, such as butane diol, act as co-initiators or chain transfer agents. The resulting polymer has hydroxyl terminal groups (Biedron et al., 1991; Francis et al., 2003). In this kind of experiment, water can act as an initiator in polymerization reaction and, further, as a deactivating agent, introducing the OH group at the end of the final product ___________________

(Biedron, 1991; Guanaes et al., 2007; Okamoto, 1984; Qureshi and Ochel, 1996). The aim of this work was to investigate the influence of water and monomer/initiator relationship in PECH synthesis. In these experiments, it was used, separately, anhydrous and hydrate initiator, respectively, SnCl4 and SnCl4.5H2O, and their results were compared. The experimental procedure was the same and reagents were similar, with the exception of the initiator.

EXPERIMENTAL Materials ECH, 1,2 dichloroethane and trifluoroacetic acid (Sigma). Anhydrous stannic chloride (Riedel). Pentahydrate stannic chloride (Vetec), as well ethylenediaminetetracetic tetrasodium salt hydrate (EDTA tetrasodium) and acid hydrochloride. Ammonium hydroxide (J. T. Baker). All reagents were used as received, without treatment. Only 1,4 butanediol (Sigma) was treated with calcium sulphate before being purified by vacuum distillation. Synthesis of PECH

Received: 21/12/11. Accepted: 06/01/12 *author for correspondence: jairojs@iae.cta.br/Pç. Mal. Eduardo Gomes, 50. CEP: 12.228-901 - São José dos Campos/SP - Brazil

Reaction of the ring-opening polymerization of ECH was carried out in a 300mL three-neck round bottom-flask,

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 41-44, Jan. - Mar., 2012

Sciamareli, J. et al.

containing a magnetic stir bar and a dropping funnel. 1,4 butanediol and 1,2 dichloroethane were stirred to obtain homogeneous solution. First, stannic chloride was introduced to round bottom-flask and, in sequence, trifluoroacetic acid was added. The system was warmed by an oil bath up to 70oC. ECH was added, drop by drop, using the dropping funnel at a rate of 30g per hour. When the ECH addition was finished, the system was warmed for another hour. The reaction was quenched by adding distilled water and acid hydrochloride. After being stirred for ten minutes, solution was transferred to a separatory funnel. Separated organic phase was stirred again with a water solution containing EDTA tetrasodium and ammonium hydroxide. One more time, resulted solution was transferred to a separatory funnel and, after being separated, organic phase was washed with distillated water. This last part was repeated for another four times. At last, polymer was obtained by removing the solvent by vacuum distillation.

Spectra in Fig. 2 are very similar too. In a general way, all the spectra show resemblance aspect. They present the same absorptions at the same position, in evidence that water does not up set structural backbone formation of PECH.

cm-1

Analysis of the polymers

Figure 1. PECH spectra with monomer∕initiator relationship of 20:1.

FT-IR analyses were performed by PerkinElmer spectrophotometer, Spectrum 2000 in the spectral region of 4000 to 400 cm-1, gain 1, resolution of 4 cm-1 and 40 scans. Samples were analyzed by transmission technique as liquid film. Gel permeation chromatography analyses were conducted by a PerkinElmer Series 200 equipped with NUCLEOGEL GPC 103-5 VA300/7.7 column, tetrahydrofuran was the mobile phase and it was calibrated with polystyrene standards. Thermal analyses were carried out in a DSC TA Instruments Q100, from -90 to 40ºC, at 20ºC/min heating rate, in nitrogen atmosphere, using 2-3mg of sample. CH analyses were recorded by a PerkinElmer 2100 II, at 925ºC under oxygen and helium flow using 2-3mg of sample. OH analyses were made according to open literature (Dee et al., 1980).

RESULTS AND DISCUSSION This work was carried out with a monomer/initiator relationship of 20:1 and 60:1. Figure 1 compares FT-IR spectra obtained to 20:1 and Fig. 2 compares spectra to reaction with 60:1 relationship, where (A) is a spectrum relative to use SnCl4.5H2O and (B) is relative to use SnCl4 anhydrous. Spectra in Fig. 1 are very similar. Both present the characteristics PECH peaks at 3470-3442 cm-1, 1119-1115cm-1 and 749-747cm-1, corresponding to OH, C−O−C and CH2Cl functional groups, respectively (Silverstein et al., 1981). 42

cm-1

Figure 2. PECH spectra with monomer/initiator relationship of 60:1.

According to results in Table 1, when it is compared to the action of anhydrous and hydrate initiator, it is possible to realize a clearly tendency to higher molecular weight when anhydrous SnCl4 is used. When it is compared to the quantity of the initiator, it is possible to observe that higher concentration of this reactant also carries out to higher molecular weight. This is reported elsewhere (Sciamareli, 2009b). Anyway, these results were expected, since it is known that water act as a chain end, avoiding their enlargement. To the same quantity of monomer, shorter chain means a higher number of chains with lower size. Furthermore, it shows the importance of concentration of initiator in reaction.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 41- 44, Jan. - Mar, 2012

Water Influence in Poly(epichlorohydrin) Synthesis: An Intermediate to Energetic Propellants Table 1. PECH molecular weight, glass transition temperature (Tg) and hydroxyl value Relationship monomer∕initiator 20:1

60:1

Initiator

Molecular weights(Mn)(g.mol-1)

Mw/Mn

Tg (ºC)

Hydroxyl value(mmol/g)

SnCl4

2974

1.09

-37

0.73 (0.90)

SnCl4.5H2O

1625

1.11

-37

1.12 (1.24)

SnCl4

1417

1.20

-41

1.16 (1.24)

SnCl4.5H2O

1033

1.18

-42

1.19 (1.29)

The effect of initiator content on polymer glass transition temperature is clearly observed by DSC curves, which are showed in Fig. 3. Higher initiator content causes an increase of polymer Tg due to higher molecular weight. This effect can be observed as the critical molecular weight was not attained in these polymers. The polymer obtained with the hydrate and anhydrous initiators showed almost the same glass transition temperature, suggesting that water has no influence in glass transition. These results are summarized in Table 1.

and results to hydroxyl value to this new synthesis process are exhibited between parentheses, with the same tendency. Results from elemental analysis of carbon and hydrogen showed close values, where, practically, there is no difference to the same elements. Table 2 exhibits these results.

CONCLUSIONS Water exert influence in PECH synthesis, carrying out to a lower molecular weight and higher hydroxyl content, but, without altering its backbone structure. Both PECHs, made with hydrate and anhydrous initiator, present the same carbon and hydrogen contents. However, the monomer/initiator relationship deeply alters properties as molecular weights and Tg. In a general way, PECH, made with hydrate initiators, presents lower molecular weight and higher hydroxyl value of the ideal range to use in synthesis of GAP.

REFERENCES Figure 3. DSC curves for PECH obtained with different catalyst contents.

In Table 1, we can observe that hydroxyl values seem to be directly related to molecular weights, independent of the type of initiator used, hydrate or anhydrous. The rule is: lower molecular weights, higher hydroxyl values. To exclude any doubts, PECH was synthesized again in the same conditions,

Biedron, T. et al., 1991, “Polyepichlorohydrin Diols Free of Cyclics: Synthesis and Characterization”, Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 29, No. 5, pp. 619-628. Dee, L.A. et al., 1980, “N-Methylimidazole as a Catalyst for Acetylation of Hydroxyl Terminated Polymers”, Analytical Chemistry, Vol. 52, No. 4, pp. 572-573.

Table 2. Percentage of carbon and hydrogen in samples of PECH. Relationship monomer/initiator 20:1

60:1

Carbon (%)

Hydrogen (%)

SnCl4

Initiator

38.3

5.7

SnCl4.5H2O

38.7

5.9

SnCl4

38.6

5.4

SnCl4.5H2O

38.4

5.8

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 41-44, Jan. - Mar., 2012

Sciamareli, J. et al.

Eroglu, M. S., Guven, O., 1996, “Spectroscopy and Thermal Characterization of Poly(glycidyl azide) Converted from Polyepichlorohydrin”, Journal of Applied Polymer Science, Vol. 60, No. 9, pp. 1361-1367.

Okamoto, Y., 1984, “Cationic Ring-Opening Polymerization of Epichlorohydrin in the Presence of Ethylene Glycol’, Polymer Preprints, Vol. 25, No. 1, pp. 264-265.

Francis, A. U. et al., 2003, “Structural Characterization of Hydroxyl Terminated Polyepichlorohydrin Obtained Using Boron Trifluoride Etherate and Stannic Chloride as Initiators”, European Polymer Journal, Vol. 39, No. 4, pp. 831-841.

Qureshi, M. Y., Ochel, M., 1996, “Synthesis and Characterization of High Molecular Weight Poly (Trimethylene Oxide)”, European Polymer Journal, Vol. 32, No. 6, pp. 691-693.

Frankel, M. B. et al., 1992, “Historical Development of Glycidyl Azide Polymer’, Journal of Propulsion and Power, Vol. 8, No. 3, pp. 560-563.

Sciamareli, J. et al., 2009a, “Síntese e Caracterização do Polímero Energético Metil Azoteto de Glicidila (GAP) Via Análises Instrumentais”, Polímeros: Ciência e Tecnologia, Vol. 19, No. 2, pp.117-120.

Guanaes, D., et al. 2007, “Influence of Polymerization Conditions on the Molecular Weight and Polydispersity of Polyepichlorohydrin”, European Polymer Journal, Vol. 43, No. 5, pp. 2141-2148. Mohan, Y.M. et al., 2004, “Synthesis, Spectral and DSC Analysis of Glycidyl Azide Polymers Containing Different Initiating Diol Units”, Journal of Applied Polymer Science, Vol. 93, No. 5, pp. 2157-2163.

Sciamareli, J., et al. 2009b, “Otimização do Processo de Obtenção do Pré-Polímero Metil Azoteto de Glicidila”, Journal of Aerospace Technology and Management, Vol.1, No. 1, pp. 29-34. Silverstein, R.M. et al.,1981, “Spectrometric Identification of Organic Compounds”, John Wiley & Sons Inc., New York, USA.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 41- 44, Jan. - Mar, 2012

doi: 10.5028/jatm.2012.04015211

An Analysis of the Initiation Process of Electro-explosive Devices Paulo Cesar de Carvalho Faria*, Koshun Iha, José Atílio Fritz Fidel Rocco Instituto Tecnológico de Aeronáutica - São José dos Campos/SP - Brazil Abstract: Electro-explosive devices (an electric resistance encapsulated by a primary explosive) fundamentally convert electrical energy into thermal energy, to start off an explosive chemical reaction. Obviously, the activation of those devices shall not happen by accident or, even worse, by intentional exogenous influence. From an ordinary differential equation, which describes the electro-explosive thermal behavior, a remarkable, but certainly not intuitive, dependence of the temperature response on the time constant of the heat transfer process is verified: the temperature profile dramatically changes as the time constant spans a wide range of values, from much lesser than the pulse width to much greater than the pulse period. Based on this dependence, important recommendations, concerning the efficient and safety operation of electro-explosive devices, are proposed. Keywords: Electro-Explosive Devices, Squib, Thermal Model, Pulsed Initiation, Temperature Stacking.

LIST OF SYMBOLS a = 1/τEE ave

Inverse of the EED time constant Average

(s-1) (---)

Time

(s)

Pulse period

(s)

Ignition temperature threshold

(---)

And

(---)

Barrel

(---)

THRESHOLD

EED thermal capacitance

(J.ºC-1)

Duty-Cycle

(---)

Electro-explosive

(---)

DC = PW/T H

Height of the water column inside the barrel

(m)

γ η θ

Exponent

(non-linear

hydraulic

resistance)

(---)

Operational efficiency (temperature gain) (---) EED temperature increment (above ambient temperature)

(ºC)

Height of the barrel

(m)

Input

(---)

≈3.1415

(---)

Cycle counting (pulsed excitation)

(---)

Water density

(kg.m-3)

Output

(---)

τb

Barrel time constant

(s)

Pave

EED time constant

(s)

Average power

(W)

Pi (t)

EED electrical input power

(W)

Peak power

(W)

Pulse width

(s)_

Input volumetric flow rate

(m3.s-1)

Output volumetric flow rate

(m3.s-1)

Radius of the barrel

(m)

R RT V

Hydraulic resistance Thermal resistance

(m-2.s) (ºC-1.W)

Volume of water inside the barrel

(m3)

___________________ Received: 04/11/11. Accepted: 23/01/12 *author for correspondence: carvalho@ita.br/Pç. Mal. Eduardo Gomes, 50. CEP: 12.228-900 - São José dos Campos/SP - Brazil

τEE

INTRODUCTION In order to study the ignition process of electro-explosive devices (EEDs), a consistent model is needed, one capable of coherently reproducing the temperatures of EDDs in response to the electrical power applied to them. In fact, EEDs merely convert energy from one type to another, which means that their electro-thermal model is a direct consequence of the energy conservation principle. There is a complete theory behind such model and it is founded on Rosenthal’s (1961) and Prince & Leeuw’s (1988) work. Hoberman (1965) and Potter & Scott (2004) developed

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 45-50, Jan. - Mar., 2012

Faria, P.C.C. et al.

a comprehensive study of this matter too. However, as the focus here is on the application of the model, not in the theory sustaining it, only the necessary level of abstraction will be retained from the original effort of those researchers and authors. From the solution of that model, an outstanding advantage, in terms of temperature gain, will be noticed in favor of the electrical pulsed excitation (time to ignition not been a problem) whenever the width and period of the train of pulses are adjusted, taking into consideration the EED thermal time constant τEE . Thus the choice of the most appropriate power source (continuous or pulsed) to drive a specific EED will be primarily affected by its thermal time constant. Accordingly, the thorough knowledge of this constant will be essential for reliable and risk free (secure) applications of EEDs. In this way, the steps to be followed to succeed in reaching the analysis of the EED ignition are: to adopt a simplified version of the currently used electro-thermal model of EEDs (an ordinary differential equation – an ODE, for short); to solve that model for different sorts of excitation (continuous and pulsed); and to verify the strong influence of both the EED (thermal time constant) and power source (pulse width and period of the train of pulses) parameters in the EED temperature response. It follows that the key objective of the analysis to be carried out later in this paper is to guide users of EEDs, when faced with the choice of the right source of power to the right squib (vice-versa). But, before analyzing the electro-explosive as an energy converter device, the investigation of a topic much more familiar to everyone, the barrel problem (Moran and Shapiro, 2008), can be valuable.

A CONVENIENT ANALOGY Consider a barrel (a cylinder with transversal area, πr2, height, Hb, and top surface removed), initially empty, having a small hole in its base. Assume a given steady water flow into the barrel. Naturally, a question will arise: how is the level of the water inside the barrel going to evolve as a function of time? To facilitate the analysis of this problem, let the hole at the barrel base be modeled by a linear hydraulic resistance ( 1 - c) R (actually, R = R0 $ H , with γ adjusted in accordance with each specific situation (Cochin, 1980), and let the water flux through the hole, Qo, be proportional to H : 46

H = H H Q0 = (1 - c) = R R0 R0 .H

c=1

= H R0

(1)

On the other hand, the water volumetric flow into the barrel is constant and equal to Q. Now, making use of the facts that, 2 from the cylindrical geometry, dV/dt = rr $ dH/dt , and, from the mass-rate balance principle, t $ dV/dt = t $ (Qi - Qo) , the ODE describing the barrel problem is dH + H = Qi 2 xb dt rr

(2)

^x b = R0 .rr2h t

This ODE has H = Qi $ R0 $ (1 - e xb ) for solution (t > 0), from which the necessary condition for water to overflow is Qi > Hb /R0 . But what barrels have to do with EEDs? Next

section comes out to unveil this seeming mystery.

THE ANALYSIS OF THE EED THERMAL MODEL By coating an electric resistance with a thick layer of high sensitivity explosive and applying an electric current to this circuit (bridgewire), the tiny explosive mass can certainly be initiated, but what is the best way to do that, by continuous or pulsed excitation? To answer this question, a simplified representation of the thermal phenomena involved in the EED ignition will be depicted. Applying the energy-rate balance principle to the EED, the injected energy rate must be equal to the stored and the dissipated energy rates added together or, as in Eq. 3, di + i di + i = Pi & Pi = CT . x EE dt RT dt CT = (x EE RT .CT )

(3)

The electro-thermal model corresponding to this differential equation is given in Fig. 1 (for experimental procedures to determine x EE = Rt $ Ct , see Rosenthal’s (1961) and Prince and Leeuw’s (1988)). Hence, the EED temperature behaves exactly as the level of the fluid accumulated inside the barrel, since differential Eq. 2 and Eq. 3, which represent both phenomena, hydraulic and thermal, are analogous. As a matter of fact, this similarity will be very useful in grasping the subtleties of the heat transfer process involved in the EED ignition. As it has already been shown, - t

i (t) = Pi .RT . (1 - e x EE ) (4)

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 45-50, Jan. - Mar, 2012

An Analysis of the Initiation Process of Electro-explosive Devices

method should be used (Wylie, 1975). Here, one is going to derive an approximated solution in the time domain. From the theory of differential equations (Wylie, 1975), the total temperature can be factored into transient (the transient part is always negative and steadily approaches zero) and periodic components. However, after a large number of pulses (n >> 1 , n is the pulse-cycle counting), the transient portion will have faded into insignificance (Fig. 3).

Figure 1. EED model.

Equation 4 is the solution of Eq.3, where Pi is constant. Therefore, the EED temperature is driven by the injected power. From Eq.4, with t→∞, the maximum temperature (continuous input) is given by i MAXCONT = Pi $ RT = Pave $ RT

(5)

Note how the EED temperature, due to the heat dissipation to the environment, via RT , does not increase indefinitely – it is bounded (Fig. 2). Nevertheless, it suffices to make θMAXCONT > θTHRESHOLD , by adjusting Pi, for ignition to occur.

Figure 3. EED temperature after the settle down of the transient phase (pulsed excitation).

The problem was thus reduced to solve the following system of two ODEs with their associated boundary conditions (considering only the periodic component):

Z di PP i ] dt + x = C , 60 # t # PW @ / 6i (0) = i1, i (PW ) = i2 @ EE T ] [ (6) ]] di + i = 0, 6 PW # t # T @ / 6i (PW ) = i2, i (T) = i1 @ \ dt x EE Figure 2. Signal flow graph (SFG) corresponding to Eq. 3 – the integrating effect of the thermal capacitance (CT) is counterbalanced by the heat conduction to the surroundings ( 1/RT).

What happens when Pi(t) is a train of power pulses (a switched power supply, for instance)? Despite the incontestable simplicity of the model adapted to our purposes, in this case, a certain mathematical expertise was required to solve it. For the complete solution of the EED ODE with a pulsed forcing function, the Laplace transform

From Eq. 6, it is not difficult to show that θ1 and θ2 must satisfy Eq. 7:

Z - PW - PW ] ] i 2 - ^ e x EE h .i1 = PP .RT .^1 - e x EE h [ ]] ^e (Tx-EEPw) h .i - i = 0; 2 1 \

(7)

Solving Eq. 7 for θ2 = θMAXPULS (maximum temperature): PW

1 - e- x EE (8) i MAXPULS = i2 = PP .RT . e T o 1 - e- x EE

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 45-50, Jan. - Mar., 2012

Faria, P.C.C. et al.

Equation 8, with Pave = Pp $ DC = Pp $ (Pw /T) fixed, becomes Eq. 9: T

.DC 1 1 - e- x EE i MAXPULS = i MAXCONT . .e o T DC 1 - e- x EE

(9)

Since (see Fig. 1 and Fig. 3)

` T >> 1j & ^T >> x EEh

x EE ` PW << 1j & ` PW $ x EE << 1j & ` PW << 1j & ^x EE >> PW h x EE T x EE T

PW PW PP $ T T = Pave = Pave = PP = PP $ , PW PW PW DC T T T i MAXCONT = Pave $ RT

PP $ RT =

Pave 1 = 1 $ RT = Pave $ RT $ , and i MAXCONT $ DC DC DC PW P = T $ W = T $ DC. x EE x EE T x EE

Now, defining the operational efficiency η (a ratio of two temperatures, a kind of temperature gain) as in Eq. 10, T

(10)

- - x EE $ DC i MAXPULS = 1 $ e1 e o, T DC i MAXCONT 1 - e- x EE

temperature (this is true only if T is an independent variable, therefore under project control). In principle, for high efficiency, one just needs to operate the EED in the high T/τEE and low DC=Pw/T domains (or short pulse domain). These conditions are equivalent to T >> τEE>> Pw , since

Considering that both Pave = Pp $ DC = Pp $ (Pw /T) and T are fixed, short pulses (PW ↓) also mean high peak-power (PP↑ ). Therefore, this specific arrangement of T/τEE and DC makes the train of pulses function as a train of impulses. Then, the EED is quasi-instantaneously responding to the energy of each incoming pulse. One can see that as follows: from Eq. 8, with T >> τEE>> Pw,

i MAXPULS

J , 1 - Pw N xEE Pw K O j 1 - `1 - Pw x EE = x EE G 1 e K O = PP $ R T $ , PP $ RT $ T KK 1 - e- x EE OO 1 ,0 L P

Pw (11) , P $R $ P

EE RT $ CT

Pulse Energy

it is evident that, for certain combinations of DC and T/τEE, η can be made >> 1 (Fig. 4).

Figure 4. Operational efficiency (η) as a function of the duty-cycle (DC ) and the normalized period T/τEE .

Apparently, the pulsed excitation of the EED can be much more efficient in terms of the maximum attainable 48

1 ? , $ PP $ P W & CT i MAXPULS

\ Pulse Energy

This is exactly the case of electrostatic discharges (ESD), actuating as Pi(t), whenever CT is small and RT is large, a narrow pulse of high energy, an impulse for all practical purposes, can cause a rapid squib temperature increase, since the EED will not be able to dissipate, through 1/RT , the energy delivered to it, and the feedback loop in Fig. 2 can be ignored (by rapidly discharging a huge amount of water into a barrel, the liquid reaches its top almost instantaneously, because the leak through the hole at its base is irrelevant). Likewise, the operation of electrical capacitive discharge fuses, a particular case of the pulsed excitation (a pulsed excitation with T→∞, PP →∞ and PW →0 ), is now clear: almost all the thermal energy transferred to the EED will be converted, in just a single pulse, into a temperature increase, that is,

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 45-50, Jan. - Mar, 2012

An Analysis of the Initiation Process of Electro-explosive Devices

di ,

PP CT

# PW

# di = i MAXPULSE , i IMPULSE , C1T $ # PP dt = C1T $ PP $ PW

0 (12) i MAXPULSE , i IMPULSE \ ^ PP $ PW h &

Now, forcing RT $ PP > i THRESHOLD , which will require high peak power PP, ignition is assured (this corresponds to a barrel of small cross-section with a large hole in its base - the liquid level follows the input pattern).

i MAXPULSE , i IMPULSE \ Pulse Energy

CONCLUSIONS AND FURTHER RESEARCH i2

Note that to calculate

# di = i2 - i1 = i MAXPULS,

the

EED temperature θ1 (above ambient temperature), at the beginning of the power pulse,was assumed to be null (the EED was not able to stock-pile thermal energy during the previous cycles), a valid supposition, as T >>τEE. Once again, if θMAXPULSE > θTHRESHOLD , ignition is assured (only one short high-energy pulse, one-shot trigger impluse, is sufficient to initiate the explosive charge). Also, if τEE >> T > PW , then, from Eq. 6, P i dt = P dt & x EE CT S negligible

di + i = di + x EE dt di ,

PP dt CT

Di ,

PP $ Dt, CT

and the EED temperature increases progressively by, Di = ^ PP /CT h $ PW whenever an electric pulse of peak power PP (no further assumption on the peak power magnitude was necessary) and duration PW is applied to it (between pulses, di , 0 , and the EED temperature, θ, remains approximately constant). After many pulses, therefore, relatively slowly, the temperature θ can certainly cross the ignition threshold, causing, without notice, the squib initiation. This adiabatic effect (or temperature stacking – the EED is storing thermal energy) is nothing else than a gradual accumulation of thermal energy, cycle after cycle. The EED has become, via CT , a good longterm temperature integrator (a large, in diameter, barrel with a very small role in its base can gradually hold water inside it, 2 since x b = R0 $ rr ). This is a very interesting, singular EED behavior, with many important applications in the aerospace industry, because high peak power is no longer required. Finally, if τEE is very small (τEE << PW), the EED temperature practically follows the source of electrical power pulses Pi(t), once

di + dt negligible

Pi (t) i = Pi (t) i & , & i , RT $ Pi (t) x EE x EE CT CT do min ant factor

Under strict conditions, pulsed excitation of EEDs can be advantageous, provided τEE is known: If τEE >> T > PW (with T being an independent variable, therefore under project control), remember that temperature gain η can be even notable, but high peak power will be demanded (remenber that Pave is constant). On the other hand, if τEE >> T , EED temperature can softly reach the ignition threshold (either by friend or foe action), after several cycles, due to the gradual integration of the energy conveyed by each pulse, a kind of discrete electrothermal energy pumping - the temperature stacking effect. If time-to-ignition is not critical, any reasonable amount of peak power will do. And, if τEE << PW , the EED temperature simply follows the source of electrical power pulses Pi(t). It is also possible to initiate EEDs after one single highenergy impulse (one-shot fuse operation), particularly when CT is small, and RT is large. Certainly, for optimized operation, the electrical driving source shall match, in terms of either energy or power, EEDs distinguishing features. Furthermore, EEDs shall be shielded from their environment (coded ignition commands shall always be the rule): each EED must have its own local driving circuit module, only the coded commands shall come from the outside, preventing their soft ignition by foe intervention. Following this conservative line, a matter that indeed deserves further investigation is the EED (one without any kind of protection) sensitivity to electromagnetic radiation (radar-like pulse-modulated radio frequency). Although closely linked with the subject developed in this article, it does not come within its scope (Thompson, 1973).

REFERENCES Cochin, I., 1980, “Analysis and Design of Dynamic Systems”, New York, Harper & Row, Publishers, pp. 211.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 45-50, Jan. - Mar., 2012

Faria, P.C.C. et al.

Haberman, C. M., 1965, “Engineering Systems Analysis”, Ohio, C.E. Merrill Books, pp. 20-49.

Wire Model”, Propellants, Explosives, Pyrotechnics, Vol. 13, No. 4, pp. 120-125.

Moran, M.J., Shapiro, H.N., 2008, “Fundamentals of. Engineering Thermodynamics”, New York, John Wiley & Sons, pp. 131-132.

Rosenthal, L. A., 1961, “Thermal Response of Bridge-wires Used in Electro-explosive Devices”, Rev Sci Instrum, Vol. 32, No. 9, pp. 1033-1036.

Potter, M. C., Scott, E. P., 2004, “Thermal Sciences: An Introduction to Thermodynamics, Fluid Mechanics, and Heat Transfer”, Kentucky, Brooks & Cole, pp. 71-135.

Thompson, R. H., 1973, “Evaluation and Determination of Sensitivity and Electromagnetic Interactions of Commercial Blasting CAPS”, USBM, Washington, D. C., Rep. no. F-C3102.

Prince, W. C., Leeuw, M. W., 1988, “Analysis of the Functioning of the Bridge-wire Igniters Based on the Fitted

Wylie, C. R., 1975, “Advanced Engineering Mathematics”, Tokyo, McGraw-Hill Kogakusha, 4th ed., pp. 295-302.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 45-50, Jan. - Mar, 2012

doi: 10.5028/jatm.2012.04014311

Methodology for Structural Integrity Analysis of Gas Turbine Blades Tiago de Oliveira Vale*, Gustavo da Costa Villar, João Carlos Menezes Instituto Tecnológico de Aeronáutica - São José dos Campos/SP – Brazil Abstract: One of the major sources of stress arising in turbomachinery blades are the centrifugal loads acting at any section of the airfoil. Accounting for this phenomenon stress evaluation of the blade attachment region in the disc has to be performed in order to avoid blade failure. Turbomachinery blades are generally twisted, and the cross section area varies from the root of the blade to the tip. The blade root shape at the attachment region is of great concern. Stress concentrations are predictable at this contact region. In this paper, a finite element model has been created for the purpose of assessing stress at the joint region connecting the blade to the disc slot. Particular attention was paid to the geometric modeling of the “fir-tree” fixing, which is now used in the majority of gas turbine engines. This study has been performed using the commercial software ANSYS 13.0. The disc and blade assembly are forced to move with a certain rotational velocity. Contact connections are predicted on the common faces of the blade and on the disc at the root. Solutions can be obtained to allow the evaluation of stresses. Results can be compared with the mechanical properties of the adopted material.. Keywords: Stress Analysis, Finite Element Method, Gas Turbine Blade, Fir-tree Joint.

INTRODUCTION One of the main factors concerning mechanical integrity of aeroengine turbines is the interface region between the blade and the rotor disc. Stresses generated in this region are mainly produced by the centrifugal force resulting from the rotational speed of rotor mass of the blade, thermal stress, and and bending loads and torsion due to the gas pressure. According to Meguid et al. (2000), the joint between a turbine blade and the disc represents the most critical load path within this assembly, and it is crucial to the operational safety and service life of gas turbine engines. In order to accommodate safe stresses values in this contact region, an adequate geometry has to be developed for the joint. A variety of methods has been adopted for fastening blades to discs. The dovetail slot type is composed of a single tooth, and this type of fastening is usually used in rotors of compressors and fans, according to Beisheim and Sinclair (2008). Most frequently, fir-tree fasteners are implemented for they provide adequate multiple and larger areas of contact between the blade and the disc, which results in better ______________________ Received: 19/09/11. Accepted: 12/11/11 *author for correspondence: tiago.vale@yahoo.com.br/Pç. Mal. Eduardo Gomes, 50. CEP: 12.228-901 - São José dos Campos/SP -Brazil

stress distribution on the several teeth, according to Kanth (1998). The present study focused on the study of stresses arising from the centrifugal loadings in a fir-tree joint using a 3D Finite Element model in the commercial code ANSYS 13.0. Earlier works carried out by Singh and Rawtani (1982a,b) present results of photoelastic stress analysis of fir-tree assemblies. Very frequently, this kind of experimental technique is used to validate numerical results obtained by the Finite Element Method. In the work of Parks and Sanford (1978), the full distribution of stress field in a dovetail joint of a turbine was assessed using the photoelastic method. The results allowed the removal of material in the interface region and thus obtained 27% reduction in stress concentration, as well as to reduce the weight of the component. Uchino et al. (1986) developed a 3D photoelastic analysis in a dovetail joint of a gas turbine disc groove. The authors presented results of stress variation and compared them with 3D finite element analysis along the thickness. The work shows the influence of the inclination skew angle from 0 to 45°. Results for skew angle of 45° present 2.5 times higher stress compared to the straight skew

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 51-59, Jan. - Mar., 2012

Vale, T.O. et al.

angle. Therefore, the skew angle of 45° has the higher stress of those analyzed. Many researches developed structural analysis studies related to dovetail joints through the finite element method. Meguid et al. (1996), analyzed the stresses in an aeroengine compressor. The geometry was modeled using the software IDEAS, and the finite element model was created in ABAQUS. The authors showed that the maximum stress occurs just below the lower contact point between the blade and the disc. Papanikos et al. (1998) conducted a stress analysis study at a dovetail joint using the software ANSYS. It revealed that the lowest stress occurs for a 70° contact angle. Another important factor shown is the influence of friction, revealing that in the absence of this factor the maximum stress was 30% higher. The effect of the skew angle inclination was carried out by Anandavel and Prakash (2010). The study shows that the groove with skew angle of 20° presents a stress two times greater than the straight skew angle, and the stress with friction is about 30% less than frictionless. Shankar et al. (2010) illustrated the stress distribution through 3D finite element analysis of low-pressure steam turbine bladed disc assembly with loading and constant speed of 6,000 RPM in ANSYS 12.0. The results showed that the peak stress of 1,187MPa is seen at blade root fillet pressure face and 1,055MPa at suction side. Other studies using finite element method were carried out for fir-tree joints. Chan and Tuba (1971) checked the stress in a fir-tree root through finite element method by changing the coefficient of friction and the clearances. Clearance changes showed more significant effects of stress than the coefficient of friction. Singh and Rawtani (1982a) evaluated the deformation in the bottom tooth of a fir-tree root. In their second work, they (1982b) studied the complete fir-tree root. Delhelay (1999) performed a thermo-mechanical analysis in the fir-tree root region. This work presented the study of stresses in 2D and 3D geometries, considering non-linear contact conditions. It was concluded that the bottom tooth is subjected to the highest stresses and the coefficient of friction does not significantly alter the stress values. Meguid et al. (2000) evaluated the von Mises stress distribution undertaken of the fir-tree region in aeroengine turbine disc assembly with different geometries, using the software ANSYS. The studies show the influence of the contact angle α, top (γ) and bottom flank angle (β), straight and skew angle, coefficient of friction, and number of teeth. 52

Results show that the stress peak occurs right below the lower contact point along the bottom tooth of the turbine disc, and the coefficient of friction alters the stress value. The search for optimized geometries of fir-tree joints has been the main purpose of designers and engineers. Thinking of this, Song et al. (2002) developed an optimization program of fir-tree root of a turbine, using a CAD system with several optimization algorithms and stress analysis by finite element method. A similar work was done by Brujic et al. (2010), the author developed an optimization program of the fir-tree root of a gas turbine with automatized results. The program was written in MATLAB, modeled in CATIA, and finite element meshing and stress analysis were generated by PATRAN/NASTRAN.

VALIDATION OF THE METODOLOGY BY 3D FINITE ELEMENT ANALYSIS Analysis of the dovetail rim region In this section, the stress distribution was analyzed in the interface region of the blade and disc in a dovetail joint, considering a rectangular geometry for the blade and disc, employing a Finite Element Model. For this task, the commercial software ANSYS 13.0 was used. Blade-disc dovetail joint is an important component and is usually found in the fan and compressor disc assemblies of aero-engine. In this study, the 3D model has been created on the software CATIA V5R18. A similar reference model was examined by Papanikos et al. (1998), and Anandavel and Prakash (2010). This study is concerned with the validation of the methodology applied in the contact between the disc and blade using contact elements. In addition, a mesh convergence test was carried out to obtain a proper mesh for the model. In the next section, the validation of the method found here will be imposed in the fir-tree joint, which is the main goal of this work. One may comment on the great difficulty to find in the literature a geometry that represents the interface region between the blade and disc, and results generated by centrifugal loading. The dimensions of the blade geometries and disc developed are presented in Fig. 1. Boundary conditions For the present study, the material used was the same employed by Papanikos et al. (1998). The properties of the

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 51-59, Jan. - Mar, 2012

Methodology for Structural Integrity Analysis of Gas Turbine Blades

tool was used, greatly reducing computational time, compared to a full disc/blades model that would require a much higher computational effort. Contact elements

Figure 1. Geometry of dovetail joint.

materials used for modeling the blade and disc were that of titanium alloy Ti-6Al-4V. In this work, the following values were used: Young’s modulus E=114GPa, Poisson’s ratio ν=0.33, and density ρ=4429kg/m3. All examined models were submitted to centrifugal loading with specific angular velocity, where ω was selected to be 1,000rpm. In view of symmetry of geometry and loading, only one sector of the disc supporting 12 blades was modeled as shown in Fig. 1. This model was imported to ANSYS 13.0 and the cyclic symmetry

The contact surfaces in the dovetail region were modeled using contact elements. The contact region between the blade and the disc is recognized by ANSYS 13.0, at the time the geometry is imported from CAD software. Friction and frictionless conditions were selected on the interface, and Lagrange multiplier method was used to obtain solutions for normal and friction contacts. The dovetail interface was defined as the stiffness matrices of each element is updated at each iteration. Given these definitions, the contact region is considered as a nonlinear contact condition (Fig. 2). Mesh There are several methods for discretization available for the generation of finite element meshes in non-linear contact analysis. A free meshing routine was used due to the necessity to model a complex geometry with large transitions in the stress field.

Figure 2.Contact elements. J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 51-59, Jan. - Mar., 2012

Vale, T.O. et al.

Figure 3. Types of mesh studied in dovetail joint.

Figure 3 shows typical 3D meshes used to test convergence. Three types of meshes have been studied. The meshes HexaMulti 3 and 1mm contain hexahedron elements with 20 nodes per element. A third case, the TetraRef mesh, contains tetrahedral elements with ten nodes, which was refined in the region of contact between the blade and disc. The number of elements and nodes for each mesh type is shown in Table 1. Table 1. Number of elements and nodes for each type of mesh in dovetail joint. HexaMult 3mm HexaMult 1mm

TetraRef

Nº of elements

5964

106172

93299

Nº of nodes

28535

455218

149580

It should be noted that the Von Mises stress values for all figures were normalized by σ/ρω2a2, ρω2a2 = 70.5kPa, where ρ is the density of material, ω the angular velocity in rad/s and a is the inner radius of the disc. The curve represented in Fig. 4 illustrates the normalized Von Mises stress across thickness at the lower contact

line, through interface between the blade/disc for two types of analysis. The three cases described before, considering a coefficient of friction μ = 0.25 and μ = 0.0 (frictionless) were compared to the results published by Papanikos et al. (1998). One may notice a very good agreement between the present result and those reported by Papanikos et al. (1998). Figure 5 shows the normalized Von Mises stresses along the interface between the blade and the disc. In this case, the stresses were obtained by comparing the 3D models studied in this paper with those found in the literature for 2D and 3D models. In these cases, the present results were obtained for coefficient of friction μ = 0.25. According to this analysis, one may observe that the curves of stresses for HexaMulti 1mm and TetraRef meshes are very close in agreement with the 3D and 2D models analyzed by Papanikos et al. (1998). Only the HexaMulti 3mm mesh has not shown good results, it was observed through the peak stresses. Table 2 shows the percentage differences between the maximum stresses obtained by the several cases with maximum stress used as a reference to validate the applied methodology.

Table 2. Differences in percentage between the maximum stresses. Relationship between literature and

Normalized maximum stress

the meshes studied

along thickness μ = 0.25

along thickness μ = 0.0

along interface blade/disc

Papanikos et al. HexaMult 3mm

26 25.7

37.14 37.56

18.4 21.7

Difference [%]

1.15%

-1.14%

-18.16%

HexaMult 1mm

25.26

36.9

21.27

Difference [%]

2.85%

0.64%

-15.85%

TetraRef Difference [%]

26.29 -1.1%

37.62 -1.31%

19.61 -6.74%

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 51-59, Jan. - Mar, 2012

Methodology for Structural Integrity Analysis of Gas Turbine Blades

Figure 4. Normalized Von Mises stress along thickness at the lower contact line for different meshes.

Figure 5. Normalized Von Mises stress along the interface at different meshes.

For each case, the normalized maximum Von Mises stresses are presented. Taking the case presented by Papanikos et al. (1998) as a reference, the percentage difference between results is reported. These results show that stresses across thickness with friction presented a small difference in all cases, and the largest one (2.853%) was found when the HexaMulti 1mm mesh was used for the frictionless case, with the difference being less than 1.31%. However, the analysis of stresses along the interface using the HexaMulti 3mm mesh presents a very high percentage difference of 29.71%, which was not observed for the other two meshes. It can be concluded that for the methodology used, very consistent and safe results may be obtained.

3D NONLINEAR FINITE ELEMENT ANALYSIS IN FIR-TREE JOINT In this section, a finite element analysis of the fir-tree region was carried out. Fir-tree joint is very used in the compressor disc assemblies in turbine axial, because it has a larger contact area between the teeth of the blade and the disc, better distributing the stresses. In this study, the geometrical model created is similar to that used in the works of Venkatesh (1988) and Meguid et al. (2000). However, one may observe the lack of more proper reported results, and even for the cited authors, important dimensions for the blade geometry are omitted, making it difficult to find the most perfect reproduction of the earlier proposed geometry. The adopted geometry is as shown

J. Aerosp. Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.4, No 1, pp. 51-59, Jan. - Mar., 2012

Vale, T.O. et al.

Figure 6. Disc and blade geometry fir-tree joint.

in Fig.6. For the disc and blade, a thickness of 10 mm was adopted. Top and bottom flank angles were taken as β=γ=40°, and rotation speed of the disc was considered as 360rpm. The material properties used for modeling the blade and the disc were that of a typical nickel alloy used in disc design; namely, INCONEL 720. This material is creep and fatigue-resistant. In this work, the following values were used: Young's modulus E=220GPa, 0.2% proof stress (σ=635MPa at 213°C), Poisson's ratio (ν=0.29), density (ρ=8510kg/m³). The boundary conditions, contact elements, and mesh applied here, were the same imposed by the dovetail model.

RESULTS

Figure 7. Types of mesh studied in fir-tree joint.

According to the results reported by previous studies of the meshes, we decided to chosen the HexaMulti 1mm and TetraRef meshes. These meshes obtained better results than the HexaMulti 3mm one. Figure 7 illustrates two cases of the meshes and the number of elements and nodes are shown in Table 3. Figures 8 and 9 present the normalized Von Mises stress obtained for the two types of mesh, which were previously described compared to results in reported by Meguid et al. (2000). Figure 8 presents the stress behavior along the

Table 3. Number of elements and nodes for each type of mesh.

HexaMult 1mm

TetraRef

Nº of elements

30160

348440

Nº of nodes

139074

598374

entire profile of the teeth on the disc side, which reveal higher levels of stress. Figure 9 focuses on the bottom region of the profile. One may notice a better similarity of the normalized Von Mises stress curve obtained by TetraRef mesh with the study

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 51-59, Jan. - Mar, 2012

Methodology for Structural Integrity Analysis of Gas Turbine Blades

Figure 8. Normalized Von Mises stress along the teeth profile for different meshes.

Figure 9. Normalized Von Mises stress along the bottom tooth profile for different meshes.

of Meguid et al. (2000). Taking this observation into account, the following studies were performed using TetraRef meshes. Seven other geometries were generated combining different values of contact angle, top flank angle, superior flank angle, and clearance between the blade and the disc. Table 4 shows the different angles studied. Figures 10 and 11 show the results of stress along the front and back interfaces on the left side of the fir-tree joint. The graphs show that there is little variation in stresses between the front and back interfaces, because the studied

geometry is symmetrical. Such variations can be seen in Table 5, which presents with details the differences of maximum Von Mises stresses between the front and back sides. Through this analysis, it was observed that the Case 4 already demonstrated good results in the front, but with considerable variation in the back interface. Case 7 showed the best results with a 10% average in both interfaces, since, without the clearance between the blade/disc, the contact area in the interface region is bigger than the others cases, decreasing stress.

Table 4. Studied geometries. Case

Coefficient of

Contact angle,

Bottom flank

Top flank

Inferior flank

Superior flank

Clearance between

friction μ

angle, β

angle, γ

angle, Li

angle, Ls

blade/disc

17.5º

40º

3.556

2.08

0.38

17.5º

40º

30º

3.556

2.61

0.38

15º

40º

3.556

2.25

0.38

15º

40º

30º

3.556

2.77

0.38

20º

40º

3.556

1.89

0.38

20º

40º

30º

3.556

2.43

0.38

15º

40º

30º

3.556

2.77

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 51-59, Jan. - Mar., 2012

Vale, T.O. et al.

Figure 10. Normalized Von Mises stress along the teeth profile for different meshes on the front side for several cases.

Figure 11. Normalized Von Mises stress along the teeth profile for different meshes on the back side for several cases. Table 5. Maximum normalized Von Mises stress for the cases analyzed. Normalized maximum stress along entire interface [x/I14] Front

Back

Difference

case 1 case 2

53.15 52.03

case 1 case 2

52.07 53.36

-2.02% 2.56%

case 3

59.31

case 3

51.84

-12.6%

case 4 case 5 case 6

48.1 62.32 54.03

case 4 case 4 case 6

57.41 62.25 58.94

19.4% -0.12% 9.08%

case 7

44.3

case 7

46.73

5.5%

CONCLUSIONS Results from this paper revealed what may be commented as follows: • the maximum stress value occurs for Case 5, which shows that angles β and γ = 40º are not appropriate; 58

• the minimum stress occurs with a contact angle between 15° and 17.5º; • stresses between the front and back region do not vary greatly, except for Case 4. This study shows the importance of calculating the clearance between the teeth of the blade and the disc, for the calculation of thermal expansion of the bodiesindicates the smallest possible value of clearance, thus obtaining a lower stress for the geometry to bedeveloped. However, this factor was not evaluated in this study. Future papers should clarify the influence of the clearance on the stress values under conditions of thermal expansion and coefficient of friction with variation between 0.0 and 0.5. The presented model uses the commercial code ANSYS 13.0, which provides several characteristic tools. Among them, there is the possibility of analyzing structures under centrifugal loading. Another useful tool is the friendly interface for modeling contact elements, which is fundamental for this type of analysis.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 51-59, Jan. - Mar, 2012

Methodology for Structural Integrity Analysis of Gas Turbine Blades

The present work is a preliminary study of the fir-tree joint between blades and disc. Much more effort must be performed in order to offer the optimum geometry for more general cases.

REFERENCES Anandavel, K. and Prakash, R.V., 2010, “Effect of three-dimensional loading on macroscopic fretting aspects of an aero-engine blade-disc dovetail interface”, Tribology International. Beisheim, J.R. and Sinclair, G.B., 2008, “Three-Dimensional Finite Element Analysis of Dovetail Attachments With and Without Crowning”, Journal of Turbomachinery, Vol.130, No. 2. Brujic, D. et al., 2010, “CAD based shape optimization for gas turbine component design”, Struct Multidisc Optim, Vol. 41, No. 4, pp. 647-659. Chan, S.K. and Tuba, I.S., 1971, “A finite element method for contact problems of solid bodies - Part II. Application to turbine blade fastenings”, International Journal of Mechanical Sciences, Vol.13, No. 7, pp. 627-639. Delhelay, D. S., 1999, “Nonlinear Finite Element Analysis of the Coupled Thermomechanical Behaviour of Turbine Disc Assemblies”, Thesis University of Toronto. Toronto, pp. 95. Kanth, P.S., 1998, “2D & 3D FE analysis of fir-tree joints in aeroengine discs”, Thesis University of Toronto. Meguid, S.A., et al. 2000, “Finite element analysis of fir-tree region in turbine discs”, Finite Elements in Analysis and Design”, Vol. 35, No. 4, pp. 305-317. Meguid, S.A., et al. 1996, “Theoretical and experimental studies of structural integrity of dovetail joints

in aeroengine discs”, Journal of Materials Processing Technology, Vol. 56, No. 1-4, pp. 668-677. Papanikos, P., et al.1998, “Three-dimensional nonlinear Finite element analysis of dovetail joints in aeroengine discs”, Finite Elements in Analysis and Design, Vol. 29, No. 3-4, pp.173-186. Parks, V.J. and Sanford, R.J., 1978, “Photoelastic and Holographic Analysis of a Turbine-engine Component”, Experimental Mechanics, Vol.18, No. 9, pp. 328-334. Shankar, M. et al. 2010, “T-root blades in a steam turbine rotor: A case study”, Engineering Failure Analysis, Vol.17, pp. 1205-1212. Singh, G.D. and Rawtani, S., 1982a, “Fir tree fastening of turbomachinery blades - I deflection analysis”, International Journal of Mechanical Sciences, Vol. 24, No. 6, pp. 377-384. Singh, G.D. and Rawtani, S., 1982b, “Fir tree fastening of turbomachinery blades - II step load analysis”, International Journal of Mechanical Sciences, Vol. 24, No. 6, pp. 385-391. Song, W. et al., 2002, “Turbine blade fir-tree root design optimisation using intelligent CAD and finite element analysis”, Computers & Structures, Vol. 80, No. 24, pp. 1853-1867. Uchino, K. et al.,1986, “Three dimensional photoelastic analysis of aeroengine rotary parts”, Proceeding of the International Symposium on Photoelasticity, Tokyo, Japan, pp. 209-214. Venkatesh, S., 1988, “Structural integrity analysis of an aeroengine disc”, Thesis Cranfield Institute of Technology.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 51-59, Jan. - Mar., 2012

doi: 10.5028/jatm.2012.04013211

Experimental Determination of Temperature During Rotary Friction Welding of AA1050 Aluminum with AISI 304 Stainless Steel Eder Paduan Alves1,*, Francisco Piorino Neto1, Chen Ying An2, Euclides Castorino da Silva1 Instituto de Aeronáutica e Espaço - São José dos Campos/SP - Brazil Instituto Nacional de Pesquisas Espaciais - São José dos Campos/SP - Brazil

1 2

Abstract: The purpose of this study was the temperature monitoring at bonding interface during the rotary friction welding process of dissimilar materials: AA1050 aluminum with AISI 304 stainless steel. As it is directly related to the mechanical strenght of the junction, its experimental determination in real time is of fundamental importance for understanding and characterizing the main process steps, and the definition and optimization of parameters. The temperature gradients were obtained using a system called Thermocouple Data-Logger, which allowed monitoring and recording data in real-time operation. In the graph temperature versus time obtained, the heating rates, cooling were analyzed, and the maximum temperature was determined that occurred during welding, and characterized every phases of the process. The efficiency of this system demonstrated by experimental tests and the knowledge of the temperature at the bonding interface open new lines of research to understand the process of friction welding. Keywords: Friction Welding, Aluminum, Stainless Steel, Dissimilar Materials, Temperature. INTRODUCTION The rotary friction welding (RFW) is a special process that occurs in the solid state. It provides high productivity, excellent repeatability, low cost, and its greatest application is found in the production of dissimilar materials joints used in aerospace, nuclear, marine, and automotive fields. All process happens at temperatures lower than the melting point of the materials involved and the joints produced are of excellent quality, featuring superior mechanical properties of the metals that were joined. In the RFW, one part is fixed and rotated by a motor unit to a predetermined speed, and the other is positioned, aligned, and moved by a hydraulic piston to touch the part that is spinning. After that, a P1 pressure is applied for a given time (t1), the machine is braked until it reaches zero speed, and P2 pressure is applied during a t2 time, finishing the welding (Alves, 2010a). Parameters of welding (rotational speed = rotation per minute (RPM), P1 and P2 pressures, t1 and t2 times) are defined by welding procedures established for each material or materials and according to the type of the equipment employed. Figure 1 shows the phases of the process.

Figure 1. Phases of conventional friction welding process. A: period of approximation; B: P1, t1, application; C: end of P1, t1, and breaking of the machine (RPM=0); D: P2, t2 application and finish welding (Alves, 2010a).

_____________________ Received: 11/07/11. Accepted: 29/12/11 *author for correspondence: ederavcp12@terra.com.br/ Pç. Mal. Eduardo Gomes, 50. CEP: 12.228-901 - São José dos Campos/SP– Brazil

Figure 2 shows the basic layout of RFW equipment, whose flawless performance and timing are of fundamental importance in the quality of joints obtained by this process.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 61-67, Jan. - Mar., 2012

Alves, E.P. et al.

Figure 2. Equipment of RFW (Alves, 2010a).

When using this process to the union of two materials, it is very important to know the temperature in the bonding interface, because it directly interferes with the formation of the crystal structure of the heat affected zone (HAZ), influencing the mechanical and metallurgical properties of the welded joint (Alves, 2010a). All the heat necessary for welding is produced by the direct conversion of mechanical energy into the thermal one. It is a complex metallurgical process, which involves a series of variables, such as pressure, time, travel speed, rotational speed, accompanied by physical phenomena: heat generation by friction, atomic diffusion, plastic deformation and formation of intermetallic compounds. During the relative motion of surfaces, a significant amount of heat is dissipated causing temperature increase, even with small values of loads and sliding speeds (Burakowiski and Wierzchon, 1998). In friction welding, the heat generation occurs differently from conventional welding processes for fusion, however there is a similarity in the temperature distribution on the joint of union of base metals (Alves, 2010a). The amount of heat generated at the bonding interface or heat input in RFW is a consequence of the friction and work of plastic deformation due to the relative motion between both materials. The temperature at the surface depends on the applied pressure, rotational speed, thermal conductivity, and also on the coefficient of friction. Heat dissipation is an automatic process since friction and adhesion occur in places where micro welds cause an increase in the rate of heat dissipation, which contributes to an increase in micro welds and bond of two surfaces (Burakowiski and Wierzchon, 1998). 62

There are many published studies on RFW and on its thermal effects through experimental and analytical methods, which were conducted by researchers in different countries according to their importance in understanding the mechanisms that involve the process. Vill, 1962; Ylbas et al., 1994; Sahin, 2004; Chalmers, 2001; Zepeda, 2001; Nikolaev and Olshansky, 1977; Aritoshi and Okita, 2003; Ambroziak et al., 2007; Isshiki et al., 2008; KusĂ§u et al., 2007; Sluzalec, 1990; Lee, 2003; Kimura et al., 2010; Ochi et al., 1998; Banker et al., 2002, conducted several studies involving the joining of dissimilar materials. These authors wrote articles on the mechanical properties and on the metallurgical and thermal effects of the welded parts of the RFW. In this study, a method for monitoring the temperature in the bonding interface in real-time operation, called Thermocouple Data-Logger (TDL), was used (Alves, 2010a). Through this system, we can monitor, determine the beginning and end of each stage of welding, analyze the different heating rates and cooling, and characterize all stages of the process through time versus temperature curves.

TEMPERATURE AT BONDING INTERFACE When two dissimilar materials are joined, such as AA1050 aluminum and AISI 304 stainless steel, with different properties by this process, the heat generated by friction between the two materials is spread differently in each material. The thermal conductivity of aluminum is three times higher than in stainless steel, influencing directly the

J. Aerosp. Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.4, No 1, pp. 61-67, Jan. - Mar, 2012

Experimental Determination of Temperature During Rotary Friction Welding of AA1050 Aluminum with AISI 304 Stainless Steel

EXPERIMENTAL PROCEDURE

rate of heating and cooling that occur during the process. The surface roughness of the interfaces that will be attached can also change the heating rates in the initial stages of the welding operation and influence the diffusion mechanism, which occurs mainly in the first phase of welding (heating phase). The temperature gradient and thermoplastic deformations determine microstructural changes, diffusion phenomena, and mechanical properties of the final product (Lindemann et al., 2006). During the welding of aluminum with stainless steel, the rise of temperature in the bonding interface causes a large plastic deformation and flash formation in AA1050 aluminum. Part of the heat is dissipated into the flash and into the contacts of the materials with the components of the welding equipment. As can be seen in Fig. 3, during the welding of AA1050 aluminum with stainless steel AISI 304, the initial temperature is higher in the peripheral region due to the higher tangential velocity, and then it extends to the centre of interface increasing with the heating time (t1, t2, t3, t4, t5, t6, t7, t8, t9). After a given time, the difference between the temperatures is going to be very small, especially on the aluminum side that has a high thermal diffusivity (Lindemann et al., 2006). When the material reaches the critical temperature Tc, the material begins to undergo severe plastic deformation leading to formation of the flash, which dissipation is also responsible for part of the of heat generated by the process.

Materials and surfaces preparation The materials used in this study were: AA1050 aluminum (commercially pure aluminum, 99.5%) and AISI 304 austenitic stainless steel. Both materials were machined with a diameter of 14.8 mm and lengths of 100 and 110 mm, respectively. After machining, they were subjected to be cleaned with acetone in order to remove organic contaminants, such as oils, greases, and so on. Tables 1 and 2 present chemical compositions and mechanical properties of the materials. Friction welding equipment A rotary friction welding machine, GATWIK brand, was used with fixed rotational speed of 3,200RPM, P1=2.1MPa, t1=32 seconds, P2=1.4MPa and t2=2 seconds. These parameters refer to welding procedures by friction between the related materials described in a previous paper (Alves, 2010a), optimized and qualified with the fracture occurring in the AA1050 aluminum, away from the bonding interface with the mechanical resistance superior of AA1050 aluminum (Alves, 2010b). Temperature in the bonding interface was also monitored with the heating time t1 extended to 52 seconds for analysis and comparison of the curves and rates of warming and cooling during welding. The materials were placed as shown in Fig. 4.

Figure 3. Distribution of temperature on the bonding interface: RT: room temperature; Tc: critical temperature. Source: Adapted from Fukumoto et al., 1997. Figure 4. Schematic view of the materials placed before welding. Table 1. Nominal chemical compositions of materials. AA1050 Aluminum

AISI304 Stainless Steel

Si(wt%)

Fe (wt%)

Cu (wt%)

Mn (wt%)

Mg (wt%)

Cr (wt%)

Zn (wt%)

Ti (wt%)

0.07

0.26

< 0.001

< 0.002

< 0.007

Si (wt%)

S (wt%)

P (wt%)

Mn (wt%)

C (wt%)

Cr (wt%)

Ni (wt%)

0.38

0.024

0.036

1.67

0.054

18.2

8.0

J. Aerosp. Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.4, No 1, pp. 61-67, Jan. - Mar., 2012

Alves, E.P. et al. Table 2. Mechanical properties of materials. Strenght σ (MPa) Yield Maximum

Elongation ε (%) Maximum Fracture

Modulus of elasticity E (GPa)

AA1050 Aluminum

44.70

78.48

59.12

AISI304 Stainless Steel

354.69

643.79

177.10

Temperature monitoring during friction welding process Temperature is the most important parameter of a joint in the solid state by controlling the kinetics of thermally activated processes involved in diffused junctions. In the joints that occur at high temperatures, the atomic mobility increases and helps the movement displacement of atoms through the bonding interface (Bagnato, 2002). As the temperature in the interface is directly related to the characteristics of the HAZ and with the strength of joints obtained by the RFW, its monitoring on a trial is extremely important for understanding the characteristics of this process. For data acquisition, the TDL system was used, coupled to a notebook that provided real-time graph of the temperature during the process. We used a thermocouple type K (cromelAlumel) measured and calibrated, ECIL brand, positioned on the pin of AISI 304 stainless steel, axial region, at a distance of 0.12mm of the interface (Fig. 5) and a data logger, brand NOVUS, with 16k of storage capacity, which allowed the collection of data in 0.5 second intervals during all the process. It was also used thermal paste of brand IMPLASTEC to improve the area of contact between the thermocouple tip and the surface of stainless steel pin. A total of five measurements was realized during the welding operation. The TDL system was settled up by software in a Windows environment and it provides resources for collecting, plotting, analyzing, and exporting logs. Communication between the data logger and the notebook is realized in a few seconds via infrared optical non-contact (Alves, 2010a). Figure 5 shows an illustration of the TDL system used for temperature monitoring in real-time (Alves, 2010a).

RESULTS In welding tests, which were performed with temperature monitoring by the TDL system, the pins of AISI 304 stainless steel showed changes in the color of the surface near the bonding interface due to displacement of heat flow. Pins made from the AA1050 aluminum also show displacement of 64

heat flows, but they are not visible on the surface due to the characteristics of the material. The thermocouple fixed in the axial region of the cylindrical pin of AISI 304 stainless steel (Fig. 5) recorded a maximum temperature of 376 °C during the welding process in real-time of 34 seconds (Approach + t1 + t2), shown in Fig.6 (Alves, 2010a).

Figure 5. TDL system composed of K-type thermocouple, data logger, infrared (IR) reader IR, and notebook to monitor the temperature.

In the graph obtained, all phases of the welding process, the approach with a time of ten seconds, the first phase of heating with application of pressure (P1=2.1MPa) at the time of 22 seconds, the second phase of forging with application of pressure (P2=1.4MPa) at the time of two seconds, and the completion of welding were characterized.It was observed that when the rotational speed was stopped at the end of the first phase (point A – 376 °C), a drop in temperature during the forging phase (B – 350 °C) occurs, finishing the welding. Hence, the phase of cooling to room temperature starts. This type of cooling does not interfere in the characteristics of the HAZ and the mechanical properties of the junction between the materials involved, mainly because the AA 1050 aluminum is not heat-treatable and has high purity (minimum 99.50% Al), which was proven by the results of the analysis and testing in this study. However, in the case of welding

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 61-67, Jan. - Mar, 2012

Experimental Determination of Temperature During Rotary Friction Welding of AA1050 Aluminum with AISI 304 Stainless Steel

Figure 6. Graph temperature versus time – AA 1050 aluminum

and AISI 304 stainless steel with total time of 34 seconds. AISI 304 stainless steel with heat-treatable alloys (series 2XXX, 6XXX and 7XXX), the slow cooling can change the characteristics of the HAZ, as the proximity of the temperature values obtained in the bonding interface with the values used for heat treatment of solubilization and aging of these alloys (Alves, 2010a). With time t1 extended to 52 seconds, the process temperature with total time of 62 seconds (approach=eight seconds; t1=52 seconds; t2=2 seconds), using the same welding parameters of P1 and P2, there is an increase of temperature during the heating phase, and over time it was stabilized to the temperature of 410°C, point "A". After applying the pressure P2 and the time t2, the welding was completed with a temperature of 392°C, point "B". The air cooling performed at room temperature (30°C) showed similar cooling rates to the previous example (total time of 34 seconds). Figure 7 (Alves, 2010a) shows all process phases and temperatures monitored during welding of AA1050 aluminum with AISI 304 stainless steel. Another important result of the analysis performed is that the maximum temperature monitored is within the range of the hot forging temperature of the alloy AA 1050, between 315 to 430°C (ASM, 1993). This knowledge allows the default parameters of RFW for a given diameter using data supplied by the graph, which allows to eliminate a series of preliminary stages in obtaining parameters for the welding of dissimilar materials. In our observations, we found that although each welding parameter individually has its importance, the relationship between them and their subsequent phases result in the formation of a joint with good mechanical properties and ideal for an application.

Figure 7. Graph temperature versus time – AA 1050 aluminum and AISI 304 stainless steel with total time of 62 seconds.

Figure 8. (a) shape of the flash in RFW; (b) bonding interface after machining (samples on graph paper)

In the heating phase, the relationship among the parameters (P1 pressure, t1 time, travel speed and rotational speed) aims at increasing the temperature at the bonding interface through the friction between the contact surfaces. This rapid increase in temperature with the constant application of pressure promotes a severe plastic deformation and removal of oxides and impurities through the flash, providing ideal conditions for the occurrence of the phenomenon of diffusion. In the early phase of forging, the temperature at the interface should have reached a certain level so that with the application of forging pressure P2 in the time interval t2, the union between the materials was successfully completed. If this relationship does not occur satisfactorily in the heating phase, and between it and the forging phase the cycle of welding ends, the layers of oxides and impurities can not be removed completely from the surfaces. The temperatures necessary for the occurrence of diffusion and forging may not be enough for the perfect union of the materials involved, resulting in a junction with poor mechanical

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 61-67, Jan. - Mar., 2012

Alves, E.P. et al.

properties. Figure 8 shows the flashes formed by plastic deformation and the bonding interface after the machining. As the temperature at bonding interface is directly related to the mechanical resistance of the junction, prior knowledge can be used to optimize processes and parameters, reducing operating times and production costs, which are the key factors for growth and development of industries working with high productivity in an increasingly globalized world.

CONCLUSIONS The temperature monitoring interface in real-time connection with the TDL system proved to be very efficient, because the results of this study showed the importance of characterization of all phases of the process in real-time by means of curves time versus temperature. The temperature monitoring during welding tests as used parameters recorded the maximum temperature of 374 °C. This result confirmed that the temperature at the bonding interface during welding coincides with the range of hot forging of AA1050 aluminum (315-430 °C), as quoted in the literature (ASM, 1993). The highest rates of heating occurring in the first ten seconds of the first phase of welding (heating phase) tend to stabilize as a function of deformation and plastic flow of the AA 1050 aluminum. Knowing the temperature curves for certain joints between dissimilar materials, they can be used to obtain optimization and qualification of parameters, reducing stop times for equipment set up in different tests that involve RFW, machining, and mechanical testing. The results of this study were of fundamental importance for understanding and comprehending the RFW, allowing the characterization of the different phases that involve this process and observation of heating and cooling rates. This knowledge will allow the opening of new lines of research, optimization, cost reduction, and increased productivity.

Alves, E. P. et al., 2010b, “Welding of AA1050 aluminum with AISI 304 stainless steel by rotary friction welding process”, Journal of Aerospace Technology and Management, São José dos Campos, Vol. 2, No. 3, pp.301-306. Aritoshi, M. and Okita, K., 2003, “Friction Welding of Dissimilar Metals”, Journal of the Japan Welding Society, pp. 24, Translated by Welding International,pp.271-275. Ambroziak, A. et al., 2007, “Friction welding of dissimilar metal joints with intermediate layers”, Journal of Achievements in Materials and Manufacturing Engineering, Vol. 21, pp. 37-40, Issue 2.ASM, 1993, “Metals Handbook: Forming and Forging”, Vol. 14, 9th ed. Processing of aluminum alloy forgings, Metals Park, Ohio, USA. Bagnato, O. R., 2002, “Propriedades Mecânicas de juntas AlSi12 / Al2O3 soldadas por difusão”, Tese (Doutorado em Engenharia Mecânica) – Universidade Estadual de Campinas, Faculdade de Engenharia Mecânica, Campinas. Banker, J. et al., 2002, “Aluminum-Steel Electric Transition Joints, Effects of Temperature and Time upon Mechanical Properties”. In: Draft 131ST Annual Meeting, February 17-21, 2002, Seattle, WA, USA. Proceedings… Seattle: [son]. Burakowiski, T., Wierzchon, T., 1998, “Surface engineering of metals – principles, equipment, technologies”. Thermal Effects of Friction, Retrieved in November 25, 2010, from http://books.google. com/books? Id=MyMn8FwROC&printsec=frontcover&dq= friction+welding-+dissimilar. Chalmers, R. E., 2001, “The Friction Welding Advantage”, p. 64-70, Retrieved in October 20, 2009, from: www.sme.org/ manufacturingengineering. Fukumoto, S. et al., 1997, “Evaluation of Friction Weld Interface of Aluminum to Austenitic Stainless Steel Joint”. Materials Science and Technology, Vol.13, No. 8, pp. 686.

REFERENCES Alves, E. P., 2010a, “Junções de Materiais Dissimilares utilizando o Processo de Soldagem por Fricção Rotativa”, Dissertação (Mestrado em Engenharia e Tecnologia Espaciais/ Ciência e Tecnologia de Materiais e Sensores) – Instituto de Atividades Espaciais – INPE, São José dos Campos, Brasil. 66

Isshiki, Y. et al., 2008, “Prediction Method of Transient Temperature Distribution in Friction Welding of Two Similar Materials of Steel”, The International Society of Offshore and Polar Engineers (ISOPE) Conference, pp. 261-267, Vancouver, BC, Canada

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 61-67, Jan. - Mar, 2012

Experimental Determination of Temperature During Rotary Friction Welding of AA1050 Aluminum with AISI 304 Stainless Steel

Kimura, M., et al., 2010, “Analysis Method of Friction Torque and Weld Interface Temperature during Friction Process of Steel Friction Welding”, Journal of Solid Mechanics and Materials Engineering, Vol. 4, No.3. Kusçu, H. et al., 2007, “Experimental Determination of Temperature by the Friction Welding Method”. Baltic Heat. Presented at the Baltic Heat Transfer Conference, St. Petersburg, Russia, Retrieved in December 3, 2010, from: http://dl.begehouse.com. Lee, W. B., 2003, “Effect of Friction Welding Parameters on Mechanical and Metallurgical Properties of Aluminum Alloy 5052 - A36 Steel Joint”, Materials Science and Technology, Institute of Materials Minerals and Mining, Vol. 19, pp. 778. Lindemann, et al., 2006, “Thermo-mechanical phenomena in the process of friction welding of corundum ceramics and aluminum”, Bulletin of the Polish Academy of Sciences Technical Sciences, Vol. 54, No.1, Institute of Mechanics and Design – Warsaw University of Technology, Poland. Nikolaev, G., Olshansky, N., 1977, “Advanced Welding Process”. Moscow: MIR.

Ochi, H., et al., 1998, “Friction Welding of Aluminum Alloyand Steel”, International Journal of Offshore and Polar Engineering, Vol. 8, No. 2. Sahin, M., 2004, “Joining with friction welding of high-speed steel and medium carbon steel”, Journal of Materials Processing Technology, Vol. 168, pp. 202-210, Retrieved in August 20, 2009, from: www.elsevier. com/locate/jmatprotec. Sluzalec, A., 1990, “Thermal Effects in Friction Welding”, Journal of Mechanical Science, Vol. 32, No. 6, pp. 467-478. Zepeda, C.M., 2001, “The Effect of Interlayers on Dissimilar Friction Weld Proprierties”, Thesis (Doctor in Applied Science), Graduate Department of Metallurgy & Materials Science, University of Toronto, Canada. Vill, V.I., 1962, “Friction Welding of Metals”, AWS, New York. Ylbas, B. S. et al., 1994, “Friction Welding of St-Al and Al-Cu Materials”, Journal of Materials Processing Technology, Vol. 49, pp. 431-443.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 61-67, Jan. - Mar., 2012

doi: 10.5028/jatm.2012.04014911

Characterization of Surface Level Wind in the Centro de Lançamento de Alcântara for Use in Rocket Structure Loading and Dispersion Studies Edson R. Marciotto*, Gilberto Fisch, Luiz E. Medeiros Instituto de Aeronáutica e Espaço / São José dos Campos/SP - Brazil Abstract: We present wind data collected for ten days during the dry season in 2008 during the Murici II Campaign, which was carried out in the area of the Centro de Lançamento de Alcântara (CLA). The main goals are to better understand processes governing the wind regime in the CLA and the development of tools for analyzing the impact of wind on rocket structure and on the dispersion of pollutants released during the launch. A set of 11 aerovanes (ten at 10-m height and one at 1.5-m height) plus a sonic anemometer at 1.5-m height were deployed to measure wind speed and direction, which were stored as ten-minute data. Turbulence intensity, gust factor, and gust amplitude were computed from the available dataset. Statistical analysis shows that the wind direction is predominant from East-Northeast (ENE), with the mean vector wind direction of 60º, in agreement with the trade wind regime. The diurnal cycle of all statistical properties of the wind are strongly marked. Wind speed, turbulence intensity, and gusts are peaked at about 1000 LST. The presence of a non-diurnal cycle of four days has been noticed and might be associated with synoptic systems acting on the region. A simple heuristic formula was proposed to compute Lagrangian time-scale from Eulerian time-scale, and from which we compute the Lagragian standard deviation, a final product to be used as input in diffusion models. Keywords: Centro de Lançamento de Alcântara, Surface Wind, Rocket Load, Plume Dispersion.

INTRODUCTION Brazilian rockets are launched from Centro de Lançamento de Alcântara (CLA), whose location close to equator is privileged for launching geosynchronous satellites. Turbulence and sudden change of wind velocity may affect straightforward rockets' structure as well as its flight trajectory. The atmospheric boundary layer, the nearest atmospheric layer from surface, is the one with higher time and space variability. Furthermore, the geographic and topographic features around make the CLA a special case from the micrometeorological point of view, as it is located in the Atlantic Ocean sea board nearby a cliff about 40-m high. Detailed numerical simulations of a flow passing such a cliff have already been done by Pires et al. (2009). Surface winds and turbulence in the atmospheric boundary layer govern the forces experienced by rockets in the first few seconds of the flight. Also, when there is only a very small ___________________ Received: 17/10/11. Accepted: 05/12/11 *author for correspondence: e.r.marciotto@gmail.com - Pç. Mal. Eduardo Gomes, 50. CEP: 12.228-901 - São José dos Campos/SP Brazil

clearance between the rocket and the launch tower winds, it can cause unaccepted tilts. Immediately before the launch, the rocket is taken from the tower shelter (Mobile Integration Tower) being exposed to the wind again (Kingwell et al., 1991). Thus, not only the rockets’ launch is of concern of the aerospace meteorology, but also the maintenance of rockets in the launch pad. Following Kingwell et al. (1991), the wind factors influencing the rocket launch are, among others, flight trajectory, vehicle controllability, structural loadings on vehicle and towers, rate of salt deposition (and therefore of corrosion) on launch structures, and human and environmental protection. Ground wind studies have been carried out by NASA concerning a safety environment for both launch procedures and dispersion of toxic gases and particulate matter rising from fuel burning (Adelfang et al., 2008). Rocket fuels produce gases of varying toxicity degrees. Generally, weather conditions such as entrainment of clear air and vertical mixing must be appropriate to keep concentrations of the combustion releases within the thresholds of toxicity. In this connection,

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 69-79, Jan. - Mar., 2012

Marciotto, E.R. et al.

Moreira et al. (2011) have developed a multi-layer model to simulate rocket’s fuel-burning releases. Note that turbulence can be very critical in a go/no go decision: concomitantly to its desired effect on pollutants released during the fuel burning, there is an undesired effect on the structure of the rocket. The importance of the wind profiles for vehicle launch has also been pointed out by Decker and Leach (2005) and Decker and Barbré (2011), who have been used wind profiles as input to a computer code to generate and validate vehicle steering commands to alleviate loads on the vehicle during ascent phase of missions in connection with NASA program. In particular, with respect to the intensity of the wind in the CLA, Fisch (1999) has studied the vertical profile of the wind in CLA and provided the first calculation of turbulence and gusts there. In a recent study, Fisch (2010) has compared two different sensors (aerovane and sonic anemometer) by analyzing the mean difference of wind speed and maximum wind speed. That study aimed at contributing to a better understanding of the wind time-series as they are obtained by aerovane or sonic anemometer, in order that the modernization of the wind data at the CLA could preserve their homogeneity for climatological purposes. A more detailed study of turbulent properties of the wind around the CLA has been addressed by Magnago et al. (2010), who used data of August, in 1999 (beginning of dry season), to calculate the spectrum of u, v, and w wind components and the ratios σu/u*, σv/u*, and σw/u*; all these parameters are relevant to characterize the surface layer flow. A wider range of data covering five years (from 1995 to 1999) from a 70-m high anemometric tower was used by Gisler et al. (2011) for statistical analysis. Those studies have helped to construct a wind climatology of the CLA. Nevertheless, they do not address the use of those data for practical/operational purposes. Therefore, many questions are still open and have to be answered. For example, what is the wind profile nearby the launch pad and how does it change from ocean upstream passing by the cliff? What are the concerning characteristics of turbulence and gusts, spatial distribution, and time scales for launching a rocket? Besides questions related to the launch safety, there are still problems of the safety of the population living in the neighborhood (Alcântara Town). Then, atmospheric data can be used to verify dispersion conditions of the boundary layer. One way to do it is by means of the computation of dispersion parameters served as input to numerical models, such as the Gaussian plume. The present study has also addressed this problem. Thus, the characterization of the surface boundary layer and, particularly, the internal boundary layer, is very important. 70

In order to investigate these points, we present data of wind turbulence and gusts collected for ten days during the dry season in 2008, in the CLA.

SITE AND DATA The CLA is located at the geographic coordinates 02o19'05" N; 44o22'04" W. In this latitude, wind trades interact with the sea breeze circulation, strengthening the easterly wind during the day (Gisler et al., 2011). The launch pad is located about 150m from the sea cost, and at the edge of sea-land there is a cliff about 40-m high (Fig. 1). The data analyzed here come from an aerovane mesh at 10-m high and from another couple of aerovanes (sonic and aerovane) 1.5-m high installed in this site.

Figure 1. The 40-m high cliff upwind the Centro de Lançamento de Alcântara.

Aerovane mesh The data presented here were collected during the Murici II Campaign, in September 2008 from 16 to 25 (DOYs 260 to 269), corresponding to the local dry season. Synoptic conditions were fair with clear sky and no precipitation during the whole campaign. This experiment deployed, among other anemometers, ten aerovanes (model 05103 from R.M. Young) were displayed in a triangular mesh of masts 10-m high, and spaced 10-m from each other. Wind speed and direction were sampled at 0.5-Hz rate. Time-series of raw data were averaged over a ten-minute time interval and stored in a data-logger CR-7 Campbell Scientific Instrument. Figure 2 shows the layout of the aerovane mast mesh. The background of figure shows approximately the West view of the site so that the prevalent wind direction enters the page. Direct observations are: wind speed as given by the rotation rate of the aerovane

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 69-79, Jan. - Mar, 2012

Characterization of Surface Level Wind in the Centro de Lançamento de Alcântara for Use in Rocket Structure Loading and Dispersion Studies

Figure 2. Site studied and the mast mesh layout. The view is of West portion of the site. Roughly speaking, the Trades are entering the page.

propeller after calibration applied, and wind direction, given by the aerovane potentiometer. The aerovanes were calibrated before and after the experiment in a wind tunnel in the Aerodynamics Division at Institute of Aeronautics and Space (IAE) using a Pitot tube as standard, not showing any tendency. The zonal (u) and meridional (v) components of the wind velocity were computed from instantaneous values of the observed wind speed and direction. The vector mean wind speed was computed from mean components. Table 1 provides a summary of the calculation procedures. Aerovane direction in those formulas has been set with respect of the magnetic North, therefore a correction for magnetic declination had to be applied. The new vector mean wind direction speed and new vector mean wind speed are written as Eq. 1: - cos d - sin d ur = url ur (1) mc m Rd c m e o=c sin d - cos d vr vrl vr

Table 1.

values were taken over a 10-min time interval.

VS = ur =

vr =

1/ S N n n

Scalar mean wind speed (observed)

1 N

/- S sin i

1 N

/- S cos i

Zonal mean wind component

Meridional mean wind component

Vector mean wind direction 1

2 2 VR = ^ur + vr h2

vn Sn

Gn =

1 ir = tan- ur vr

I= where, Rδ is identified with the counterclockwise Oz-rotation matrix. The magnetic declination, δ=20o50’W (–20.833o) was obtained from NOAA National Geophysical Data Center calculator (http://www.ngdc.noaa.gov/geomagmodels/ Declination.jsp). Despite of the fact that wind components have changed due to the magnetic declination correction, the magnitude of the wind velocity is kept unchanged. The tanθ (ratio between the components) has an increment straight in the argument. Therefore, it is enough to sum δ to the old angle (θ'=θ+δ=θ-20.883º). Besides, the correction for the magnetic declination, wind direction timeseries, showed a systematic difference amongst the aerovanes. This is probably due to some aerovanes

Basic averaging methods of the observations. All mean

S pn VS

An = S pn - VS

Vector mean wind speed Turbulence intensity

Gust factor, is a 10-min peak velocity

Gust Amplitude

being a little misaligned. In fact, after subtracting the mean value of each aerovane and summing the overall averaged direction, all aerovane direction becomes consistent with each other (Fig. 3). The ten-minute mean amplitude of wind direction along the Murici II Campaign is relatively large, reaching almost 80º by the day of year 266. Note that even after correction, a random variability

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 69-79, Jan. - Mar., 2012

Marciotto, E.R. et al.

of up to 10º still remains. It is observed that in the two days of the campaign (DOYs 267 and 268) the variability of wind direction has decreased. Maximum and minimum instantaneous wind speeds were taken from each ten-minute intervals and were averaged hourly for each aerovane and then averaged over all ten aerovanes. Turbulence intensity (standard deviation over mean wind speed), gust factor (maximum wind speed over mean wind speed), and gust amplitude (maximum wind speed minus mean wind speed) were computed from the available dataset. Since the aerovane sampling rate is 0.5Hz, the time interval between two consecutive readings is two seconds and, thus, the gust we consider here has automatically a time width of two seconds. It seems that there is not a standard in defining the duration of the gust. Usually the gust duration is defined as a function of instrumentation/data availability or specific objectives and two or three seconds wind peak in a time-interval of ten minutes is often used.

Figure 3. Direction time series after correction for magnetic declination and for offsets. The symbol ‘dm’ means mean direction of each aerovane.

Sonic anemometer and aerovane at 1.5-m height We also considered in this analysis data from a sonic anemometer and aerovane at 1.5-m height. They were close to each other about 2m and about 20m apart from the aerovane mesh. They were operated at the same sampling rate as the aerovanes of the mesh, but in this case the raw data were stored. These dataset allowed us to carry out a more refined spectral analysis, which would not be possible with the ten-minute averaged mesh data. 72

ESTIMATING σy2 AS INPUT FOR DISPERSION MODELS One of the most common dispersion models is the Gaussian plume, as described in Hanna et al. (1982); Seinfeld and Pandis (1999) and Arya (1999). To apply the Gaussian model, the ground surface should be relatively homogeneous so that the calculation of the statistical properties at a given position can be used at other ones. The main statistical parameter to be determined from observations is the mean wind (easy task) and the lateral (σy2) and vertical (σZ2) variances (tough task). Conceptually, these variances can be determined from observing the diffusion of a point source, for example, and measuring the concentration with a fast response device. Then, one computes the mean value and the variance of concentration for a suitable averaging time, usually ten minutes. The difficulty in this procedure is that to get known the spatial variation of the variances, an array of devices must be setup, which is often not possible. Furthermore, measurements in real time provide a diagnostic of the concentration, instead a prognostic as is usually desirable. In the modeling study of dispersion over the CLA area carried out by Moreira et al. (2011), σy2 is expressed in terms of an empirical formula that depends on many unknown parameters in practice, such as the convective velocity scale and the Obukhov length. Both are not measured in a regular basis, and, besides, they are related to the unstable condition of the surface layer, a condition not clearly satisfied in the CLA area. For instance, Magnago et al. (2010) found that 93% of data collected in August, 1999 (beginning of dry season) lie to the stability parameter interval of –0.3 < z/L < 0.1. Thereby, the use of σy2 as it is provided by the statistical model of diffusion (Hanna et al., 1982; Arya, 1999), instead of semi-empirical formulas, would be more reliable, with the advantage that wind speed and direction are measured routinely in the CLA. According to Taylor’s approach (Arya, 1999), σy can be obtained from σV by means of the Lagrangian autocorrelation function, as in Eq. 2: t^t0, g h =

vl (t0) vl (t0 + g) 2 vv

(2)

where, ξ = t – t0. In the case of a stationary flow, the starting time is not important and the autocorrelation can be expressed only as a function of ξ Lateral concentration variance is then given by Eq. 3 (Hanna et al., 1982; Arya, 1999):

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 69-79, Jan. - Mar, 2012

Characterization of Surface Level Wind in the Centro de Lançamento de Alcântara for Use in Rocket Structure Loading and Dispersion Studies

()=

(3)

Taylor (1921) assumed an exponential form for ρ(ξ), although another form, exp(–ct2 /τL) has been suggested as well (Arya, 1999). The latter form seems not to fit our observations, because the rounded shape close to ξ = 0 is not observed. We thereby adopt the simple form assumed by Taylor: ρ(ξ) = exp(–ξ / τL), where τL is the Lagrangian time scale. Therefore, the concentration variance will be given by Eq. 4: t 2 2 2 t v y = 2v v (t) x L 8 - 1 + exp `- jB xL xL

(4)

Two well-known limit cases, also found by Taylor (1921), are: for t << τL; in this case, the autocorrelation function is close to the unit (ρ ≈ 1). And, for t >> τL, when the ρ → 0. The Lagrangian concentration variance within this limit case can be written as in Eq. 5: 2 2

v t 2 v y (t) . ) v 2 2vv x L t

for t % x L

(5)

for t & x L

All those parameters refer to a Lagrangian system of coordinates. Actual measurements are carried out at a fixed point (Eulerian coordinates). There is no general exact procedure for obtaining Lagrangian coordinates from Eulerian coordinates or vice-versa in a turbulent flow. To do this, we will make use of a heuristic approach based on Hanna et al. (1982), but somewhat modified. Consider a turbulent eddy as a circulating vortex of radius R in a flow approaching an anemometer with mean speed U. A particle at the edge of the eddy will have a tangential velocity, whose OY-component is v. The time for the particle to perform the entire eddy in the OY direction is a measure of the Lagrangian time-scale, being equal to 2R/v. On the other hand, for the anemometer to capture the entire eddy, it will take a time given by 2R/(U+u), where u is OX component of the eddy tangential velocity. Thus, the ratio τL/τE for one single eddy will be something of the order of (U+u)/v. If we take into account a large range of eddies in average (using the same time-interval to compute U=ū), we can estimate Lagrangian-to-Eulerian time-scale ratio as Eq. 6: b/

v xL = u (6) xE vv

Note that the σ u and σ v, which appear in the previous equation, are the Eulerian standard

deviations, that is, they are measured in a fixed frame of reference.

RESULTS AND DISCUSSION Wind characteristics The diurnal cycle of wind direction averaged over all aerovanes is strongly marked var-ying from 25 to 55º, that is, on the central half of the first quadrant. Its peak value occurs at about 1000 LST. (Fig. 3). Scalar and vector mean wind speed (not shown) are very close throughout the time-series, and the maximum mean difference for the diurnal cycle is not greater than about 2%. Such wind persistence is a characteristic of trade winds and already a known fact for a much larger area covering the CLA site. Hereafter, we will then refer only to the vector mean wind speed in the following discussion. Figure 4 shows ten-minute time-series for the magnitude of the vector mean velocity and the spatial average over them. A not negligible variability is observed among the aerovanes, it is not clear its origin. A misalign-ment of aerovanes would generate systematic differences between each couple of aerovanes. We have investigated those differences, and found that they are either negative or positive throughout the timeseries, i.e., no systematic difference was found. Therefore, a misalign-ment does not seem to be the answer. Furthermore, this large scattering is also observed in the scalar wind speed and suggests a length-scale of few meters.

Figure 4. Aerovane mesh-min averaged time series. Red line is the spatial average, and black symbols, individual aerovane values.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 69-79, Jan. - Mar., 2012

Marciotto, E.R. et al.

Figure 5 presents the vector mean wind speed, maximum and minimum wind, and direction averaged over all days and over all aerovanes. We have applied offsets of -3 and + 3m/s to the maximum and minimum speed, respectively, in order to enhance the diurnal cycle on the graphic. Mean wind speed presents values ranging from 5.0 to 6.5m/s. Minimum and maximum wind speed are about 3.0 and 9.0m/s, respectively. These numbers show that wind is rather strong in the CLA site. The main contribution comes from the trade winds, whereas the variation along the diurnal cycle is due to the mesoscale regime of the land-sea breeze circulation, which strengths the wind during the daytime and weakens it during nighttime.

Figure 5. Diurnal cycle of mean wind, maximum and minimum wind, and wind direction.

Figure 6. Distribution of the vector mean wind speed (above) and the vector mean wind direction (below) with their correspondent normalized Gaussian curve. Wind speed trends to be below

To complete the statistical characterization of the wind, histograms for the mean wind speed and the wind direction for the whole time series after averaging for the ensemble of all ten aerovanes were drawn in Fig. 6. Accordingly, the wind average and standard deviation are 5.8 and 1.0 m/s, respectively. For direction, average and standard deviation are 36.6º and 12.4º, respectively. Gaussian distributions based on those values have also been plotted. As it can be seen, both wind speed and direction are not symmetrically distributed about the mean value; wind speed tends to be below the mean value (skewness = 0.50) and wind direction tends to be above the mean value (skewness = –0.43). Gust and turbulence features There are still very few studies of the impact of wind on the rocket and launch tower in the Brazilian space program. Wind engineering has devoted most studies to other kind of 74

the mean value (skewness=0.51) and wind direction trends to be above the mean value (skewness = –0.45).

structures, such as transmission line towers or tall buildings. Even if the results for transmission line were extended to other vertical structures like a rocket or a launch pad, the particular features found in the CLA would likely fall out the code criteria. Gusts can affect rocket structure due to stresses caused either because of the strength or because of the frequency of the gusts (Hrinda, 2009). A particular concern is about the structure natural frequency, which may not withstand to resonance. Turbulence intensity (standard deviation over mean wind speed), gust factor (maximum wind speed over mean wind speed), and gust amplitude (maximum wind speed minus mean wind speed) were computed from the available dataset (Table 1). Time-series show that the gusts and turbulence are intensified between 0800 and 1300 LST, likely due to the interaction between the sea breeze and the Trades

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 69-79, Jan. - Mar, 2012

Characterization of Surface Level Wind in the Centro de Lançamento de Alcântara for Use in Rocket Structure Loading and Dispersion Studies

(Fig. 7). Turbulence intensity, gust factor, and gust amplitude are strongly correlated to each other: their peak values take place at about 1000 LST. Turbulence intensity is mainly associated with the wind shear, and its value is between 0.13 and 0.27. This value can validate wind tunnel simulations by Pires et al. (2009), who obtained a turbulence intensity of about 0.20 along a stream wise distance of 200m from the cliff. In fact, the mesh is closer than the cliff, ~100m, and for this distance the value obtained by Pires et al. (2009) is about 0.15. Gust factor and gust amplitude are between 1.3 and 1.8m/s, and 2.2 and 4.3m/s, respectively. This kind of comparison will be very important in the ongoing research, in order to find relationships between results from wind tunnel and field experiment. Figure 8. Comparison of CLA gust factors with Adelfang et al.’s empirical formula.

Figure 7. Diurnal cycle of turbulence and gust properties.

Figure 8 shows how the ensemble averaged ten-minute gust factor found in the CLA is distributed with respect to Adelfang et al. (2008) curve, who present a set of formulas supported in the literature to compute ten-minute gust factors for a range of heights (10 to 150m) and a range of peak winds (4.6 to 15.4m/s), based on the reference wind at 18.3m (60ft). Their results show a consistent gust factor decrease as the wind peak or height increases. At 10m, their maximum gust factor value is about 1.87 and a minimum value of 1.62, corresponding to the mean wind speeds of 2.5 and 9.5m/s, respectively. An extrapolation of such formula predicts a minimum gust factor of 1.6. This corresponds, indeed, to the maximum gust factor we have found.

The compilation of Adelfang et al. (2008) shows that, for wind peaks greater than 15m/s, the gust factor increases asymptotically with respect to the averaging time, there not being any difference between five and ten-minute averaging within 0.1% error. Other quantity that has been used to model wind peaks is the gust amplitude, defined as A = umax – u, in which umax is the ten-minute wind peak and u the mean wind in the same time interval. This is used particularly to model the vertical wind profile, and a gust amplitude of 6m/s nearby the ground has been reported. One important issue that should be tackled in the Brazilian space program is the planning of more laboratory and field experiments to characterize not only the flow regime in the CLA, but also to better understand how such a flow regime can affect launches as well as the rocket and launch tower structures themselves. The lack of acceptance criteria for wind conditions makes more difficult the use of existing data. As Adelfang et al. (2008) point out: “there is not a clear precedent from building codes to follow in recommending design risk for a given desired lifetime of structure.” In this way, the simple usage of already existing building codes can be misleading. Nondiurnal cycles Despite the fair weather along the field campaign, some variation in the surface pressure field has been observed between DOYs 263 and 265 suggesting a wave-like perturbation modulating the pressure oscillation, but not its amplitude (Reuter, 2011). According to the wind time-series a high pressure system should be placed over the ocean, causing an acceleration of the flow inland. Frequency-domain analysis

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 69-79, Jan. - Mar., 2012

Marciotto, E.R. et al.

(Fig. 9) reveals the presence of non-diurnal cycle forcings. The main peaks take place at 0.25, 1.0, 2.0, and 3.0 day–1, corresponding to periods of four days, one day, 12 hours, and 8 hours, respectively.

Figure 9. Averaged spectrum of the aerovane mesh. Frequencies of 0.25 and 1.0 day–1 are well apparent. Higher frequencies seem to be present, but not clear.

The averaged autocorrelation of the mesh data (not shown) is consistent with a diurnal cycle and another cycle of about four days, probably due to a large-scale system or still due to the limited length of the series. Those cycles are apparent in the frequency-domain representation (Fig. 9). The spectrum presents some higher frequencies with a time scale of 8 to 12 hours, which may be associated with the double-peak diurnal cycle of surface pressure (barotropic tide). However, they have low strength and are not clear enough in the spectrum. Surface pressure data supportthe scenario of a forcing of larger spatial scale acting. For instance, the surface pressure variation analyzed by Reuter et al. (2011) by means of Wavelet Transform in the same period of Murici II Campaign shows a period lying from 10 to 14h, consistent with what we have found from Fast Fourier Transform (FFT) applied to the wind data. The little difference (~2 h) between wind and pressure spectra may be due to the dynamics process itself, however we did not go through this question. Higher sampling rate data: diffusion parameters A 2-D sonic anemometer was placed close to an aerovane and it was operated during about the same time as the Murici II Campaign. Five DOYs from 264 to 269, 2008, were analyzed. Data from both measurements were recorded every 76

2s (0.5-Hz sampling rate), allowing a more detailed analysis of the wind statistics. We compared the performance between the anemometers, especially to assess the aerovane response to rapid variations of the wind. A similar study by Fisch (2010) deployed a sonic anemometer and an aerovane on 10-m high masts, showing that differences between them are rather small, being Usonic – Uaerovane= 0.3m/s, regardless the average, it is performed for one or ten minutes. In order to study the statistical wind properties applied to dispersion problems, they have to be considered with respect to the direction along to the prevalent wind, which we called Ox (wind component in this direction will be called u: longitudinal component) and the direction across it, Oy direction (wind component will be called v: lateral component). Here after throughout this subsection, u and v will mean longitudinal and lateral wind components, respectively, instead of zonal and meridional wind components. Since wind changes direction all the time, we considered the prevalent wind direction the time-averaged value over a time-interval Δt and, therefore, ur (Dt) = U and vr (Dt) = 0 . We have varied Δt from 1.5 minutes to five days and took the mean value: for example, there are 960 time-intervals of 7.5 minutes in five days, then the mean ρ is calculated as ∑ ρk / 960; for Δt=1 day we have ρ= ∑ ρk / 5; for Δt=5 days we have just ρ=ρk / 1. A common behavior observed in the autocorrelation function is that after a large enough lag it becomes negative. To compute the integral time scale we used only non-negative values of auto correlation function. Figure 10 shows the autocorrelation coefficient computed for several time intervals (= 5 days, 1 day, 12h, 6h, 2h, 1h, 30min, 10min, 7.5min, and 1min). Note that for the greatest time intervals the autocorrelation no longer approaches zero until 1-min lag. However, time-intervals of 12 hours to 5 days show an approaching to zero when longer lags are considered. For example, for a one-day time-interval, the curve crosses the abscissa only at a time-lag of about 5h. This shows that larger scales motions are captured by the autocorrelation function, but only when the plots are extended to longer time lags they become apparent. For instance, when the autocorrelation function for the time interval of five days like seen in Fig. 10 (continuous black line) is plotted for a time-lag as longer as the time interval considered oscillations about zero showing the diurnal cycle can be seen. Thereby, what is seen in Fig. 10 for the time lag up to 60s is the turbulence correlation superposed (or modulated) by larger scale motion. On the other hand,

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 69-79, Jan. - Mar, 2012

Characterization of Surface Level Wind in the Centro de Lançamento de Alcântara for Use in Rocket Structure Loading and Dispersion Studies

when shorter time-intervals are taken into account, the largest motion scales are filtered out and then the autocorrelation function carries information of the turbulence only. Figure 10 shows a significant difference between the series of 24-h time-interval (continuous red line) and a series of 2-h timeinterval (continuous navy line). The former attains cross the abscissa at 5h, whereas the latter cross the abscissa for a time lag of about 40 minutes. The shortest time-interval plottedis of 1.5min, which cross the abscissa at 12s approximately.

integrating the autocorrelation curves in Fig. 11 until the first zero.

Figure 11. Lagrangian integral time-scale: the subscript refers to the components u (longitu-dinal) or v (lateral).

Figure 10. Sonic-anemometer autocorrelation function plotted up to the time-lag of 60s.

Integral time-scale was obtained from wind measured in an Eulerian frame of reference, therefore, it is a Eulerian time-scale. To transform it into Lagrangian time-scale we use Eq. 6. The variation of the number β in relation to the time-interval is shown in Fig. 11. β was computed for sonic anemometer and aerovane. Aerovane presents systematic lower values. Usually for use in dispersion model sonic anemometer data are desirable because of their faster response to wind changes, but aerovanes are more commonly used for monitoring winds in the CLA as elsewhere. Thereby, to check its response and the wind properties measured by aerovanes were important points in the present analysis. The integral time-scale ratio (β) for short lags varies from about 1.1 (aerovane) to 1.3 (sonic). The maximum value is 1.55 (sonic) for a lag as long as five days (whole time-series). It can be noticed from Fig. 11 that β is not a parameter that will affect significantly the integral time-scale. Based on this, we present in Fig. 12 the integral time-scale of u- and v-component only for Lagrangian case. Time-scale was computed by

Figure 12. Time-scale ratio of Lagrangian and Eulerian frames of reference.

The relationship between dimensionless Lagrangian lateral standard deviation [σy /(σv τvL)] and the dimensionless time [t /τvL] for a time-interval of 10-min is shown in Fig. 13. The choice of ten minutes average time for representing the Lagrangian standard deviation is based on the fact that this is the most usual averaging time employed in Meteorology and in dispersion studies in particular. A similar plot for the Eulerian σy (not shown) cannot be distinguished by eye given the closeness of them. For the particular case of the CLA, where wind has a strong persistence, it seems reasonable to use the Eulerian variances to input dispersion Gaussian models. In

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 69-79, Jan. - Mar., 2012

Marciotto, E.R. et al.

situation with low wind and/or presence of obstacles nearby, the scenario can be rather different.

scale, standard deviation of concentration, and relationships between Lagrangian and Eulerian frames of reference. This last task, which should be deeply studied, is very import since model’s diffusion parameters are Lagrangian quantities whereas measurement are usually Eulerian quantities. The time-series we analyzed so far is still short for drawing general statements. In a future work, we would analyze longer time-series. Nonetheless, the dispersion parameters we have computed may be applied to dispersion models eventually for those days of year when they were collected or in general studies considering that the series length is short and may be not representative of other days of year. Finally, the characterization of general wind behavior and turbulence within the surface boundary layer must be a central concern in the CLA to guarantee the safety of the launching operations.

Figure 13. Standard deviation of the concentration in the lateral direction, assuming an exponential decay of the autocorrelation function. The Eulerian and Lagrangian are

ACKNOWLEDGEMENTS

undistinguishable by eye. Two main behaviors can be seen: for t<τv, σy increases as t, and for t>4τv, σy increases as t1/2.

CONCLUSIONS This study aimed at better understanding processes governing the wind regime in the CLA and at developing tools for analyzing the impact of wind on rocket structure and on the dispersion of pollutants released during the launch. We presented data of wind and turbulence collected in the CLA during a ten-day field experiment in the dry season of 2008. Wind speed, wind direction, turbulence intensity, gust factor, and gust amplitude present a marked diurnal cycle. Vector mean wind speed, minimum speed, and maximum speed are about 6.0, 3.0 and 9.0m/s, respectively. Mean wind direction is about 40º. The gust factor found for CLA (1.5 in average) is rather low when compared with literature (1.6 to 1.9). Spectral analysis of aerovane mesh shows two strength frequencies (0.25 and 1 day–1) and two lower strength frequencies (2 and 3 day–1). The lowest frequency may be associated with a large-scale system, while the 1 day–1 frequency is obviously associated with diurnal cycle. The two higher frequencies may be associated with the pressure diurnal cycle, which reaches two peaks a day due to the barotropic tide. The concern with pollutant dispersion and environmental impact reports lead us to investigate more direct methods to estimate diffusion parameters, such as integral time78

The authors thank J. Yamasaki for helping with data and metadata. The authors also thank the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) under Grant number 2010/16510-0 and the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) under Grant Universal 471143/2011-1, PQ 303720/2010-7, and the Project 559949/2010-3 for supporting this study.

REFERENCES Adelfang, S. et al., 2008, “Winds, in Terrestrial environment (climatic) criteria guidelines for use in aerospace vehicle development 2008 revision”, NASA technical memorandum 4511 (Johnson D.L., editor), Chap. 2. Arya, S.P., 1999, “Air pollution meteorology and dispersion”, Oxford University Press, New York, 310 pp. Decker R.K., Leach R., 2005, “Assessment of atmospheric winds aloft during NASA Space Shuttle Program day-oflaunch operations”, American Institute of Aeronautics and Astronautics (AIAA). Decker R.K., Barbré Jr., R.E., 2011, “Quality control algorithms and proposed integration process for wind profilers used by launch vehicle systems”, 15th Conference on

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 69-79, Jan. - Mar, 2012

Characterization of Surface Level Wind in the Centro de Lançamento de Alcântara for Use in Rocket Structure Loading and Dispersion Studies

Aviation, Range and Aerospace Meteorology of the American Meteorological Society (AMS), Los Angeles, 2011. Fisch, G., 1999, “Características do perfil vertical do vento no centro de lançamento de foguetes de Alcântara (CLA)”, Revista Brasileira de Meteorologia, Vol. 14, No. 1, pp. 11-21. Fisch, G., 2010, “Comparisons between aerovane and sonic anemometer wind measurements at Alcântara Launch Center”, Journal of Aerospace Technology Management, Vol. 2, No 1, pp. 105-110. Gisler, C.A.F. et al., 2011, “Análise estatística do perfil de vento na camada limite superficial no centro de lançamento de Alcântara”, Journal of Aerospace Technology Management, Vol. 3, No. 2, pp. 193-202. Hanna, S.R. et al., 1982, “Handbook on atmospheric diffusion”, U.S. Department of Energy Technical Report, 102 pp. Hrinda, G.A., 2009, “Single-point-attachment wind damper for launch vehicle on-pad motion”, NASA Langley Research Center, Technical Report. Kingwell, J. et al., 1991, “Weather factors affecting rocket operations: a review and case history”, Bulletim of the American Meteorological Society, Vol. 72, pp. 778-793.

Magnago, R. et al., 2010, “Análise espectral do vento no Centro de Lança-mento de Alcântara (CLA)”, Revista Brasileira de Meteorologia, Vol. 25, No.2, pp. 260-269. Moreira, D.M. et al., 2011, “A multilayer model to simulate rocket exhaust clouds”, Journal of Aerospace Technology Management, Vol. 3, No. 1, pp. 41-52. Pires, L.B.M. et al., 2009, “Studies using Wind Tunnel to Simulate the Atmospheric Boundary layer at the Alcântara Space Center”, Journal of the Aerospace and Technology Management, Vol. 1, No. 1, pp. 91- 98. Reuter, E., 2011, “O desempenho do modelo ETA: uma análise comparativa do vento entre medições in situ e o modelo para o Centro de Lançamento de Alcântara”, Tech. Report, related to the CNPq Ph.D. program at the Instituto de Pesquisas Espaciais (INPE), agosto, 2008, 14 pp. Taylor, G.I., 1921, “Diffusion by continuous movements”, Proceedings of the London Mathematical Society, Vol. 2, pp. 196-211. Seinfeld, J.H, Pandis, S.N., 2006, “Atmospheric Chemistry and Physics”, John Wiley & Sons, Inc., New Jersey.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 69-79, Jan. - Mar., 2012

doi: 10.5028/jatm.2012.04015111

Study of a Lower-deck Galley for Airliners Marcelo Vieira Abritta1, Jürgen Thorbeck 2, Bento Silva de Mattos1,* Instituto Tecnológico de Aeronáutica - São José dos Campos/SP - Brazil Institut für Luft- und Raumfahrt der Universität Berlin - Berlin - Germany

1 2

Abstract: The present work is concerned with the preliminary design of a new galley concept for long-haul airliners. Usually, galley systems are placed in the main deck of airliners or even in some cases in the lower-deck cargo compartment. The present concept considers placing trolleys, components that occupy significant room in galley installations, in the lower-deck compartment and transport them to the passenger deck by a dedicated lift system. The main advantage of this proposal is that more room becomes available in the passenger cabin for the accommodation of additional passengers. By a careful analysis, which considers the required structural modifications that must be incoporated into the airplane configuration to accommodate the new concept, the payoff of the present proposal is investigated. This was carried out by using the PADLAB® 2.4 software package, written for MATLAB®, and an in-house routine to evaluate aircraft performance and calculation of direct operating costs per seat mile. PADLAB® is tailored to the design of the cabin and the device and systems aimed at operating the trolleys by the passenger cabin crew; the in-house routine was validated against data obtained for the category of airliners under consideration. Keywords: Aircraft Design, Airplane Interior, Aircraft Performance, Airliner.

LIST OF SYMBOLS AND ABBREVIATIONS BOW

Basic operating weight

CLmax

Maximum lift coefficient

CAD

Computer-aided design

CRT

Cathode ray tube

DOC

Direct operating cost

EADS

European Aeronautic Defense and Space

EFIS

Electronic flight and instrument system

FAR ILR

Federal Aviation Regulation Institut für Luft - und Raumfahrt der Technischen Universität Berlin

MTOW

Maximum takeoff weight

OEW

Operational empty weight

PAX

Passengers

TAT

Turnaround time

INTRODUCTION The present work has proposed and analyzed a new galley configuration for long-haul airliners. The galley is the ___________________ Received: 11/11/11. Accepted: 03/02/12 *author for correspondence: bmattos@ita.br/Pç. Mal. Eduardo Gomes, 50. CEP: 12.228-900 - São José dos Campos/SP - Brazil

compartment of a ship, train or aircraft, where food is cooked and prepared. The Douglas Aircraft DC-3 was the first airplane with a planned galley for food service. Galleys on commercial airlines typically include not only facilities to serve and store food and beverages, but also flight attendant jump seats, emergency equipment storage, as well as anything else flight attendants may need during the flight (Wikipedia, 2012). The activities utilizing galleys can be basically divided into two categories: preparation for the next flight, and the remaining activity being characterized by flight attendants’ duties when the airplane is airborne. Galleys are also used before flight, when the passenger boarding is not cleared yet. They have a strong impact on passenger’s evaluation of the service that is provided by the flight attendants. The galley of the airplane has a crucial impact on TAT at the airport. In this context, it is of primary concern how long it takes from the moment passengers leave the cabin up to the time the passenger cabin crew is ready to service the next batch of customers. The ground service team at the airport shall clean the cabin and cater the aircraft in the shortest possible time, under stringent safety rules. In this area, there is provision for garbage disposal and communication equipment. There is usually an extra seat, in the shape of a revolving chair, for the comfort of the cabin crew during a flight.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 81-94, Jan. - Mar., 2012

Abritta,M.V. et al.

Aircraft in operation today mainly use the familiar trolley system. This system was introduced in the late 1960s, at the same time a new generation of large wide-bodied airplanes were entering into service with the airlines. The significantly larger number of passengers on these aircraft meant that meals could no longer be efficiently delivered by hand, as they had been up to that point (Wikipedia, 2012). Since then, galley concepts did not change significantly and could not be considered efficient anymore for the current air transportation scene (Fig. 1). Major drivers for a design review are workload for cabin crew is high; galley area is not pleasant for passengers and safety must be improved.

other items, and equipment for in-flight service other than additional stowage room on the upper portion. The equipment usually consists of ovens, trash compactors, and kettle for hot water. More than one of those facilities is necessary to hold the catering cargo required to serve all passengers. Galley location in the passenger cabin is defined by their total loading capacity. A typical galley is able to accommodate up to eight trolleys, which can stow food and beverage for approximately 120 passengers. There is already a trend to accommodate some facilities in the lower deck of long-haul airliners. The four-engine Airbus A340 airplane is an example. Some Airbus A340-600, which are operated by Lufthansa, accommodate in their lower deck lavatories, galleys and crew rest areas (Schliwa, 2000). Access to those facilities is provided by stairs and lift allow trolleys from going from the main to the lower deck and vice versa (Fig. 2).

Figure 1. This picture of an Airbus A-340 galley compartment reveals how inappropriate the galley design of current airliners can be (Wikipedia, 2012).

Many airliners are fitted with more than just a galley installation. The number of galleys is strongly dependent on the seating capacity of passengers. Typical figures for sizing are employed according to the numbers of passengers that a single galley can service. For smaller single-aisle airplanes, ranging from regional jets to Boeing 737 and Airbus A320, the main galley is usually located at the rear part of the fuselage, with an auxiliary smaller galley at the forward part, close to the cockpit. For two-aisled airplanes, the position may vary a lot. However, the composition of the galley, regarding its mechanical components and specially its geometrical size, is not heavily modified. This happens because in double-aisled cabins the galley cannot occupy the whole diameter of the fuselage, being restricted to the area within the two aisles. Galleys usually accommodate from six to eight trolleys, which are stowed and loaded when necessary. In addition, galleys have just a small area for preparation of food or 82

Figure 2. Lufthansa Airbus A340-600. Lower-deck lavatory entrance.

J. Aerosp. Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.4, No 1, pp. 81-94, Jan. - Mar, 2012

Study of a Lower-deck Galley for Airliners

Both the 777-300ER and 777-200LR Worldliner offer overhead crew and attendant rest areas in the fuselage crown above the passenger cabin (Fig. 3). Most airplanes have crew rest areas either in the passenger cabin or in the cargo compartment. By moving crew and attendant quarters off the main deck, 777 operators can free as many as four-to-seven revenue passenger seats (Boeing, 2011a). Alternatively, using overhead crew rest areas frees up room for additional capacity in the cargo compartment, up to six LD-3 containers. This revenue-generating capability is another innovation that the competitor’s airplane, the A340, cannot match because of the A340’s constrained cross-section design (Boeing, 2011a).

Figure 3. Boeing offers 777 customers new innovations, like the crew resting area in the overhead space of fuselage (Boeing, 2011a).

Trolleys need considerably large areas in order to be accommodated in galley installations. Therefore, the present concept focuses on them. They shall be accommodated in the lower deck instead, and be transported to the passenger deck by a special purpose lift system (Fig. 4).

GALEY DESIGN The methodology that was employed for the design of the new galley system is described in this section. In order to evaluate the impact of the new concept on airplane performance and its general characteristics, a computational code was developed and validated. Suited airplanes to the new concept Galley areas of main passenger decks were compared with different types of airliners and cabin layouts. The higher galley area-to-cabin-area ratios were found for wide-body airliners, such as the Boeing 777 (Fig. 5). These airplanes feature high-capacity double-aisle passenger cabin and are designed for long-haul flights. The duration of typical flights of such airplanes requires that two or more meals be served to a higher number of passengers. This usually results in a greater area of main decks for food and beverage stowage, approximately 25% for the typical economy class area of a B777-300 airliner (Boeing, 2011a). Galleys for the business and first classes present a greater volume per seat dedicated for stowage. In this case, items served to passengers are unique and require longer time for their preparation and handling. First and business classes would benefit rather from beverage storage on the lower deck than that for food. It is important to point out that the presence of first and business classes in the cabin makes the allocation of the galley on the lower deck less effective. Smaller density in occupation of the passenger deck implies a smaller total volume of passenger luggage, releasing room in the cargo compartment for automated galleys. Smaller aircraft that operates with low-cost airlines may also benefit from the lower-deck galley concept. If one

Figure 4. CAD model of the lower-deck galley lift system. J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 81-94, Jan. - Mar., 2012

Abritta,M.V. et al.

Figure 5. Boeing 777-200 two-class typical cabin layout (Boeing, 2011a).

considers that galleys occupy a smaller portion of the pax cabin area, the high-seat density that they present could eventually support the required investment and pay-off. Low-cost airlines record high-occupation rates, and their policy is marked by very restrictive luggage allowances, allowing the cargo compartment to be easily reconfigured for the automated galley concept. This would also be very simple due to the highly limited service provided by this kind of airline to passengers during the flight. However, due to the size of cargo compartment of mid- and small-size airliners that low-cost airlines operate, the new galley concept does not fit them. Cargo compartments of airplanes, like the Embraer E-190/195, are not able to accommodate the required trolleys and the associated elevator system.

many meals are needed for a flight and the number of meals that one trolley can accommodate. Also, the calculation of the necessary volume of food and beverages requires the knowledge of how galleys are operated when the airplane is airborne, in order to identify parameters that might influence the location of the galleys inside passenger cabins. Factors that may impact TAT must also be taken into account (Fig. 6).

General considerations of galley design Galley configuration also impacts ground operations, as well as in-flight services. The utilization of current trolleys could simplify catering, avoiding longer TAT if new trolley designs are introduced. In addition, the use of current trolley designs would not require any big adaption of cabin crew to new concept. The current equipment used in galleys must also be compatible with the automated galley, in order to reduce installation costs and avoid new certification issues. It would be mandatory that the present galley concept be easily installed and removed from affected airplanes; therefore, increasing their residual value. The understanding of in-flight service logistics is crucial for the design of any new galley system, which shall provide improved service and reduce the workload of flight attendants. In order to achieve these goals, it is essential to pinpoint the way meals are prepared and served to passengers. The total number of trolleys can be simply calculated by defining how 84

Figure 6. Boeing 777 of KLM prior departure, being catered and supplied with external electric power.

The required number of trolleys is strongly dependent on passenger capacity. The number of trolleys that are needed for a service round at the main deck must be defined. A service round is comprised of attending passengers with snacks, lunch, dinner, drinks and refreshments, and even shopping products. Considering that the majority of the trolleys will not be available in the passenger cabin, according to the new galley concept, the sequence that they must be utilized in the pax cabin is important to their positioning in the lower-deck cargo compartment and to the design of mechanism system, which will bring them to the upper deck.

J. Aerosp. Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.4, No 1, pp. 81-94, Jan. - Mar, 2012

Study of a Lower-deck Galley for Airliners

Usually, a single sandwich or a small snack pack is served to each passenger. Given the relatively low storage space that is necessary to accommodate these items, and the fast distribution procedure, usually one flight attendant can cover one aisle with one trolley. Some airlines serve snacks together with drinks or refreshments. This way, just one trolley will be used per aisle, requiring two flight attendants to do the job. For long-range flights, each passenger will be provided with one pack containing cold items, usually consisting of salad, small juice, and dessert, as well as disposable utensils. These items were prepared by catering companies and arranged inside trolleys, no additional effort is needed by the cabin crew. Usually, airlines offer two options for hot meals, which are categorized in general terms as “Pasta or Beef” or “Chicken or Fish”. Hot meals are heated in ovens before being offered to the passengers. For this, the meals are covered with protective packing. A single oven is able to heat two batches of food usually consisting of 25 packs for the economy class, or 16 ones for the first or business class. Ideally, a separate and backup oven is used for items, such as bread that are usually served as well. After that, the cabin crew arranges meals on the upper part of trolleys and service starts to passengers. For wide-body airliners, usually two trolleys with two flight attendants, each one will cover one aisle. In fact, this arrangement is suited to 120 passengers at most. If a passenger asks for a second hot meal, flight crew will usually deny until all passengers were served. However, if a passenger asks again, a second hot meal will be served if available. After a major meal is provided, usually there is a round of water and beverage service and then cabin crew starts collecting disposables. This is a simple process, it basically consists of stowing garbage and disposables in the trolleys that were employed before for meal distribution. Glasses and cups are kept on the upper surface of the trolley. Further disposables are later collected by using bags and no trolleys, with a single flight attendant covering one aisle. For the drinks and refreshments service, two trolleys shall be used for a single aisle, with two flight attendants operating them. Some drinkables are kept cold with the use of cold ice. Modern airliners are equipped with chillers, while most of the older airplanes use chillers for items such as dairy products and fruit slices for garnish, which will not hold quality if heated. Many options are usually available for different drinks, and then few or any options within each type of drink. For instance, passengers can choose from a range of up to four sodas, three juices, one red wine, one white wine, two kinds of beer, milk, and water. Each passenger will be served with approximately

200mL – unless additional water or beverage is asked for. Water and beverage service is usually performed before snacks or hot meals are served. Sometimes water and juice are offered together with snacks, and it is also served right after hot meals are distributed among passengers. In average, there are two drink offerings for each hot meal or snack round. Frequently, an additional third round of drinkables is served to passengers. Galley layout In this section, the new galley concept will be better described as well as its associated equipment and operation. The main goal of the concept is to provide room for additional passengers in the airplane. It is mandatory that all the necessary food and beverages for servicing passengers be available at the main deck, but not necessarily at the same time. Food for dinner and lunch can be stored on the lower deck while snacks are being served, reducing the size of galleys and providing extra space for revenue passengers. Using basic cargo containers of known size and small adjustments to fit the machinery used for allocation of the trolleys on the lower deck, a sample, preliminary design of the new galley concept was generated with the CAD package CATIA®. Major geometric design constraints are the width of the galley in the main deck, which is bounded by aisles; and the size of existing cargo containers. Thus, the number of trolleys that can be stored side by side on the passenger deck is limited by the size of galley and the size of the containers placed in the cargo compartment. PADLAB® (Fig. 7) is a software package used for airplane cabin design modeling that was developed at ILR. PADLAB® is able to deal with different kinds of geometric constraints for any airplane configuration, enabling users to easily customize their designs. Thus, a module to consider the trolley storage in cargo compartments as well as its associated lift system was easily incorporated into the PADLAB® package.

Figure 7. A340-300 modeled with PADLAB®.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 81-94, Jan. - Mar., 2012

Abritta,M.V. et al.

The lower deck geometry and its sizing are tailored to accommodate standard cargo containers. There are many containers of different sizes and shapes. Our concept chose the LD-26 one (Fig. 8). This container presents a suitable capacity and it would be the safest option for housing the necessary machinery, which will operate on the lower deck.

Figure 10 shows the shared area between the lower and main passenger decks and Fig. 4 provides an idea of how trolleys can be brought to galleys on the main deck. In Fig. 10, the area dedicated to the elevator is displayed. If a single elevator is employed, it would be necessary to control the position of the trolleys on the lower deck using a bi-axial system, which would compromise the level of service and would increase the workload of the cabin crew. The optimal configuration consists of the same basic layout, but using a multi-lift system to cover the full extension of the lower deck area, so that trolleys are moved in a single direction inside the cargo compartment.

Figure 8. View of LD-26 container (Driessen Cargo Equipment).

The LD-6 container shall suffer some modifications to accommodate the trolleys in organized storing positions: an opening on its upper surface is necessary to allow trolley to be lifted to the main deck; an important aspect that has to be considered is the required structural reinforcements to support elevator operations; the incorporation of a system to lock the trolleys into their storage positions; and the incorporation of a movement mechanism to allow trolleys to reach the elevator in an appropriate fashion. Trolleys are kept in special stowage hub, so that they can be easily handled by a feeding mechanism system in the lower deck. As can be seen in Fig. 9, there is a space below the floor that supports the trolleys. This area will house the necessary equipment to move the trolleys from their parked positions to the elevator, and vice versa.

Figure 8. View of LD-26 container (Driessen Cargo Equipment).

Figure 10. Main passenger deck and lower deck area of a modified airplane that incorporated the new galley concept.

Structural considerations In order to keep development and manufacturing costs as low as possible, it would be highly convenient that the new galley structural layout in the main passenger deck does not depart significantly from current designs. The frame is built using aeronautical aluminum and low-density polymers in order to reduce structural weight. The structural design is usually modular, based on the dimension of the equipment that the galley should enclose. According to the quantity of the specific equipment of the galley and taking into account geometric constraints imposed by the cabin sizing and geometry, the galley structure is possible to be defined. Taking this into account, a design tool for the frame structure of the galley was developed. The basic dimensions of the galley are input and calculations that provide specific coordinates for the galley assembly using a CAD tool. The designed frame is very simple, composed basically of the bottom area for trolley storage and the main working surface for the in-flight service. To the extent, the

J. Aerosp. Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.4, No 1, pp. 81-94, Jan. - Mar, 2012

Study of a Lower-deck Galley for Airliners

design becomes more complex; items are to be added to the frame structure. Preliminary design of aircraft interiors could greatly benefit from such tool. Based on that demand, a MATLAB® Graphic User Interface was developed and could represent a good further development for the ILR’s PADLAB® tool. Basic galley equipment design was based upon Driessen Aircraft System galley equipment. The company has agreed to provide basic information regarding its models’ geometry and technical specifications. That information has been used as a background for the CATIA® models, so that the design can be consistent in terms of dimensions. Elevator system The necessary technology for the elevator system that was envisaged has been already employed with Boeing 747. The main deck of this airliner is used to store trolleys to be employed for servicing the upper deck. This takes place when airplane is on ground. If the airplane becomes airborne, trolleys shall be in the deck where they will be utilized. This procedure poses no major effort of the cabin crew. Similar concept is used for the Airbus A380 double-deck airliner. The operation procedure is the same as that for the Boeing plane, where two trolleys are placed in the elevator. Yet, on the A380, the ground service team can reach the upper and lower decks through service doors on the fuselage, making the re-stocking process simpler to use (Stilp, 2006). Yet, there remains the high occupation of both decks by the galley kitchen, since all trolleys are still on the deck even when they are not being used. According to the new concept proposed by the authors, trolleys shall be brought from the lower deck to the pax cabin during flight, before and after their used. All this must be accomplished in a reasonable period. However, the existence of trolleys that are already employed in similar tasks contributes to reduce costs and time of development. Based on the technical data available for the current dualtrolley elevator (Jenoptik, 2007), it is possible to estimate the weight and capacity of the new equipment. The estimation was carried out considering a parametric approach, evaluating the impact of the number of trolleys on the engineering features in order to keep operational conditions. Table 1 displays the results of this estimation. The critical factor for the weight estimation is the load capacity of the elevator. It must be taken into account that all trolleys will be fully loaded to their maximum capacity, when they are being moved to the main deck. The tare weight of

the trolley is about 25kg, resulting in a total 125kg for five trolleys. Hence, elevator capacity can be estimated as 475kg. Table 1. Weight estimation for the system of the elevator.

Data for the dualTrolley Elevator

5 Trolley Lift Estimation

Weight (elevator)

< 320kg

< 800kg

Weight (trunk)

< 60kg

< 150kg

Speed

0.2 to 0.4m/s

Load

240kg during flight

600kg during flight

480kg on ground

1,200kg on ground

Handling mechanism The mechanism system for the lower deck is the only item that is not currently under aeronautical operation in commercial aircraft. However, there are similar applications under operation for freighter jets. Once it has been defined that the lower deck machinery system should allow movement in a single direction, the axis of the airplane, the ideal solution is one that can move the group of trolleys as a whole. A simple solution is a revolving chain to be attached to the trolley groups in order to hold their position and allow their movement on the container platform. This system will operate in a horizontal direction, so it can be assumed that it will have a smaller impact on weight than the trolley elevator system. Motors and required structure will be considerably lighter, since they are stable on top of a surface and must not hold the total weight of the trolleys. It is hard to pinpoint the weight of such a system. This would require the full development of the material, and such task is beyond the purpose of a preliminary design of the lower deck galley. However, estimations are to be considered as well as a sensitivity analysis regarding the estimation parameters. A good estimation can be achieved by determining the weight of the total lower deck equipment as the sum of the tare weight of the container and items belonging to the mechanism system. Equipment weight can be estimated co-relating it to the elevator weight, considering that the operation is restricted to horizontal displacement, as described. The tare weight of the container, however, is already known. By varying the fraction of weight of the lift that is assumed to represent the horizontal positioning control system, it is possible to determine the total increase of weight produced by

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 81-94, Jan. - Mar., 2012

Abritta,M.V. et al.

each occurrence of the lower deck equipment. Based on these estimations, the total weight of the lower deck system may vary from 400 to 900kg (Table 2). Table 2. Estimation of equipment weight (kg).

Percentage of elevator weight 80%

60%

40%

20%

Equipment weight

640

480

320

160

Equipment + Container

890

730

570

410

System operation Initially, it was thought to introduce a fully automated new galley system. This includes automated use of ovens for heating meals and of automated system to displace items in lower deck, such as cold drinks from the refrigerator to the trolleys. This approach could contribute to lower DOCs, considering that it would require fewer flight attendants. For instance, a fully automated system would require considerable knowledge of Robotics and Control Theory. In addition, any new system has to comply with aeronautical certification rules, besides presenting low weight and extreme high reliability. For these reasons, a simpler concept was chosen, which attains the same basic goal of allowing extra room for passenger allocation on the main deck. The approach that was adopted consists of keeping trolleys in the cargo compartment, while equipment such as ovens, water heaters and others remain in the pax cabin. Cabin crew will manually put meals inside the ovens for their heating. Garnishing would be carried out by the flight attendants, as well as cold drink distribution using trolleys. Thus, lower deck equipment will be substantially reduced. Basically equipment that is needed consists of an elevator to move trolleys from the lower to the main deck, and of a horizontal controller, or, maybe, a matrix oriented controller, to move the trolleys inside the lower deck. Also, there would be more considerations regarding special stowage needs for the lower deck, which would also have to be fully automated and highly reliable to comply with safety requirements. For this option, the main factor to determine the operation of the galley is the elevators that will transport the trolleys between the lower deck and the main cabin. Current elevators that operate with Boeing 747, between the main deck and the upper cabin in the forward part of the airplane, and also the ones that operate on A340 with lower deck galleys with access to flight attendants operate lifting two trolleys at a time. This number is not the optimum 88

possible, as most of the trolleys will have to be lifted, moving two at a time would take a long time, both during flight service and in ground preparation, impacting turnaround time. Lower deck positioning of the trolleys would also have to be much more complex, since there would be only one place where all the trolleys could go in order to be taken to the upper deck. One possibility is to employ from two to three elevators for each galley. This would decrease the time that is needed to bring trolleys up and down, and simplify the moving criteria on the lower deck, due to the multiple positions of elevators. Yet, it would cause a significant increase in weight to the project, and, more importantly, in necessary equipment. This would lead to a less efficient occupation of the galley area on the lower deck and also in the container. An optimal configuration can be achieved by developing a new elevator, which is capable of transporting six trolleys to the main deck. This way, the required time to move all necessary trolleys is reduced as well as the complexity of the moving mechanism on the lower deck, since all six trolleys could be moved as a group, removing the necessity for a matrix oriented allocation system, replaced by a single direction operator. The drawback lies in the necessity of the development of a new tool, yet, due to its simplicity, it would probably not represent a high demand on research. Using this configuration, each galley could provide service to 150 passengers, which is considerably more than the typical figure of 120. This could boost efficiency for the overall cabin crew service. Slots of up to six trolleys could be prepared by the ground operation team to accommodate each round or service. At each round, the galley would send to the main deck only the trolleys required, keeping the others on the lower deck, meanwhile moving them to accommodate the upcoming disposals, when the current service is completed. A door to allow ground personal to load and unload trolleys without entering the airplane is fully feasible, but it was not considered. Deeper structural calculations are required for its design and integration into the airplane configuration. These issues are beyond the scope of the present work.

MTOW AND BOW ESTIMATION Besides the equipment related to the new galley concept, structural reinforcements are required to accommodate the increased payload. Payload increase could top three tones for airplanes like the Boeing 777. Thus, MTOW will vary if payload is increased and/or structural weight is added to the

J. Aerosp. Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.4, No 1, pp. 81-94, Jan. - Mar, 2012

Study of a Lower-deck Galley for Airliners

configuration. The lift and drag coefficients will also vary and they will impact range, takeoff, and climb performance. All this will lead to additional fuel to fulfill the mission. Furthermore, everything changes and the new airplane parameters must be computed. For this purpose, a computational code was developed in MATLAB® language. The code was Christianized WEST, which iteratively calculates the MTOW and other core weight figures. Airplane component weights are calculated according to Roskam’s Class II methodology (Roskam, 1985). Calculation of aerodynamic coefficients also fits into a Class II approach (Roskam, 1985). Some component weight calculations were carried out using methods developed by Torenbeek (1982). Figure 11 presents the workflow utilized for WEST. In the present work, engine was modeled into a simpler way, with the specific fuel consumption being provided for some flight phases. Thus, no sophisticated engine deck was elaborated. Vertical and horizontal tail stabilizers areas were obtained using the tail volume coefficient approach (Roskam, 1985). WEST was employed for weight estimation of the following airliners: Airbus A340-300; Boeing 777-200 and Boeing 767-200.

A module was added to WEST in order to incorporate the impact on component weights caused by the incorporation of the new galley concept. An extra 2.5 tones accounted for the systems and structural modifications. This represents an overestimation of the galley components total weight, and it also encompasses two lower deck galley installations in the same aircraft, if necessary. Airbus A340-300 The Airbus A340-300 is a long-range four-engine wide-body commercial passenger airliner designed and manufactured by Airbus, a subsidiary of EADS (Fig. 12). It seats up to 335 passengers in a two-class layout (the stretched 600 series carries 440), and it has a range between 12,400 and 16,600 km. It is very similar in configuration and systems to the twin-engine A330 with which it was concurrently designed. Initial A340 versions share the fuselage and wing of the A330, while later models are longer and have larger wings. Over 370 A340s are in operation worldwide as of September, 2010.

Figure 11. WEST code workflow for MTOW and OEW estimation. J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 81-94, Jan. - Mar., 2012

Abritta,M.V. et al.

service with United on September 26, 1982. Certification with Pratt & Whitney engines was awarded on July 30, 1982.

Figure 12. Airbus A340-300 profile.

Table 3 displays the estimation carried out for weight estimation for the A340-300 with the WEST code. Table 4 contains the results of a sensitivity study of passenger capacity for the A340-300 series. The extra weight due to the lower deck galley concept was considered in the model.

Figure 13. Boeing 767-200 side view.

Table 5 displays the validation effort carried out for the Boeing 767-200 MTOW estimation. Error margins are acceptable, considering that Class II methodology was employed for the present calculations. The estimated empty weight is now slightly higher than that for the actual airplane (Boeing, 2011b). Table 6 shows the results of the sensitivity study carried out for Boeing 767-200. The increase in the required fuel is now higher than that calculated for the A340300, especially for the 20% passenger increase. If both airliners are able to transport 20% more passengers thanks to the new galley concept, the Boeing 767-200 will require 6.5% more fuel while the Airbus aircraft will consume 9% more kerosene. However, considering that the Boeing aircraft carries typically 200 passengers and the Airbus airplane about 300, the impact of payload increase on DOC could reveal another trend. The results signalize potential DOC reduction, provided that the increase in passenger capacity is always higher than the associated increase in the mass of the components. The results of the DOC analysis will be later presented.

Table 3. A340-300 weight estimation with WEST code (all values in tones).

Empty weight MTOW Mission fuel

Airbus data 126 251.7 72.5

WEST results 123.8 246.1 70.0

Error (%) -1.8 -2.2 -3.4

The increase in the required fuel to fulfill the mission falls close to the error margin as seen in Table 4. The increase observed for MTOW and OEW values takes into account the structural modifications to accommodate the increased payload and new galley concept. Boeing 767-200 Launched in July, 1978, the Boeing 767 (Fig. 13) was developed in tandem with the narrow body 757 with which it shares a common two crew EFIS flight deck (with six color CRT displays) and many systems. The 767 also features a unique width fuselage typically seating seven abreast in economy, and a new wing design with greater sweepback than that of the 757 wing. The 767 first flew on September 26, 1981, and entered

Table 5. Boeing 767-200 weight estimation (all figures in tones).

Boeing data 80.1 142.9 29.9

Empty weight MTOW Mission fuel

WEST results 81.5 138.9 28.8

Error (%) -1.8 -2.8 -3.9

Table 4. Sensitivity analysis for increasing the passenger capacity of Airbus A340-300 (values in tones).

5% Passenger capacity OEW MTOW Mission fuel

10%

15%

20%

Total

Increase

Total

Increase

Total

Increase

Total

Increase

305

319

334

348

128.1 252.8 72.4

3.5% 2.7% 3.5%

128.4 255.2 73.4

3.8% 3.7% 4.9%

128.5 256.7 73.9

3.9% 4.3% 5.5%

128.7 258.6 74.6

4.0% 5.1% 6.5%

J. Aerosp. Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.4, No 1, pp. 81-94, Jan. - Mar, 2012

Study of a Lower-deck Galley for Airliners Table 6. Sensitivity analysis for increasing the passenger capacity of Boeing 767-200.

5% Passenger number Empty weight (t) MTOW (t) Mission fuel (t)

10%

15%

20%

Total

Increase

Total

Increase

Total

Increase

Total

Increase

210

220

230

240

85.3 142.9 29.4

4.6% 2.9% 2.4%

85.6 144.4 29.8

4.9% 4.0% 3.7%

86.1 146.8 30.5

5.5% 5.7% 6.1%

86.1 147.3 31.3

5.5% 6.1% 9.0%

Boeing 777-200

Table 7. Boeing 777-200 weight validation with the WEST code.

Boeing 777 (Fig. 14) is the world’s largest twinjet. The airplane offers seating for over 300 passengers and has a range from 5,235 to 9,380 nautical miles (9,695 to 17,370 km), depending on the model. It isable to accommodate 550 passengers in a single-class cabin layout. Its distinguishing features include the largest diameter turbofan engines of any aircraft, six wheels on each main landing gear, a circular fuselage cross-section, and blade-shaped tail cone. Developed in consultation with eight major airlines, the 777 was designed to replace older wide-body airliners and bridge the capacity gap between the 767 and 747. The 777 was Boeing’s first fly-by-wire airliner.

Boeing data

WEST results

Error (%)

Empty weight

139.2

136

-2.3

MTOW

247.2

240

-2.9

Mission fuel

71.2

69.7

-2.0

Once the model has been pre-qualified for the analysis, the lower deck galley is implemented on the design, and then the tests are made considering the possible results in terms of passenger capacity increase. Table 7 illustrates the results obtained. The additional fuel for B-777 if payload is increased is considered when compared to the remained models studied here (Table 8). This can be credited to its higher capacity and range.

PASSENGER CAPACITY INCREASE AND DOC CALCULATION Figure 14. Boeing 777-200ER in Air France livery.

Figure 15 displays the rear portion of Airbus A340-300 passenger cabin, modeled with PADLAB®. Outer airplane skin was set transparent in order to generate a cutaway of the airplane interior. This is an example of the PADLAB® output file for CATIA. The program generates a complete aircraft layout for the entire cabin. Specifications, such as economy

For the 777, the WEST code slightly underestimated MTOW, OEW, and the required fuel mass (Table 7). However, the accuracy is very satisfactory considering the low fidelity approach of the modeling; all errors fall below 3% for the three airliners studied here. Table 8. Results from the B777-200 weight impact analysis.

5% Passenger number Empty weight (t) MTOW (t) Mission fuel (t)

10%

15%

20%

Total

Increase

Total

Increase

Total

Increase

Total

Increase

383

402

420

438

141.2 248.4 72.6

3.8% 3.5% 4.1%

142.6 255.0 75.5

4.9% 6.2% 8.3%

143.2 255.5 76.8

5.3% 6.5% 10.1%

144.0 262.6 78.6

5.9% 9.4% 12.8%

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 81-94, Jan. - Mar., 2012

Abritta,M.V. et al.

class seats, seating pitch, and the number of seats abreast, can be defined by users using a user-friendly graphical interface. The code generates an output file, which is read by a CATIA® macro code, producing the CAD surfaces as shown in Fig. 15.

Figure 15. A340-300 aft pax cabin as modeled by PADLAB®.

Figure 17. Original cabin layout of the Airbus A340-300 airliner (top) and the additional seats that were added after the

New cabin layouts were elaborated incorporating the proposed galley concept. They were useful for the evaluation of the increase in passenger capacity thanks to the new concept. Figures 16 to 18 show the original cabin layout compared to its reconfiguration after the new galley concept is incorporated into the airplane configurations, which were studied here. Table 9 shows how many additional passengers can be transported in the main cabin of the three airliners studied in the present work after changing the galley configuration. The galley concept described in the present work proved to be advantageous for the three airliners that were studied here (Table 10). DOC was calculated using the methodology suggested by Roskam (1985).

incorporation of the new galley concept (bottom).

Figure 18. Original cabin layout of the Airbus Boeing 767-200 airliner (top) and the additional seats that were added after the incorporation of the new galley concept (bottom).

Figure 16. Original cabin layout of the Boeing 777 airliner (top) and the additional seats that were added after the incorporation of the new galley concept (bottom).

The new galley concept can provide a significant increase in passenger’s capacity for the three long-haul airliners studied. The payload increase and its associated structural reinforcements are largely justified by the associated DOC reductions that were obtained. In addition, there will be a more healthy work environment in the passenger cabin. Airplane ownership cost will increase with the incorporation of the new galley system, but the lower DOC per seat mile will

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 81-94, Jan. - Mar, 2012

Study of a Lower-deck Galley for Airliners

overcome this. DOC per seat mile improvement ranged from 11.8 to 15.4%. Table 9. The present concept enables increasing the number of passengers that can be transported in the main cabin of three major airliners.

A340-300

Boeing 767-200

Boeing 777-200

Typical Economy class capacity

238

175

280

Additional seats in the economy class

+11.8%

+15.4%

+12.1%

Seating capacity variation

Table 10. The galley concept described in the present work proved to be advantageous for the three airliners that were studied.

A340-300

Boeing 767-200

Boeing 777-200

DOC per seat mile improvement

-4.17%

-5.42%

-3.25%

Variation ofSeating capacity

+11.8%

+15.4%

+12.1%

galley products for future airliners. Airbus SPICE innovative catering equipment (Airbus, 2012) can be utilized in the design of trolley, and galley equipment and structure enable an additional reduction of the system overall weight, enabling yet better DOC figures. Detailed design of the mechanism system for the trolley displacement inside the cargo compartment is needed to enable a better weight estimation of the new system and its impact on the overall airplane weight, performance, and operating costs. This is a hard task, considering that the concept is new. Certification issues may require additional reinforcements to the airplane structure as well as its systems. It is likely that this would be the critical stage of the design and creation of a prototype, once all the other necessary equipment is available on the market. Another area that could be very sensitive for the success of the concept presented on this report is that of safety regulations, especially the evacuation requirements. FAR 25 regulations require that all passengers must be able to evacuate the cabin within 90 seconds. Santos (2004) utilizes software simulation to evaluate the possibility of evacuation from wide-bodied aircraft equipped with a lower deck passenger seating area. It is possible that the same routine can be adapted for the calculation of the evacuation time for the models described, providing important data to support further research on this subject, or indicating problems that must be solved.

REFERENCES CONCLUDING REMARKS The new galley concept can provide a significant increase in passenger’s capacity for the three long-haul airliners studied. The payload increase and its associated structural reinforcements are largely justified by the associated DOC reductions that were obtained. In addition, there will be a more healthy work environment in the passenger cabin. Airplane ownership cost will increase with the incorporation of the new galley system, but the lower DOC per seat mile will overcome this. DOC per seat mile improvement ranged from 11.8 to 15.4%. In order to reduce development and manufacturing costs, the present work was restricted to the conceptual design and it used available equipment of existing galleys. However, there is room for improving the concept by the incorporation of new technologies and smarter designs. Further development of the lower-deck trolley allocation is the utilization of modern

Airbus, 2012,“Airplane Innovative Catering Equipment”, EADS Airbus, Retrieved in January 20, 2012, from http:// www.airbus.com/innovation/well-being/inside/spice/. Boeing Co., 2011a, “Boeing 777-200LR/-300LR/ Freighter, Airplane Characteristics for Airport Planning,” Boeing Commercial Airplanes, Retrieved in September 20, 2011, from http://www.boeing. com/commercial/airports/acaps/777_23.pdf. Boeing Co., 2011b, “Boeing 767, Airplane Characteristics for Airport Planning,” Boeing Commercial Airplanes, Retrieved in August 22, 2011, from http://www.boeing.com/commercial/ airports/acaps/767.pdf. Jenoptik, 2007, “Cart Lift Systems for World Air Carriers,” Hamburg: s.n., .

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 81-94, Jan. - Mar., 2012

Abritta,M.V. et al.

Roskam, J., 1985, “Airplane Design,” Vol. I-VIII, University of Kansas.

Stilp, T., 2006, “A380 Ground Handling – A Review of Past Milestones,” Orlando.

Santos, M. C., 2004, “Passenger Evacuation Process Simulation for a Wide Bodied Aircraft with a Lower Deck Seating Compartment,” Undergraduation thesis, Technological Institute of Aeronautics, São José dos Campos.

Torenbeek, E., 1982, “Synthesis of Subsonic Airplane Design,” Dordrecht: Kluwer Academic Publishers.

Schliwa, R., 2000, “Making use of the Lower-deck Area.” Hamburg.

Wikipedia, “Galley (kitchen)”, 2012, Retrieved in January 29, 2012, from http://en.wikipedia.org/wiki/ Galley_%28kitchen%29.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 81-94, Jan. - Mar, 2012

doi: 10.5028/jatm.2012.04015011

Proposal of a Methodology of Stakeholder Analysis for the Brazilian Satellite Space Program Mônica Elizabeth Rocha de Oliveira*, Leonel Fernando Perondi Instituto Nacional de Pesquisas Espaciais/São José dos Campos/SP - Brazil Abstract: To ensure the continuity and growth of space activities in Brazil, it is fundamental to persuade the Brazilian society and its representatives in Government about the importance of investments in space activities. Also, it is important to convince talented professionals to place space activities as an object of their interest; the best schools should also be convinced to offer courses related to the space sector; finally, innovative companies should be convinced to take part in space sector activities, looking to returns, mainly in terms of market differentiation and qualification, as a path to take part in high-technology and high-complexity projects. On the one hand, this process of convincing or, more importantly, committing these actors to space activities, implies a thorough understanding of their expectations and needs, in order to plan how the system/organization can meet them. On the other hand, if stakeholders understand how much they can benefit from this relationship, their consequent commitment will very much strengthen the action of the system/organization. With this framework in perspective, this paper proposes a methodology of stakeholder analysis for the Brazilian satellite space program. In the exercise developed in the article, stakeholders have been identified from a study of the legal framework of the Brazilian space program. Subsequently, the proposed methodology has been applied to the planning of actions by a public organization. Keywords: Brazilian Space Program, Stakeholders Analysis, Management of Organizations.

INTRODUCTION Modern management concepts indicate that the success of companies and organizations will be greater the better is their relationship with customers, suppliers, employees, shareholders, and community (Svendsen, 1998; Fleisher and Bensoussan, 2007). These actors, technically known as interested parties or stakeholders, will cooperate with the organization more effectively if they are convinced that their needs and expectations, in their relationship with the system/ organization, will be fulfilled. The resulting commitment will ensure the sustainability of the organization, i.e., its economic survival in a competitive environment. This applies both to private and public organizations. In societies with deficiencies in basic sectors, such as health, education, security, employment, among others, the survival of public organizations is related, additionally, to their operational effectiveness and to their capacity of attracting funding, especially in the space sector, where returns are ___________________ Received: 21/10/12. Accepted: 14/01/12 *author for correspondence: monica.rocha@dir.inpe.br/Av. dos Astronautas, 1758 - CEP: 12.227-010 São José dos Campos/SP - Brazil

not so easily perceived by society and public agents, who, ultimately, are responsible for the buget allocation. With the increasing range of applications of space technologies, the justification for the existence and maintenance of space activities go beyond geo-political and scientific reasons, to include also economic, commercial, and social issues. In addition to financial resources, institutions responsible for space activities also depend on the existence and availability of other factors, which are also essential, such as skilled workforce and highly qualified suppliers, among others. Therefore, for a country to develop space activities, beyond convincing society and Congress representatives of the importance of these activities, it is also necessary to convince talented professionals to place space activities as an object of their interest; the best schools should also be convinced to offer courses related to the space sector; finally, innovative companies should be convinced to take part in space sector activities, looking for returns, mainly in terms of market differentiation and

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 95-105, Jan. - Mar., 2012

Oliveira, M.E.R. and Perondi, L.F.

qualification, as a path to take part in high-technology and high-complexity projects. On the one hand, this process of convincing or, more importantly, committing these actors to Space Activities, implies a thorough understanding of their expectations and needs, in order to plan how the system can meet them. On the other hand, if the stakeholders understand how much they can benefit from this relationship, their consequent commitment will strengthen the system in a way that the organization could never do it alone. Based on these observations and hypotheses, this paper proposes a methodology for the analysis of stakeholders of the Brazilian Space Program, specifically those related to the Instituto Nacional de Pesquisas Espaciais (INPE). The stakeholders were identified from a careful scrutinizing of the legal framework of the Brazilian Space Program. Once the stakeholders were identified, answers to the following questions were sought for: What are their interests? What are their influence over the system/organization? Which are the opportunities and challenges related to the functions that the stakeholders expect from the system/organization? These answers may, in principle, be used to define and balance requirements for the system/organization, as well as to design and plan functions to be implemented. In conclusion, the main objective of this study has been to identify stakeholders of the Brazilian Space Program – satellite by-product, and then to illustrate the potential of stakeholder analysis as applied to the design and planning of functions to be performed by a public system/organization in the space sector, aimed at meeting the needs and expectations of its stakeholders. In this analysis, although the identification of stakeholders has been carried out with some accuracy, the design and planning of functions have been carried out via a free exercise by the authors, and not by objective evidence produced through interviews or other methodologies.

STAKEHOLDERS - CONCEPTS, CLASSIFICATIONS, AND METHODS OF ANALYSIS The identification and evaluation of the needs of the totality of actors interacting with a system/organization is not a new subject in the areas of Business Ethics and Organization Theory. Freeman (1988) proposed the concept of stakeholder as any group or individuals who can affect or be affected by a company. Examples include shareholders, creditors, 96

managers, customers, suppliers, local community, and general public. This concept has been further elaborated by Weiss (in Branco et al., 2009), who consider that stakeholders are individuals, groups or organizations, which can influence the stages of development of a company. Several authors dealing with this subject argue that the environment of a company is heavily conditioned by the action of stakeholders (Araújo Jr., 2008; Freeman, 1988; Svendsen, 1998; Boaventura and Fischmann, 2007). According to Araújo Jr. (2008), analyzing a stakeholder is equivalent to analysing the environment in which the company operates, identifying the main actors in this environment and, then, evaluating the influence of these actors on the company and on the environment itself. Branco et al. (2009) state that a system/ organisation exists to generate value for its stakeholders and that the functions performed by the system/organization, most valued by its stakeholders, must be the basis to develop the system/organization. Thus, the stakeholder analysis, comprising the processes of identifying stakeholders, understanding their needs and evaluating their influence over the system/organization, may contribute significantly to the design and planning of the functions to be performed by the system/organization. In this article, such an application of a stakeholder analysis will be demonstrated in the scope of the Brazilian satellite space sector. According to Wood (1990), stakeholders can be classified into two main categories: primary – owners, customers, suppliers, employees and competitors – and secondary– internal and foreign governments, media, community, nonprofitable organizations, financial analysts and institutions. In this paper, it will be proposed a classification adapted to the specificities of a public institution, containing ten categories. Some authors have developed models for stakeholder analysis, as in Boaventura and Fischmann (2007), Freeman (1984), Weiss (2009), Svendsen (1998) and Carroll and Buchholtz (2000). The initial step of the analysis, according to these authors, is the identification of stakeholders. As a criterion for identifying a stakeholder, Mitroff (Boaventura and Fischmann, 2007) proposes that each stakeholder shall present at least one relevant characteristic for the system/ organization, among the following: aims and motivations; benefits or potential benefits; the resources they control (material, political, and skills); distinctive knowledge; legal commitments or others; and relationship with other stakeholders in terms of power, authority, accountability, and credibility.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 95-105, Jan. - Mar., 2012

Proposal of a Methodology of Stakeholder Analysis for the Brazilian Satellite Space Program

These definitions, concepts and methods of analysis form the basis of the methodology of stakeholder analysis developed in the following sections.

the potential utility of a stakeholder analysis, however, interests of each category of stakeholder, and the ranking of these interests, will be conjectured; for each category of stakeholders a figure of merit representing the level of interest of that category in the system/organisation will be computed through a simple arithmetic average of the weights attributed to interests in the previous step;

PROPOSED METHODOLOGY FOR STAKEHOLDER ANALYSIS According to Boaventura and Fischmann (2007), the models of stakeholder analysis, when used for analyzing a system/organization, usually comprise the following steps: identification of stakeholders; description of their interests, policies, and behaviours; identification of the level of their satisfaction; identification of how they can influence the sector, considering the possible interactions between stakeholders; assessment of the impact of each stakeholder in the sector/ industry; and hierarchical classification of the stakeholders. In this study, the proposed methodology for stakeholder analysis for the Brazilian Space Program, satellite by-product, will follow the directives of Boaventura and Fischmann, and will comprise the following steps: • identification of stakeholders, from an analysis and study of the legal frameworks of the Brazilian Space Program, satellite by-product; •

classification of the identified stakeholders into more general categories to facilitate the process and analysis; the organization in categories will be made according to criteria based on the dimensions proposed by Mitroff, described to identify stakeholders;

• identification and ranking of the interests of each category of stakeholders; the set of interests of each category of stakeholders is obtained by simple agglutination of the interests of all stakeholders in that category; interests correspond to the aspects the stakeholder value most and expect to be fulfilled in its relationship with the system/ organisation; each interest will be ranked according to its relative level of importance from the standpoint of the stakeholder, by assigning a weight according to the following scale: 1 - very little importance; 2 - little importance; 3 - fair importance; 4 - great importance; and 5 - very great importance; ideally, this step should be implemented through structured interviews with representatives of each category of stakeholders; in this study, concerned mainly with a demonstration of

•

identification and ranking of the influences of each category of stakeholders on the system/organization, as perceived by the system/organization; the set of influences of each category of stakeholders is obtained by simple agglutination of the influences of all stakeholders in that category; what each category of stakeholders has to offer to the system will be estimated, and the relative level of importance of these assets to the system/organization will be ranked; the ranking will be implemented using the same scale as used for ranking the interests of stakeholders; ideally, this step would be accomplished through structured interviews, carried out within the organization, with skilled staff; as in the previous step, however, possible influences of each category of stakeholders over the system/organization, and the ranking of these influences, will be conjectured, in a free exercise; for each category of stakeholders a figure of merit representing the level of influence of that category on the system/organization will be computed through a simple arithmetic average of the weights attributed to influences in the previous step;

• hierarchical classification of stakeholders, considering the levels of interest and influence estimated in the previous steps; since the ranking of interests and influences of categories of stakeholders, in previous steps, has been accomplished through a free exercise, the classification that emerges from this step has to be seen as the result of a free exercise, as well; • identification of the functions to be performed by the system/organization in order to meet the expectations of the stakeholders; the interests and influences of each category of stakeholders, identified in previous steps, will guide the design and planning of functions within the system/ organization; the process of designing and planning functions should be carried out making the best possible use of the opportunities made available through the influences of each stakeholder category; finally, for implementing

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 95-105, Jan. - Mar., 2012

Oliveira, M.E.R. and Perondi, L.F.

this step, it is necessary to evaluate the already existing management processes in the system/organization and to identify, for each function to be performed, those already implemented, those to be implemented for the first time, as well as the opportunities for improvements. Next, as a demonstration of the application of this methodology to space activities, it is developed an exercise in which the proposed methodology is applied to the Brazilian satellite space program.

Table 1. Legal framework of the Brazilian Space Program related to satellites. Decree No. 51.133,

Creation of the Organizing Group of

from 08/03/1961

the National Commission of Space Activities (GOCNAE).

Decree No. 68.532,

The Institute for Space Research (INPE)

from 04/22/1971

is created in substitution of the CNAE.

Decree No. 68.099,

Creation of the Brazilian Commission

from 01/20/1971

for Space Activities (COBAE), with the

EXERCISE OF APPLICATION OF STAKEHOLDER ANALYSIS TO THE BRAZILIAN SATELLITE SPACE PROGRAM In this section, the proposed methodology, described in the previous section, is used in the identification of categories of stakeholders of the Brazilian Space Program, satellite by-products. Its applicability to the design and planning of the functions of a system/organization is demonstrated through an exercise, in which interests and influences of each category of stakeholders are conjectured and ranked, and used in modelling the processes in a system/organization to meet the needs and expectations of each category of stakeholders. The exercise of application of the proposed methodology follows the steps described previously, although, as already advanced, the information related to the steps of identifying and ranking interests and influences has been conjectured, rather than obtained through structured interviews with representatives of categories of stakeholders and of the system/ organization. INPE has been taken as the system/organization of reference for this exercise. Identification of stakeholders and grouping in general categories Stakeholders of the Brazilian Space Program – satellite by-products – were identified from a careful reading and studying of the legal framework to the Brazilian Space Program related to satellites. Table 1 lists the set of documents used for this purpose. In total, 42 stakeholders explicit or inferred from the text were identified, who were grouped into ten different categories, as follows: Federal Government and Congress, Industrial Sector, Education, Science, International Partners, Staff, Security, Media, and Brazilian Society. This categorization 98

responsibility of establishing policies and programs related to space activities in Brazil. Decree No. 1.332,

The National Policy for Development

from 12/08/1994

of Space Activities (PNDAE) and the Brazilian Space Agency (AEB) are established.

Decree No. 1.953,

The National System for development

from 07/10/1996

of the Space Activities (SINDAE) is established.

National Program of

The current ten-year planning for the

Space Activities 2005-

Brazilian Space Program is established.

2014

has been developed based on the criteria of performance area and distinctive knowledge, i.e., stakeholders with major intersection in terms of performance area and distinctive knowledge, from the perspective of the system/organization, were gathered in the same category. Table 2 displays the stakeholders identified and their categorisation, according to the directives given above. Additionally, a few stakeholders, that have not been identified from the legal framework but that can have a competitive attitude in relation to the system/organization, such as foreign space technology competitors and foreign commercial competitors, are also displayed in Table 2. Interests and influences of stakeholders The interests of a stakeholder have been identified with the expectations of the stakeholder in relation to the system/

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 95-105, Jan. - Mar., 2012

Proposal of a Methodology of Stakeholder Analysis for the Brazilian Satellite Space Program Table 2. Stakeholders of the Brazilian Space Program, satellite by-product. Federal Government and Congress

1. Presidency of the Republic; 2. Legislators; 3. Brazilian Space Agency (AEB); 4. Ministry of Science, Technology and Innovation; Ministry of Environment and other ministries; Secretary of the Republic Presidency and Inter-ministerial Commissions; 5. States, Federal Districts, and Cities.

Industrial Sector

1. Suppliers; 2. Industrial Sector benefitted by industrial policies; 3. Industrial Sector as suppliers of products or specialized services; 4. Industrial Sector as costumers of the facilities of INPE for qualification of its products or services.

Education – related to INPE’s post-grad activities

1. National and International Entities for technical and scientific cooperation; 2. Research and Education Fostering Agencies; 3. Conferences, symposiums and other national and international academic meetings; 4. Technical and scientific publications; 5. Technical and scientific national and international events; 6. Post-grad students; 7. Other institutions for team training.

Science – related to the INPE’s

1. Conferences, symposiums, and other national and international academic meetings;

scientific and research activities 2. National and international scientific organizations; 3. Technical and scientific publications; 4. Technical and scientific community; 5. Researchers and technologists; 6. Support Institutions for scholarships; 7. Post-grad students.

International Partners

1. National and International Entities of technical and scientific cooperation; 2. Space Agencies of other countries; 3. Foreign Companies with (potential) commercial relationship with Brazil; 4. Partners International Institutions.

Staff

1. Skilled Human Resources; 2. Researchers and technologists; 3. Institutions for team training; 4. Human Capital – Staff of INPE.

Media

Brazilian Society

1. Magazines, newspaper and television, as well as other technical and scientific publications; 2. Instruments for disclosure of results and services provided by the Institute. 1. Skilled Human Resources; 2. Costumers of space products, weather forecasting, global climate changes, and other products and services provided by the Institute; 3. Brazilian Society.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 95-105, Jan. - Mar., 2012

Oliveira, M.E.R. and Perondi, L.F. Table 3. Interests and influences of stakeholders of the Brazilian Space Program of INPE. Stakeholder

Interests

Influence

Federal

Popular approval; national and international political support; Laws to facilitate the mechanisms of hiring innovative

Government

achievement of goals and budget execution; to provide products products and services; approval of funds; hiring staff approval;

and Congress

and services for the Brazilian society - data and technology; international agreement for cooperation in space activities; international agreements; political visibility; industrial establishment of priorities for space activities; political competitiveness and jobs; monitoring of Brazilian natural incentives for development of innovative activities; creation of resource.

conditions for building capabilities, as new universities, etc.

Industrial

Jobs; new business; productivity and profitability; improvement Jobs; skilled labor; autonomy to develop projects and

Sector

of industrial infrastructure; improvement of infrastructure support programs; commitment with the quality, schedule, and costs (research and testing laboratories); new technologies; contracts with in the development of space products; investment of private the public sector; development of challenging projects; spin-offs. resources for developing projects and programs; response to the technological demands for the development of new projects.

Education

Training; scientific indicators; growth of scientific and Skilled labor; Scholarship; emphasis in disciplines related to technological knowledge; creation of infrastructure in schools and space activities. universities; inspiration for young people.

Science

Technologies to meet scientific demands; scientific indicators; Stock of scientific and technological knowledge; conferences participation in national and international conferences; research and other scientific conclaves related to the space sector; skilled interesting to agencies and industries; stock of scientific labor. knowledge; scholarship.

International

International political support; commercial opportunities; sharing Training of qualified personnel; technology transfer;

Partners

knowledge; cooperation for development of strategic products; partnerships for exchange of technologies; supply of qualified services that Brazil can provide for countries with smaller components and equipment; financial resources; international experience in space technologies.

Staff

Jobs; best salaries; good working conditions; professional Skilled labor; experience and capability; dedication; challenges; training/education.

Security

discussions on matters related to the sector.

commitment with the objectives of the organization.

Cooperation for the development of satellite launch vehicles; Creation of demand for new products and projects for national technology to develop products and services for military use; defense; financial resources for new products and projects monitoring of the Brazilian territory; technological autonomy for related to national security. development of projects of interest to national security.

Media

News of interest to the general population and/or political impact; Encouraging public opinion; dissemination and relationship credibility.

with the general public.

Brazilian

Jobs; national pride; defense of sovereignty; generation of services Skilled labor; popular support; demand for products and

Society

for general society; monitoring of natural resources; inspiration services related to space activities; taxes. for the young people.

Countries

Supremacy; national security concerns; prevention of the Export control of information and material pertaining to defence

Leaders

development of competitors in exploitation of space activities.

in Space

and military related technologies, including parts and materials qualified for space uses.

Technology Countries

Interfere in the growth of the participation of Brazilian industries Export control of information and material pertaining to defense

that compete

in the international trade; interfere in the development of the and military related technologies, including parts and materials

economically

Brazilian economy.

qualified for space uses.

with Brazil

100

J. Aerosp. Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.4, No 1, pp. 95-105, Jan. - Mar., 2012

Proposal of a Methodology of Stakeholder Analysis for the Brazilian Satellite Space Program

Hierarchical classification of stakeholders according to the level of interest and influence on the system/organization An overall figure of merit representing the importance of each category of stakeholders for the system/organisation has been obtained by multiplying the figures of merit computed for interests and influences, and is given in Table 5. Classifying the set of category of stakeholders according to their importance for the system/organisation results in the ordered list shown in Table 6. The information thus obtained may be taken as an important tool for the management of the system/organisation, and may be used to align the objectives of the organization with the needs and expectations of the most important stakeholders, which, depending on the situation, may be either those that have more interest in the system, or those that can affect most the system, or else those best scored in a balanced average, that considers both the interests and the influences. It is important to note, that care must be exercised when defining the balance between interest and influence. This point may be illustrated by the exercise considered here, in which

Table 4. Assignment of factor for interests and influences related for each group of stakeholder.

Federal Government and Congress

Main interests

Factor

Popular approval

National and international political support

Achievement of goals and budget execution

Provide products and services for Brazilian

society – data and technology International agreements Political visibility

3 3

Industrial competitiveness and jobs

Monitoring of Brazilian natural resources

Average Influence over the system/organization Laws to facilitate the mechanisms of hiring

Federal Government and Congress

organization under perspective. The identification of interests occurs concomitantly with the identification of the needs that the stakeholder would like to see addressed in its interaction with the system/organization. The influence, in turn, is related to the power of the stakeholder in affecting the organization, directly or indirectly, through actions in its sphere of influence. As already mentioned, in an actual application of the methodology, the identification of interests and influences shall be accomplished through interviews with representatives of each category of stakeholders and representatives of the organization. In this work, the interests and influences for each group of stakeholders were conjectured in a free exercise. Table 3 gives the result of the free exercise carried out in the scope of this work. Each interest or influence listed has been ranked according to its level of importance from the standpoint of the stakeholder or system/organization, respectively, following the weights given in the section dealing with the methodology. Table 4 exemplifies the process of assigning weights for one of the ten groups of stakeholders. Table 5 displays, for each category of stakeholders, figures of merit, representing the level of interest and the level of influence of that category on the system/organisation, computed by averaging the weights, for both interests and influences, exemplified in Table 4.

3.50 Factor 5

innovative products and services Approval of funds

Hiring staff approval

International agreement for cooperation in space

activities Establishment of priorities for space activities

Political incentives for development of innovative

activities Creation of conditions for building capabilities,

as new universities, etc. Average

categories of stakeholders have been ranked using the product of the scores obtained for interests and influences. Using figures of merit defined in this way, a category of stakeholders that has a high rank as regards the level of influence or the level of interest may get a low rank as regards the level of importance for the system/organisation. For instance, in the current example, “countries leaders in space technology” ranked first, as far as the level of influence is concerned, and eighth, as far as the overall level of importance is concerned. Thus, in an environment in which effects of stakeholders’ actions, on operation of the system/organization, are critical, it would not be appropriate computing the overall level of importance by the product of the scores obtained for interests and influences. Rather, in such a situation, the level of importance should be computed using a different relationship between interests and influences, which reflects the relative importance between these two factors in the considered

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 95-105, Jan. - Mar., 2012

101

Oliveira, M.E.R. and Perondi, L.F. Table 5. Order of the groups of stakeholders, considering the interests/

Table 6. Hierarchical classification of stakeholders according to the

influence over the system/organization. Stakeholders

Score

level of interest and influence over the system/organization.

Stakeholders ordered by

Score

Hierarchical classification

Ordering – interests versus

ordered by

level of influence over

of the identified groups of

influence

level of interest

the system/organization

stakeholders

Science

3.71

Federal Government and

4.00

Congress 2

Federal

3.50

4.00

and Congress Industrial

3.33

Sector 4

Brazilian

14.00

Congress

Staff

Government

Federal Government and

Countries leaders in

4.00

space technology 3.33

Industrial Sector

3.75

Society

Science

13.62

Staff

12.80

Industrial Sector

12.49

Brazilian Society

10.82

International Partners

10.50

Staff

3.20

Science

3.67

Education

9.99

International

3.00

International Partners

3.50

Countries leaders in space

9.32

Security

8.25

Media

6.68

Countries that compete

3.00

Partners 7

Education

3.00

Education

3.33

Security

2.75

Brazilian Society

3.25

Media

2.67

Security

3.00

2.33

Media

2.50

1.50

Countries that compete

2.00

10 Countries leaders in space

technology

economically with Brazil

technology 11 Countries that compete

economically with Brazil

environment. In a generic scenario, one may take the overall level of importance as a general function of the weights for effects and interests relative to each particular stakeholder. Functions that the system/organization must perform to meet the needs of its stakeholders As a final step of the exercise, several functions that the system/organization must perform to meet the expectations of its stakeholders were identified. Again, we emphasize that these functions are fictitious, since all data inputs were conjectured rather than obtained through interviews or other methodologies. 102

Continuing the exercise, the functions were then classified according to their level of implementation within the system/organization, following these criteria: "Implemented" – functions already performed by the system/ organization; "Improving" – functions implemented, but that need improvement to better meet the expectations of the stakeholders of the system/organisation; "To Implement" – functions that the system/organization does not perform and need to be implemented. Table 7 illustrates the functions that the system/ organisation must perform to meet the needs/expectations of some selected categories of stakeholders. The same procedure has to be carried out for all categories of stakeholders. After completing the whole process so far indicated, there will emerge a list of functions that the system/organisation must consider in order to meet the needs/expectations of its diverse categories of stakeholders. By considering the hierarchical list of stakeholders, according to their importance for the system/organization, one may finally order

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 95-105, Jan. - Mar., 2012

Proposal of a Methodology of Stakeholder Analysis for the Brazilian Satellite Space Program Table 7. Functions that the system/organization must perform. Functions that the system/organization must perform to meet the

Implemented Improving To Implement

needs/expectations of stakeholders x

Create mechanisms for helping technological areas to create demands for scientific areas. Science

Encourage employee participation in conferences and other scientific conclaves in

order to keep them current on best practices. x

Create mechanisms for dissemination of scientific works developed within the Institution Promote conferences, workshops, and other scientific conclaves for discussing best

practices in areas interesting to the Institute.

Invest in the improvement of post-graduate courses offered by INPE .

x x

Brazilian Society

Expand the Visitor Center in order to better meet (in quantity and quality) students and the general society.

To improve exchanges between Institute-Schools-Universities to inspire young people in some fields, such as astronomy, space engineering, etc. Promote forms of dissemination of research, projects, products and services to the

Create mechanisms for qualification of suppliers, mainly to ensure autonomy in technologies

Countries leaders in space

general society.

strategic technologies. x

Make efforts for creating/setting legislation that allow more efficient means for contracting the development of innovative and complex products and services. Expand the prospective activities of the Institute, to define new projects and new

technologies to be explored and developed. Expand the relationship with countries leaders in space technologies to present the

predominately peaceful characteristics of space activities in Brazil.

these functions according to the priority with which they should be addressed. Finally, in conclusion to the exercise, after mapping the functions that the system/organization must perform to meet the needs/expectations of its stakeholders and after evaluating the level of implementation and prioritisation of each of these functions within the organization, an action plan for the system/organisation may be developed.

CONCLUSIONS In this article, an application of stakeholder analysis to the Brazilian satellite space program has been proposed and

exemplified through an exercise. In the exercise developed in the article, stakeholders were identified from a careful scrutinising of the legal framework of the Brazilian space program. Subsequently, the proposed methodology has been applied to the design and planning of functions to be performed by a system/organization in the space sector, in order to better attend the needs and expectations its stakeholders. The exercise showed that the proposed methodology seems to be feasible and may allow institutional gains by expanding the vision of the system/organization, such as to include the external actors interacting with the system/organization. This expanded vision of the system/organisation has then been shown to have the potential of being used as

J. Aerosp. Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.4, No 1, pp. 95-105, Jan. - Mar., 2012

103

Oliveira, M.E.R. and Perondi, L.F.

animportant basis for the definition of functions to be performed by the system/organization, in order to best attend the expectations and needs of the identified categories of stakeholders. The exercise also demonstrated the feasibility of identifying and meeting the needs and expectations of the stakeholders of public institutions, taking the legal framework, inside which the public institution is defined, as a basis for the identification of the different categories of stakeholders. As a final conclusion, we remark that the stakeholder analysis might be a powerful tool for an institution whose operation relies on a large number of internal and external actors, such as skilled human resources, leading suppliers, and major financial investment.

Costa Filho, E. J., 2000, “A política científica e tecnológica no setor aeroespacial brasileiro: da institucionalização das atividades ao fim da gestão militar – uma análise do período 1961-1993”, Campinas, SP. Fleischer, C. S., Bensoussan, B., 2007, “Business and competitive analysis methods: effective application of new and classic methods”, Pearson Education: FT Press, New Jersey. Freeman, C., 1988, “Innovation and the strategy of the firm”, In: Freeman, C., The economics of industrial innovation. Harmondsworth: Penguin Books Ltda.

REFERENCES Araújo Jr., J.P., 2008, “Análise de stakeholders: um estudo exploratório”, Revista Eletrônica de Educação e Tecnologia do SENAI-SP, Vol. 2, No 4, Retrieved in February 6, 2012, from http://revistaeletronica.sp.senai.br/index.php/seer/ article/view/30/41. Boaventura, J.M.G., Fischmann, A.A., 2007, “Um método para cenários empregando stakeholder analysis: um estudo no setor de automação comercial”, Revista da Administração: São Paulo, Vol. 42, No. 2, pp. 141-154. Branco, M.S.A. et al., 2009, “Stakeholder value analysis of architecture alternatives for sustainable space systems developments”, Sixth International Aerospace Congress IAC’09, Moscow State University, August, pp. 23-27.

104

Carroll, A. B., Buchholtz A. K., 2000, “Business & Society, Ethics and Stakeholder Management”, 4th ed., South-Western Educational Publiching, Ohio.

Gremaud, A. P. et al., 2003, “Manual de economia”, 4ª ed., São Paulo, SP: Saraiva. Svendsen, A., 1998, “The Stakeholder strategy profiting from collaborative business relationships”, 1st edition, San Francisco: Berret-Koehler Publishers Inc. Weiss, J. W., 2009, “Business Ethics: A Stakeholder and Issues Management Approach”, 5th ed., Mason, Ohio: South-Western Cengage Learning Forth Worth, USA: Dryden Press. Wood, D. J., 1990, “Business and Society”, Pittsburg: Haper Collins.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 95-105, Jan. - Mar., 2012

doi: 10.5028/jatm.2012.04015111

An Overview of the Certification of VSB-30 with Emphasis on Technological Innovation Antonio Ramalho de Souza Carvalho1,*, José Henrique Damiani1, Andrea de Oliveira Netto Follador1, Marcelo Guido de Oliveira Guimarães2 Instituto Tecnológico de Aeronáutica/São José dos Campos/SP - Brazil Instituto de Fomento e Coordenação Industrial/São José dos Campos/SP - Brazil

1 2

Abstract: This article is focused on the space sector, mainly due to the devices developed and marketed that require high investments in research, development and technological innovation, supported by a permanent need to meet the standard requirement to guarantee their effectiveness, among all the correct certification. In such scenario, this article has examined, highlighting technological innovation, which are the context and elements involved in the certification of the Vehicle Survey Booster – 30 (VSB-30). The research is characterized as a case study, conducted in the first semester of 2011, and it consisted of reviewing the literature on technological innovation and certification as well as information available in various reports, some of which were public, and interviews. The article presents a brief description of the development of the VSB-30, its importance to the market and the relevant aspects of the certification and the Certification Body of Brazil. The conclusion is that the certification is a contribution to technological innovation for it provides benefits in process improvement, especially regarding the question of documentation, creating conditions for the industry to adapt to formal established and qualified processes and, in return, to become restricted to trade companies unable to adjust to these requirements. Keywords: Space Systems, Technological Innovation, Aerospace Certification.

INTRODUCTION The search for autonomy, qualification and industrial competitiveness, and also to give something in return to society in the space sector is a triad established in the National Program of Space Activities (PNAE), with the objective to gain technological independence in the space sector in Brazil, once the technology will be reversed to society, regardless of the origin being military or civilian. It is a major way to join the leading countries in the modern space conquest. One scale of this conquest is presented in Chart 1, in which Brazil has progressed to level four. It is not only about technological independence in the space sector, but also the search for competitiveness in the national aerospace industry, thus making it less technologically dependent, so that knowledge and technology (pillars of the knowledge society) be equipped, and strengthening the industry into one of the supporting pillars of society. ___________________ Received on: 01/11/11. Accepted on: 27/01/12 *author for correspondence: ramalhosjc@gmail.com - Pç. Mal. Eduardo Gomes, 50. CEP: 12.228-901 - São José dos Campos/SP - Brazil

The national aerospace industry exists in a complex market, in which the developed and sold devices require high investments in research, development and technological innovation. It is constantly subjected to meeting technological criteria, in which the acquisition of technologies and processes is refused, so they need to be developed without subsides of countries who have such knowledge. With the perspective to overcome these restrictions, Landini and Cabral (2005) reported the technological transfer and cooperation in the Brazilian Space Program. According to the authors, countries like Brazil are forced to develop sensitive technologies in case they want to concrete their programs. In the Brazilian case, there was the need to conceive, project and build processes and products with the country’s own resources. Among the products of the space sector, there are small satellites, flight equipment, sounding rockets, launching vehicles, services of satellite imaging application, propulsion, among others (Brazilian Association of Aerospace Industries, 2011).

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 105-115, Jan. - Mar., 2012

105

Carvalho, A.R.S. et al. Chart 1. Scale of countries in the modern space conquest. Level

Corresponding objective

Countries or multinational groups

Ten

Manned landing in Mars, Phobos or Deimos.

None

Nine

Permanent base on the moon, enabling visits to objects that are

None

close to the Solar System. Eight

Moon landing, with continuous orbital presence.

The United States of America

Seven

Independent ability to send astronauts into the space

Russia and China

Six

Ability to train astronauts and conduct scientific missions.

Europe and Japan

Five

Independent ability to launch satellites to the Earth’s orbit.

India and Israel

Four

Existence of a National Space Agency with its own satellites.

Argentina, Australia, Brazil, Canada, South Korea, Iran, Nigeria, Pakistan, Taiwan and Ukraine

Three

Existence of a National Space Agency without its own satellites.

South Africa, Saudi Arabia, Argelia, Bangladesh, Chile, North Korea, Indonesia, Malaysia, Peru, Thailand and Turkey

Two

Existence of Science Academy and human resources related to

Armenia, Belarus, Singapore, Colombia, Croatia,

space.

Georgia, Lebanon, Mexico, New Zealand, Servia and Venezuela

One

Existence of observatories, planetarium and/or astronomy clubs.

Albania, Azerbaijan, Bahrein, Belize, Bolivia, Bosnia, Brunei, Congo, Costa Rica, Cuba, Equator, Egipt, the United Arab Emirates, Ethiopia, Philippines, Gana, Jamaica, Jordan, Kazakhstan, Kuwait, Libya, Macedonia, Madagascar, Morocco, Moldova, Montenegro, Namibia, Oman, Qatar, Dominican Republic, Syria, Sri Lanka, Tajikistan, Tunisia, Turkmenistan, Uruguai, Uzbekistan, Vietnam and Zimbabue

This market requires elaborate products that can permanently guarantee to meet the requirements and needs imposed by the clients, considering the will of the stakeholders (including national sovereignty); the main action to ensure this guarantee is the certification. Then, the certification is seen as an instrument that enables the transfer of technology between the institution of Research and Development (R&D) and the industry, as well as an instrument to deal with barriers imposed by technology. From this point of view, this article aimed to describe which are the context and elements involved in the certification of the Vehicle Survey Booster – 30 (VSB-30), with emphasis on technological innovation. In order to complement the presented objective, the VSB-30 is described, as well as its importance 106

in the market and relevant aspects regarding certification; the Brazilian certification body is also introduced.

THEORETICAL BASE Theory is based on academic literature review about technological innovation and information concerning certification. Technological innovation Knowledge and technology become increasingly important as one of the most effective instruments to promote economic development in the world scenario. It is about a

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 105-115, Jan. - Mar, 2012

An Overview of the Certification of VSB-30 with Emphasis on Technological Innovation

view that surpasses the gates of the enterprise and the industry itself, thus leading new services, products and production processes to appear faster and faster in society (Jungmann, 2010). Taking knowledge and technology into consideration means to search for an efficient way to manage technological innovation. Technological innovation also causes great economic impacts that can pass unnoticed for the following years, or even for long periods. According to Freeman (1975), innovation may induce the creation of other innovations, so the date innovation was created should not be used as an indicator to analyze its impacts, but the period of its dissemination and diffusion in the market. Technological innovation can be defined as the implementation of a new product or process in terms of technology that brings significant improvement into the market, that is, it should be introduced in the market (innovation in the product) or used in the production (innovation in the process). As presented in the Oslo Manual, a product or process needs to be new (or substantially improved) for the company (not necessarily to the world) in order for there to be innovation (Organisation for Economic Co-operation and Development, 2005). For Tigre (2006), a company that implements technological innovation shows that it is possible to be different, thus enabling the accumulation of knowledge and practice for learning, besides the unique competitive advantage after being recognized among competitor companies. Innovation is usually a result of scientific research and the development of a product. Even if innovation is considered as a factor that can distinguish one company from another and enable a unique competitive advantage, Tigre (2006) says that the efficiency of an industry depends on the specialization pattern of a country and its demand for technology. In order to reach competitive ability levels, the mechanisms of the market are not enough; thus, public policies are necessary for specialization, technological infrastructure and measures that support innovation. By innovating and improving transferred technologies, the local industry can manufacture new products and develop new production methods, or superior means to employ the acquired technology, thus increasing the technological ability of the industry (Yeo, 1999).

Certification According to the Brazilian Association of Technical Standards (2006), the certification is a way to evaluate the compliance of an organization, regardless of the parts that are directly involved in the commercial relation. To complement this definition, according to the Department of Aerospace Science and Technology (2009), the certification corresponds to a “process by which an organization checks and confirms the application of requirements that were established for a product. It represents a very important activity in the technological and industrial development”. Likewise, based on the instructions of Aeronautics regarding the certification of products and the governmental insurance of quality (Department of Aerospace Science and Technology (2006), the certification is: “the process by which the Brazilian Aeronautical Command (COMAER) insures the application of requirements established for a product or a quality management system”. The mentioned concept is used for: • Type certificate: process to ensure the project of a product is in accordance with the requirements related to safety and to the accomplishment of the mission, thus officially recognizing this conformity; • Integration certificate: process to ensure that the integration of a platform project (aircraft, space vehicle and arms) are in accordance with the requirements related to safety and to the accomplishment of the mission, thus officially recognizing this conformity; •

Modification certificate: process to ensure that any changes in the certified product is in accordance with the established requirements;

•

Convalidation: process to ensure that an organization that supplies the product is capable of producing it in accordance with the project verified during the process of certification, unlike the one with the type certificate, thus meeting the requirements related to safety and the accomplishment of the mission established for the product;

• Supplier certificate: process to ensure that the quality management system from a supplier organization is in accordance with the established requirements;

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 105-115, Jan. - Mar., 2012

107

Carvalho, A.R.S. et al.

• Quality control: activity performed with the suppliers during the length of the contract in order to ensure the that the supplied material is in accordance with the established requirements; • Authorized operation return: consent of operation return after proving that the services were executed according to previously approved technical data; and • Product installation: is related to the infrastructure of the product. The action of certifying a product, service or system means to prove to the market and to the costumers that the organization has a controlled manufacturing system, which invested in training personnel, or even that it has an active management system, which makes sure that the specified activities are in accordance with the rules. According to the Brazilian Association of Technical Standards (2006), the main benefits brought by the certification are: to ensure the efficacy of the product, service or system; to ensure that the product, service or system will meet the standards; to introduce new products and brands into the market; to face the unfair competition; to reduce losses in the productive process and to improve management; to improve the image of the organization and its products or activities with the clients and decrease control and evaluations from the clients. The listed benefits, as well as the need to have products that are technologically innovative, show the certification should not be seen only as a bureaucratic instrument, but also as a tool to deal with the barriers imposed by technology, that is, the certification corroborates with technological innovation in terms of implementing a technologically new product (or with significant improvements) in the market, thus allowing the accumulation of knowledge and the practice for learning, in search of a competitive advantage.

METHODS Minayo (1993) informs that research is a scientific activity to solve problems and to discover reality, based on a process that will enable to discover new facts or data in any field of knowledge as a result of the combination of theory and collected data. The result of the research does not necessarily demonstrate an absolute true, since results and new discoveries are frequently renewed. 108

The research techniques vary according to the condition of the object of study; in this article, the research is characterized as a case study. Yin (2005) informs that the case study is chosen when questions such as “how” and “why” are used, and also because the control of the investigator over the events is much reduced, or when the temporal focus is in contemporary phenomena in the context of real life. This kind of investigation is particular, since its general feature is limited because the validity of its conclusions is still contingent. The research was carried out in the first semester of 2001, and consists of the literature review on technological innovation and certification, as well as information that is available in different reports, some of which are public, and interviews with experts involved in the certification of VSB-30. The analyzed reports were in the Department of Aerospace Science and Technology, the Institute of Aeronautics and Space and the Institute of Fomentation and Industrial Coordination, and some of them are public. An unstructured and qualitative interview was applied to the coordinator of the certification process. The participation of managers of the mentioned institutions guided and facilitated the research. The steps of the research aimed to understand the relevant aspects about the Brazilian certification body, about the VSB-30 and the used certification process, with the objective to describe which are the context and elements involved in the certification of the vehicle, emphasizing the technological innovation.

NATIONAL CERTIFICATION SYSTEM IN THE SPACE FIELD In the space field, the Brazilian Space Agency is responsible for the certification and normalization of space activities, and it is established under the terms of Article 3rd, Law n. 8,854, from February 10, 1994, subsections XIII and XIV, with the following attributions: to establish rules and expedite licenses and authorization concerning space activities, and to apply quality and production standards in space activities. In order to fulfill its goal in the certification and normalization fields, the Brazilian Space Agency created the National System of Certification in the Space Field (SINCESPAÇO), with the objective to promote the quality and safety of space activities in Brazil, as well as to develop the national space sector.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 105-115, Jan. - Mar, 2012

An Overview of the Certification of VSB-30 with Emphasis on Technological Innovation

The proposed certification mechanisms can include both the volunteer and the mandatory certification, depending on the standards, technical standards, and other related documents. In order to manage these normative acts the Brazilian Space Agency uses the Program to Support the Normalizing and Quality Activities in the Space Field (QUALIESPAÇO), with the objective to elaborate normative documents and promote them to ensure quality, safety and reliability of the products (goods and services) related to space activities. It is possible to observe that the actions of the certification agency are focused on: implementing and operating the access and propagation of technical standards; accelerating the actions of the Brazilian Association of Technical Standards in regards to the normalization of the space field; stimulating the Brazilian participation in the process of elaborating norms in the context of ISO; and establishing technical and administrative bases to implement national mechanisms of certification in the space field. The action of the agency in the normalization and certification segments is performed with the Brazilian Association of Aerospace Industries; the National Institute for Space Research and the Institute of Fomentation and Industrial Coordination. Certification Body A certification body is an organization authorized by the Brazilian Space Agency to certify within the National Development System of Space Activities (SINDAE). The Brazilian Space Agency nominated the Institute of Fomentation and Industrial Coordination as a certification body on January 7, 2011, resolution n. 3, which was issued by the Brazilian Space Agency (Diário Oficial da União, 2011, p. 13). The Institute of Fomentation and Industrial Coordination was considered as a Scientific and Technological institution from the Aeronautics Command (COMAER). It responds to the Institute of Aeronautics and Space and belongs to an innovative research complex of development, teaching and laboratories, as presented in Fig. 1. At COMAER, the Department of Aerospace Science and Technology is in charge of certifications, and delegates the work to the institute. The institute plans to generate knowledge and assistance especially to three sectorial objectives established by the Department of Aerospace Science and Technology (2010): to establish partnerships with the aerospace and defense

industries from the conception of studies, which encourage the aerospace scientific and technological complex; to rapidly meet the demands of the Brazilian Air Force for technical and scientific activities related to the aerospace fields; and to develop technical professional programs to improve cultural and intellectual levels. Fundraising for R&D projects at the Department of Aerospace Science and Technology comes from different actors and means; sometimes, researchers and technicians are the main sources: federal budget, bodies that lead to the development of science and technologies, partner companies and specialized technical services. The Institute of Fomentation and Industrial Coordination is internationally recognized and is part of the Brazilian System Evaluation of Conformity (SBAC), with the responsibility to certify in different fields. It is supported by the accreditation bodies such as the National Institute of Metrology, Standardization and Industrial Quality (INMETRO), the International Accreditation Forum (IAF) and the International Aerospace Quality Group (IAQG). Among the responsibilities of the institute, the ones related to the certification are: Certification of Quality Management Systems (OCS n. 0016), according to registration n. 21 – Aerospace DM 35.3 (Nace); Certification of Aerospace Quality Management System (OCE n. 0001); and Governmental Certification, based on Brazilian Regulations of Aerospace Quality (RBQA). It is possible to list some activities related to the space field certification: to give a type certificate in an aerospace product; to certificate an aerospace component; to give a modification certificate, in order to complement the type certificate, in an aerospace product; to validate the certification of an aerospace product; to follow-up and monitor difficulties in the service of a certified aerospace product of COMAER; to give technical assistance to the organization of the aerospace sector, in relation to the certification activities of an aerospace product; to give courses about certification of aerospace products and to give a declaration of technical ability to the manufacturing organizations in order to reproduce the certified aerospace product. The certification activity in the space field is being built, so there is not a definite and consensual standard among certifiers in different countries. Due to this lack of commonly accepted rules, it is important to understand the certification process of the VSB-30, but besides this understanding, it is worth to describe the vehicle and its importance to the national space industry.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 105-115, Jan. - Mar., 2012

109

Carvalho, A.R.S. et al.

Figure 1. Organizational positioning of the Institute of Fomentation and Industrial Coordination. Fonte: Ellaborated from the Aeronautics Command (2009).

VSB-30 It is a sounding vehicle, also known as research rocket. This device was designed to measure and to be used as a tool for scientific experiments during the suborbital flight. Generally, these vehicles can transport loads from 50 to 1,500km above the surface of the Earth. The VSB-30 (Fig. 2) is a Brazilian double-stage sounding vehicle, launched by rail. The first stage presents fast combustion of the propellant, and the second is widely used in other sounding rockets, such as Sonda III, VS-30 and VS-30 ORION. The main characteristics of the vehicle are (Brazilian Space Agency, 2005): total length of 12.6m; maximum diameter of 57m; two stages; total mass of 2,570kg; load mass of 400kg and apogee of 270km. For experiments in microgravity environment, the VSB-30 enables the load to remain above the altitude of 110 km for about six minutes, without atmospheric resistance, without propeller acceleration and in free fall. 110

In the European scientific program in high atmospheric layers, the operations to launch the VSB-30 happen in the Esrange launch field, in Sweden, which has a launcher with three rails, unlike the one used in Centro de Lançamento Alcântara, with only one rail. Due to the differences in the configuration of the launcher, there are two versions of the VSB-30, one to be launched in Alcântara (Brazil), and the other to be launched in Esrange (Sweden). The rocket received little international contribution for the project, and Brazil developed its own technology by adopting engineering solutions that are different from those adopted in Europe and the United States; some of ours are even preferable. The VSB-30 was launched by the Institute of Aeronautics and Space, in 2001, with the German Space Agency (DLR), to respond to the Microgravity European Program with around R$ 5 million in investments; 40% of this amount was granted by DLR. The rocket cost about R$ 750,000. The first launch was in 2004, and there were eleven launches until March 2011. All of them were successful,

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 105-115, Jan. - Mar, 2012

An Overview of the Certification of VSB-30 with Emphasis on Technological Innovation

According to the same story, around US$ 1 million is charged per launch, but the expectation is that there will be 1,500 annual flights if the price of loads per pound is reduced to US$ 250.

PROCESS OF CERTIFICATION OF THE Vvsb-30

Figure 2. Vehicle Survey Booster – 30 . Source: Brazilian Space Agency (2005).

and two were launched from Centro de Lançamento de Alcântara (MA-BR). In May, 2005, after meeting a serious of requirements related to documents and proofs of performance and safety, VSB-30 was approved by the European Space Agency (ESA) and was able to perform flights in Europe, carrying Texus and Maser scientific loads of the Microgravity European program (Kasemodel, 2010). The VBS-30 is considered as the most interesting alternative to replace the British Skylark, whose most recent launchings occurred in 2005, thus becoming the supplier in this launcher category. In 2009, COMAR and the Department of Aerospace Science and Technology announced the certification of VSB-30, leading to series production. This was the first Brazilian rocket to have this certificate. Nowadays, many companies work in its development and production: Villares, Cenic, Fibraforte, Mectron, Compsis, Avibrás, Orbital, among others. According to a story in the paper Valor Econômico, in June 27, 2009, about the certification of the VSB-30, it is estimated that the global market of suborbital sounding rockets, considering only civil applications, exceeds 100 annual launches for loads (scientific and technological experiments).

The process of certification of VSB-30 in Brazil was performed by the Institute of Fomentation and Industrial Coordination as the certification body, based on: Resolution n. 60, from May 17, 2004, of the Superior Council of the Brazilian Space Agency; and by an instruction of COMAER (ICA 80-2), approved by resolution n. 699/GC3, from July 6, 2006, of COMAER – instruction that guides the certification of aeronautic, space, infrastructure and airspace control products, and also the governmental insurance of these products. The certification was n. 001T2009, that is, “ the project of a product that is in accordance with the requirements related to safety and to the accomplishment of the mission, and officially recognizes this conformity”, as established by the Department of Aerospace Science and Technology (2006, p. 8). The process of certification counted on the evaluation of the European Space Agency (ESA), of the DLR and the Swedish Space Agency (SSC), besides the companies Kayser-Threde and EADS. The certification of VSB-30 is considered as an important step of its life cycle, so, after that, it is no longer a developing project, but operational, and its production should be totally transferred to the Brazilian aerospace industry. The certificate was officially given on October 16, 2009, in a ceremony held at the Department of Aerospace Science and Technology, with the participation of many military and civilian authorities. Some months before, the VSB-30 had been qualified by Sweden, and then integrated the products/services of that space agency. Such qualification aims to confirm that the vehicle meets the pre-established specifications, so it could be used as a small launcher in suborbital missions of space exploration (Silveira, 2009). The guidelines of the certification of the VSB-30 are the requirements related to safety and to the accomplishment of the mission, according to the steps presented in Fig. 3. Each phase – which always involves the participation of the Institute of Aeronautics and Space, the DLR and the Institute of Fomentation and Industrial Coordination – should consider some aspects, thus not being restricted to the following:

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 105-115, Jan. - Mar., 2012

111

Carvalho, A.R.S. et al.

•

Pre-requirement phase: certification plan; initial discussion on the base of certification, including aspects such as understanding the standards, applicable regulations, adjustments and equivalent levels of safety; getting to know the developing project; means to confirm the fulfillment of certification requirements (ground and flight rehearsals, among others, including the qualification by foreign organizations); information on the costs of type certification processes; list of critical items and confidentiality agreement, besides significant items. To sum up, these are the guidelines of certifications with procedures and deadlines to be considered during this process.

• Planning phase: norms and equivalent levels of safety; submission of the configuration management plan (CMP) to acceptance; definition of certification base; submission of the certification plan to be submitted to the Institute of Fomentation and Industrial Coordination and dossier of the construction of the device. It corresponds to the formal process of certification, with the definition of participants, resources and necessary activities to be delegates, besides the appropriation of adaptations in case of binational agreements to be connected with the foreign certifying organization. •

and requests to incorporate the necessary changes. • Approval phase: description of limitations defined by the requirement of applicable certifications and any other limitation and information that can identify the approved project. So, it is the deliberation of the certificate; then, the Institute of Fomentation and Industrial Coordination confirms that the requirements were fulfilled, and the process is filed. Even with the existence of a formal process in the Institute of Fomentation and Industrial Coordination, the coordinator of the certification process of the VSB-30 describes there is no structure of space certification that can be compared to that established by the International Civil Aviation Organization, nor a similar safety regulation established in the Civil Aviation that is accepted by the member countries. In fact, there are no internationally accepted predefined rules in the process of certification of the space device.

DISCUSSION

Execution phase: rehearsal proposals; simulation of analytical demonstration; requirement approval document; official rehearsals of certification with the participation of the Institute of Fomentation and Industrial Coordination; Engineering inspection to check for the fulfillment of non-confirmed requirements only by analyzing designs and reports; and approval of operational manuals presented

The fact that the Institute of Fomentation and Industrial Coordination is the certification body is in accordance with the National Policy of Space Activity Development (PNDAE), which aims to increase the participation of the government, the private sector, and especially of the Brazilian industries in space programs. The idea is to create opportunities to commercialize products and services related to space. The work of the Institute of Fomentation and Industrial Coordination as the certification body allows the presence of opportunities and threats or uncertainties that are consistent with the competitive environment, and the main ones are

by the solicitant, including reviews and supplements. It consists of the implementation of the certification plan, with inspections and rehearsals, examination of technical data, fulfillment of applicable requirements

demonstrated in Chart 2. Some aspects should be analyzed in Chart 2. The State is expected to promote and encourage scientific development, research and technological skills, aiming at the progress of

Figure 3. Macro phases of the certification process

112

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 105-115, Jan. - Mar, 2012

An Overview of the Certification of VSB-30 with Emphasis on Technological Innovation

sciences and the well-being of society, thus stimulating the internal market in order to enable the cultural and socioeconomic development, and also the technological autonomy of the country. Technology barriers were seen as a potential threat, due to the increase in commercial restrictions and also to the restrictions to access aerospace technologies. Also, besides the technological barriers, the volume and frequency of resource allocation for R&D activities oscillate, as a consequence of political and economic conjunctures. This leads to interruptions and non-programmed delays, which causes more difficulties to complete what was planned and to search for alternatives to execute the tasks. However, the PNDAE tries to consolidate and develop the Brazilian advances in space activities. In practical terms, it means to demand that the existing infrastructure be completed and updated, and also that the base of personnel dedicated to space activities be increased and improved. They also ask for more governmental and private participation in space programs, especially concerning the Brazilian industries, so that new commercial opportunities can be created for the products and services related to the space. To complement the PNDAE, the Ministry of Defense aims to be recognized as a reference organization as to the conduction of subjects related to Science, technology and innovation of interests of national defense by 2015. The domination of Science, Technology and Innovation

management should be pursued both by the Ministry of Defense and the Brazilian Space Agency, once it is seen as an instrument to promote economic development in the world scenario, besides attracting new innovations. The certification of VSB-30 can be seen as a technological innovation, once it brings significant improvements to the process of acceptance and technological transfer to the market. The mentioned process of certification complements the mechanisms in the market in relation to technological ability, in which the public institution, in this case, the Institute of Fomentation and Industrial Coordination as the certification body, provides mechanisms to leverage the technological infrastructure and to create support measures towards innovation. The certification helps to show the stakeholders the efficacy of the VSB-30, making sure there is continuous concern to improve and refine the activities, as well as the constant auditing processes that lead the involved personnel to be involved; consequently, the performance generally improved, thus removing the uncertainties and broadening market opportunities. It is worth to remember that in regards to technological innovation, the certification of VSB-30 allows the accumulation of knowledge and the practice for learning, once it is based on structured stages of pre-requirement, planning, execution and approval, which implies that different institutes (Institute of Aeronautics and Space), the German Space Agency (DLR)

Chart 2. Scale of countries in the modern space conquest. Factors and variables Political

Opportunities

Threats or uncertainties

Interest in the development of the aerospace sector.

Competition with resources for other fields (social and education).

Economic

Social

The market demands the certification;High earned value

Low investments in aerospace products Dependence on

of the products.

governmental actions.

The society demands safe products.

The society does not understand the need to invest in aerospace products.

Technological

Legal

Tendency to increase the national ability to develop

Commercial barriers for sensitive Technologies

technologies to the aerospace sector.

(technological barriers); Lack of qualified personnel.

The Law is addressed to the certification of products.

The recognition of certifications by the Institute of Fomentation and Industrial Coordination by foreign organizations.

J. Aerosp. Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.4, No 1, pp. 105-115, Jan. - Mar., 2012

113

Carvalho, A.R.S. et al.

and the Institute of Fomentation and Industrial Coordination can share information and technology. On one hand, the certification enables the companies to develop technologies and to increase technological skills in the industry; on the other hand, it is a tool of industrial isolation, in which incapable companies may be expurgated from the innovative process. It is expected that the aerospace certification can increase technological skills based on the principles presented by the “Strategic Concept: Science, Technology and Innovation of Interests in the National Defense”, established by the Ministry of Defense (MD). It is important to emphasize that the strategic conception has the mission to enable technological and scientific solutions as well as innovations to attend the needs of the country in relation to national defense and development. The national aerospace certification is a response to the technological barriers imposed by the countries that have the technology, once it makes the adopted requirements and parameters clear, transferring credibility to the costumers and stakeholders, even if there is not a certification structure that can be compared to that established by the International Civil Aviation Organization. The response to the technological barrier will be valid for the national industry, as long as it leads to synchronicity and symmetry between the skill level of R&D institutions and national companies. In this sense, the certification becomes part of the public policies that develop formation, technological infrastructure and support measures that lead to innovation. From an operational point of view, it is possible to observe that the improvement of processes was a great benefit, especially concerning documents and the reduction of failure in projects. It is a way for the stakeholders to have access to technical standards. Another operational aspect is directly related with SINCESPAÇO, in terms of establishing technical and administrative techniques to implement national mechanisms of certification in the space field, as well as to present certification stages for ABNT.

presented objective, the development of the vehicle was described, as well as its importance in the market and relevant aspects about the certification. The Brazilian certification body was also presented. We understood that Brazil is on level four, “the existence of a National Space Agency with its own satellites”, in the scale of countries towards space conquest, together with Argentina, Australia, Canada, South Korea, Iran, Nigeria, Pakistan, Taiwan and Ukraine. The next level (five) is the independent ability to launch satellites in the Earth’s orbit. The certification of VSB-30 is a way to reach that level, where India and Israel are located. The process of certification is a contribution to technological innovation since it provides benefits to improve processes, especially concerning documentation and, at the same time, creating conditions so that the industry can adapt to consolidated formal processes, and also creating barriers to the companies that are unable to adjust to the established requirements. The homologation certificate searches for international recognition of the country in the space field. It helps Brazil’s autonomy in many critic space technologies, thus creating an opportunity to strengthen the national aerospace industry, contributing with PNDAE. It is also a way to overcome the technological barrier.

FINAL CONSIDERATIONS

Câmara dos Deputados, 2009, “A política espacial brasileira: parte 1”. Relator: Rollemberg R., Veloso E.M., Filho Q.P.A, et al., Brasília, Câmara dos Deputados, Edições Câmara. Vol. 2.

The article aimed at describing which are the context and elements involved in the certification of VSB-30, with emphasis on technological innovation. To complement the 114

REFERENCES Agência Espacial Brasileira (AEB), 2005, “Veículos Lançadores”, retrieved in 15 jul 2010, http://www.aeb.gov.br/ indexx.php?secao=lancadores. Associação Brasileira de Normas Técnicas (ABNT), 2006, “ABNT Certificadora”, retrieved in 3 Jun 2011, http://www. abnt.org.br/m3.asp?cod_pagina=1001. Associação das Indústrias Aeroespaciais do Brasil (AIAB), 2011, “Números da Associação das Indústrias Aeroespaciais do Brasil. São José dos Campos – SP”, retrieved in 25 Mar 2011, http://www.aiab.org.br.

Comando da Aeronáutica (COMAER), 2009, “Regimento

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 105-115, Jan. - Mar, 2012

An Overview of the Certification of VSB-30 with Emphasis on Technological Innovation

Interno do Comando da Aeronáutica (RICA 20-36)”, Gabinete do Comandante da Aeronáutica, Brasília. Departamento de Ciência e Tecnologia Aeroespacial (DCTA), 2006, “Certificação de produto e garantia Governamental da qualidade”, São José dos Campos. Departamento de Ciência e Tecnologia Aeroespacial (DCTA), 2009, “Primeiro foguete recebe certificação no Brasil”, retrieved in 10 mai 2011, http://www.cta.br/noticias/ estrutura.php?id=55. Departamento de Ciência e Tecnologia Aeroespacial (DCTA), 2010, “Programa de Trabalho do Instituto de Fomento e Coordenação Industrial” (ICA 19-113), São José dos Campos. Diário Oficial da União (DOU), 2011, “Seção 1. Portaria nº 3/AEB, de 07 de janeiro de 2011”, Brasília, Imprensa Nacional, p. 13. Freeman, C., 1975, “La Teoría económica de la innovación industrial”, Madri, Alianza Editorial.

n. 3, retrieved in 11 Abr 2011, http://www.aeroespacial.org. br/downloads/revista/AABRevista_N03_2010-Jan-Mar.pdf. Landini, M. Z. S., Cabral, A. S., 2005, “Transferência e cooperação tecnológica no Programa Espacial Brasileiro”, In: XI Seminario de Gestión Tecnológica Altec, Salvador, Altec. Minayo, M. C. S., 1993, “O desafio do conhecimento”, São Paulo, Hucitec. Organização para Cooperação Econômica e Desenvolvimento (OCDE), 2005, “Manual de Oslo: proposta de Diretrizes para coleta e interpretação de dados sobre Inovação Tecnológica”, FINEP. Silveira, V., 2009, “Foguete Brasileiro recebe certificação”, In: Jornal Valor Econômico. Tigre, P. B., 2006, “Gestão da inovação: a economia de tecnologia no Brasil”, Rio de Janeiro, Elsevier.

Jungmann, D. M., 2010, “A caminho da inovação: proteção e negócios com bens de propriedade intelectual: guia para o empresário”, Brasília, IEL.

Yeo, E., 1999, “Technological Capabilities of Our Defence Industries”, Journal Of The Singapore Armed Forcues, Cingapura, Vol. 25, N. 2, retrieved in 24 Set 2010, http://www. mindef.gov.sg/safti/pointer/back/journals/1999/Vol25_2/9.htm.

Kasemodel, C.A., 2010, “VSB-30: o primeiro Foguete Brasileiro Certificado”, In: Associação Aeroespacial Brasileira – Revista.

Yin, R. K., 2005, “Estudo de caso: planejamento e métodos”, Tradution Daniel Grassi, Porto Alegre, Bookman.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 105-115, Jan. - Mar., 2012

115

Thesis abstracts

This section presents the abstract of most recent Master or PhD thesis related to aerospace technology and management

Development of a Novel Rans-Based Method for the Computational Aeroacoustics of High-Speed Jets Carlos Roberto Ilário da Silva Escola Politécnica da Universidade de São Paulo São Paulo/SP – Brazil carlos.ilario@yahoo.com Thesis submitted for PhD degree in Mechanical Engineering at Universidade de São Paulo, São Paulo, São Paulo State, Brazil, 2011. Advisor: Doctor Julio Romano Meneghini Abstract: A novel computational aeroacoustics tool based on Reynolds Averaged Navier-Stokes method (RANS) was developed for predicting the noise generated by complex three-dimensional jet flows. The new method is called LRT, which arises from the combination of Lighthill’s acoustic analogy with Ray-Tracing acoustics. The powerful advantage of applying the LRT method for noise predictions is that it calculates not only the noise sources, but it also models and takes into account sound-flow interaction effects without any geometric simplification, such as flow symmetries of the problem. This is now a strong requirement from aero-engines manufactures since investigations on asymmetric nozzles, as a means of noise reductions are in progress. The LRT method is a relatively fast jet noise prediction-tool based on Lighthill’s Acoustic Analogy and it uses a Reynolds-Average Navier-Stokes (RANS) computational fluid dynamics (CFD) simulation as input information. The sound-flow interaction is computed by solving the propagation using Ray-Tracing equations. The LRT method has been formulated as a general three-dimensional method and it has no restrictions on the type of the flow field or nozzle geometry for noise prediction. Successful numerical noise predictions have been carried out for a variety of jet flows (single, coaxial, and asymmetric jets), using the LRT as an engineering tool. The outcome from this thesis is a numerical tool that allows noise predictions of complex exhaust systems and the variations in sound field due to modifications of the flow field generated by the interaction

of the jet flow with high-lift surfaces. In addition, the LRT method can be applied to complement experimental analysis providing a better understanding about the flow and acoustics mechanisms for complex jets. Keywords: Aeroacoustics, Noise prediction, Jets, Aeroacoustics analogy, Sound-flow interaction.

Study on the Accelerated Aging of the Composite Solid Propellant Júlio de Barros Magalhães Instituto Tecnológico de Aeronáutica São José dos Campos/SP – Brazil juliodebarros@gmail.com Thesis submitted for Masters in Aeronautical and Mechanical Engineering at Technological Institute of Aeronautics (ITA), São José dos Campos, São Paulo State, Brazil, 2011. Advisors: Doctors Vera Lucia Lourenço and Luciene Dias Villar

Abstract: The solid propellant is a complex and stable mixture of oxidizing ingredients and reducing agents, which, when ignited, react with each other, forming gas molecules at high temperatures. The storage of rocket motors loaded with composite solid propellant for long periods may change the propellant properties, thus causing failure and affecting the safety during launch. In this study, it was carried out an accelerated aging assay was carried out in order to predict the useful lifetime and to evaluate variations on the properties of the propellant with time by means of tensile tests, hardness measurements, solvent extraction, thermal analysis TG/DSC, and testing in bomb calorimeter. The aging temperatures used were 45°, 55° and 65°C, and samples were withdraw at 1, 3, 6, 9, 10, and 12 months. Aging was also carried out at room temperature (21°C). There was no significant variation in the values of activation energy of thermal decomposition in the two methods used – Ozawa and model-free isoconversional methods – during the aging period. The variation in results obtained by the bomb calorimeter was small. The solvent extraction showed that the use of bonding agent significantly

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 117-119, Jan. - Mar., 2012

117

changed the sol fraction during aging. The mechanical properties showed that over the course of accelerated aging, the propellants presented loss of stiffness, which was verified by decreasing values of maximum stress and Young’s modulus. There was also a decrease in hardness Shore A. The variation of mechanical properties allowed the prediction of useful lifetime of the composite solid propellant. Keywords: Solid propellant, Accelerated aging, Mechanical properties, Thermal decomposition, Prediction of useful lifetime.

Evaluation of Static Strength and Fatigue Life of Carbon Fiber/Epoxy and Glass Fiber/Epoxy Composites Under Shear Stress by Using the Iosipescu Method Vanderlei de Oliveira Gonçalves Technological Institute of Aeronautics São José dos Campos/SP – Brazil vanderleivog@iae.cta.br Thesis submitted for Masters of Science in Mechanical Engineering at Technological Institute of Aeronautics (ITA), São José dos Campos, São Paulo State, Brazil, 2010.

strength, fatigue tests were performed. The S-N curves were obtained using two stress ratio, R=0.1 and R=0.5, at frequency of 12 Hz, and a cycle limit (N) of 120,000 ones. The values of (τ12) and (G12) obtained on static and fatigue tests were compared. The neat epoxy resin presented the lowest shear modulus, the lowest static shear strength and lowest fatigue life compared to the composites studied. The composites reinforced having 0/90º fiber orientation in relation to the loading axis, exhibited lower shear strength, shear modulus and fatigue life strength, compared to the composites having ±45º fiber orientation in relation to the loading axis. On fatigue tests, the fatigue life increased as the stress ratio increases from R=0.1 to R=0.5 due to the viscoelastic behavior of the epoxy resin for all materials studied. Keywords: Composites, In plane Iosipescu Shear, Fatigue.

Reduction of Autonomy-Level Requirement in LEO Sun-Synchronous Satellites Via Optimization of Longitude Distance Between Two High-Latitude Ground Segments Hossein Bonyan Khamseh Shahid Beheshti University Tehran, Iran h.bonyan@gmail.com

Advisor: Doctor Luiz Claudio Pardini Abstract: Composite materials have increasingly been used on aerospace industry due to its low density and high mechanical strength as well as high fatigue endurance. Aerospace and aeronautical structures are submitted to several types of stresses like tension, compression, flexure and shear, during service life. These stresses can also occur simultaneously. Among these stresses, shear is the most complex to be evaluated in composites due to its heterogeneity, anisotropy, and premature structural damage can be achieved at low shear stresses. Compared to other shear test methods for composites, the Iosipescu shear test has increasingly been used by researchers, due to its simplicity to determine the in-plane or interlaminar shear strength (τ) and shear modulus (G). The purpose of the present work was to determine the static in-plane shear, in-plane modulus and to measure the in-plane shear fatigue life using the Iosipescu method, for a neat epoxy resin and for composites made of epoxy matrix reinforced with glass and carbon fiber fabrics. From the results of static in plane shear 118

Thesis submitted for the degree of Master of Science at Shahid Beheshti University, Tehran, Iran, 2010. Advisors: Doctors Navabi and Ghannadpour Abstract: The required level of in-orbit autonomy for LEO sun-synchronous satellites was studied. Requirements to accommodate a given level of autonomy onboard a typical satellite were investigated. With respect to the pattern of ground segment access to the satellite, longest accessibility gap was proposed as the appropriate metric to study the autonomy requirement onboard the satellites. The concept of repeatability cycle was employed to derive the proposed metric in a time-independent manner. A method was proposed to reduce the requirement of onboard autonomy. In the proposed method, it was shown that longitude of the ground segment does not affect the pattern of access to the satellite, in a repeatability cycle. Consequently, optimization of ground segment location was done in the latitude-direction only. In

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 117-119, Jan. - Mar., 2012

accordance with these results, very-high-latitude regions, i.e. Polar Regions, were ideal location for ground segment, to reduce the autonomy requirement. To avoid the practical operational difficulties of ground segments at arctic regions, mission architectures with two high-latitude (but not polar) ground segments were proposed. Optimization of the relative longitude-distance between the two ground segments was performed to reduce the requirement of satellite onboard autonomy. It was shown that by taking advantage of such mission architectures, considerable reduction in the onboard autonomy requirement may be achieved, while avoiding increased cost of ground segments at Polar Regions. It is being recalled that reduction in the requirement of onboard autonomy is translated into reduction of monetary and schedule requirements of the satellite program.

while implication in longitudinal, lateral and complete model of aircraft. In addition to optimal control problem constraints, comfort index (CI) for passenger convenience has been modeled mathematically and it has been applied in the problem in order to find optimal trajectory and control history. Results of the proposed optimal periodic maneuver, in terms of required flight controls and their associated costs, are presented with and without passenger comfort consideration.

Keywords: Autonomy Requirement, Low Earth Orbit, Sun-synchronous Satellite, Ground Segment Location.

Élcio Jeronimo de Oliveira National Institute for Space Research São José dos Campos/SP – Brazil elcioejo@iae.cta.br

Adaptive Performance Optimization for a Transport Aircraft Seyed Hossein Mortazavi Sharif University of Technology Tehran, Iran h.mortazavi@sri.ac.ir Thesis submitted for degree of Master in Engineering at Sharif University of Technology, Tehran, Iran, 2006. Advisor: Doctor Seid Hossein Pourtakdoust Abstract: Performance optimization in cruise phase is one of the most important challenges in fuel reduction for airlines. The significance of this issue will be so critical during long hours of flight that make a serious effect on profit of airlines. Many works in air-vehicle performance optimization have already been done, however newer methods are always needed for achieving better optimization rate. In this Thesis, with inspiration of motion of fishes in water, a periodic optimization method for a transport aircraft has been proposed. This method delivers an optimal maneuver to improve the cruise fuel consumption of a transport category aircraft. The maneuver is taken by analogy to the natural phenomena of fish movement in the water, since we know that natural locomotion of some animals is energy-efficient motion. This method leads in considerable improvement in aircraft performance

Keywords: Periodic maneuver, Performance optimization, Comfort index, Fish locomotion, Optimal control.

Fault Detection and Isolation on Inertial Measurement Units with Minimal Redundancy of Fiber Optic Gyros

Thesis submitted for PhD degree in Space Engineering and Technologies at National Institute for Space Research (INPE), São José dos Campos, São Paulo State, Brazil, 2011. Advisors: Doctors Ijar Milagre da Fonseca and Hélio Koiti Kuga

Abstract: In this thesis, it is shown the study related to fault detection and isolation process in the inertial measurement units with minimal redundancy of fiber optic gyros. In the course of this work, as accessory elements, the models for unit construction and calibration were addressed. In the main focus of this thesis, topics of signal processing were approached in order to design a fault detection and isolation algorithm applicable to an inertial measurement unit. The main theoretical bases for this work were abrupt change detection algorithms, median filters, wavelet transform, singular value decomposition (SVD), and principal component analysis (PCA). As a result, it was possible to detect small abrupt variations in the inertial measurement unit sensors (gyros) with high confidence, meeting the requirements of simplicity, velocity, and independence of dynamics models in the fault detection and isolation process. Keywords: Fault detection and isolation, Fiber optic gyros, Wavelet packet, Inertial measurement units tetrad, Principal component analysis.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 117-119, Jan. - Mar., 2012

119

INSTRUCTIONS TO AUTHORS (Revised in March, 2012)

SCOPE AND EDITORIAL POLICY The Journal of Aerospace Technology and Management (JATM) is the official publication of the Departamento de Ciência e Tecnologia Aeroespacial (DCTA), in São José dos Campos, São Paulo State, Brazil. The journal is quarterly published (March, June, September, and December) and is devoted to research and management on different aspects of aerospace technologies. The authors are solely responsible for the contents of their contribution. It is assumed that they have the necessary authority for publication. When submitting the contribution, authors should classify it according to the area selected from the following topics: • • • • • • • • • • • • • •

Acoustics Aerospace Circuitry Aerospace Systems Aerospace Meteorology Aerodynamics Air and Space Defense Systems Applied Computation Astrodynamics Ceramic Materials Composites Eletromagnetic Compatibility Energetics Fluid Dynamics and Turbulence Guidance Navigation and Control

• • • • • • • • • • • •

Management Systems Metallic Materials Photonics Polymeric Materials Processing of Aerospace Materials Propulsion and Combustion Radars and Tracking Systems Robotics and Automation Structures Synthesis and Characterization of Aerospace Materials Thermal Sciences Vibration and Structural Dynamics

Papers in specialties that are not mentioned above but clearly related to JATM´s scope are also of interest. PAPER SUBMISSION The manuscript should be digitalized using a Microsoft Word (.doc) software program and submitted electronically in English. See the instructions at www.jatm.com.br/, paper submission. If there is any conflict of interest regarding the evaluation of the manuscript, the author must send a declaration indicating the reasons so that the review process occurs fairly. After submitting the manuscript, the corresponding author will receive an e-mail with the Term of Copyright Transfer, in which the author agrees to transfer copyrights to the DCTA in case of acceptance for publication, thus being forbidden any means of reproduction (printed or electronic) without previous authorization of the Editor in Chief. If the reproduction is allowed, it is mandatory to mention the JATM. The author also declares that the manuscript is an original paper, its content is not being considered for publication in other periodicals and that all co-authors participated satisfactorily in the paper elaboration as to make public the responsibility for its content. The declaration must be printed, signed by the main author, and sent back by mailing to the following address: Instituto de Aeronáutica e Espaço/ATTN: Helena Prado/ Praça Marechal Eduardo Gomes, 50 – Vila das Acácias – CEP: 12228-901 – São José dos Campos/SP, Brazil. Papers already presented at conferences will be accepted if they were not published in complete form in the Annals of the conference or if they are extended with additional results or new findings. These articles will be evaluated as the others. Articles from guest authors will be published after approval of one specialist associate editor. The JATM does not publish translated articles from other journals. 120

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 120-123, Jan. - Mar., 2012

PEER REVIEW Manuscripts will be reviewed by at least two expert consultants, members of the Editorial Committee or external evaluators (ad hoc referees) in double-blind peer review mode, ensuring complete anonymity. In case of disagreement on the results of the evaluation, the manuscript will be forwarded to a third reviewer, and it will be accepted for publication only if two approvals are received. The evaluators can accept the manuscript in the form it was submitted, they can reject it or request revisions. The manuscript that requires revision will be sent to the author that is supposed to submit a new version and, in the case the author does not agree with the suggestions, it is necessary to send a “letter to editor”, explaining the reasons. The Editor in Chief will approve after verifying in the new version the adherence to the reviewers’ suggestions or will send to another evaluation round if the changes have not been sufficiently addressed. Accepted manuscripts can be edited to comply with the format of the journal, remove redundancies, and improve clarity and understanding without altering meaning. Authors are also strongly advised to use abbreviations sparingly whenever possible to avoid jargon and improve the readability of the manuscript. All abbreviations must be defined the first time that they are used. The edited text will be presented to authors for approval. MANUSCRIPT CATEGORIES Editorial: Any researcher may write the editorial on the invitation of the Editor in Chief. Editorials should cover broad aspects of Aerospace Technology. Such manuscripts are not submitted to peer review. Review articles: These should cover subjects that are relevant to the scope of the journal. Authors should bear in mind that they are expected to have expertise in the reviewed field. The article may be of any length required for the concise presentation of the subject. Original papers: These articles should report results of the scientific research. The article may be of any length required for the concise presentation and discussion of the data, but succinct papers are favored in terms of impact as well as in readability. Communications: They should report previous results of ongoing research and should not exceed eight pages. Thesis abstracts: The journal welcomes recent Masters and PhD thesis abstracts for publication. Such contribution will not be submitted to peer review. MANUSCRIPT STRUCTURE Whenever it is possible, articles should include the following subsections, however articles from some areas should follow their usual format. Title and names of authors: The title should not contain abbreviations. All authors should be identified with full name, e-mail, institution to which they are related, city, state, and country. One of them should be indicated as the author for correspondence and his/her full address is required. Abstract: They are limited to 250 words and structured into objectives, methods, results, and conclusions. Citations or abbreviations (except internationally recognized abbreviations, such as weights, measures, and physical or chemical ones) are not permitted. Keywords: Three to six items that should be based on NASA Thesaurus volume 2 – Access Vocabulary.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 120-123, Jan. - Mar., 2012

121

Introduction: It should set the purpose of the study, providing a brief summary (not a review) of previous relevant studies, and stating the new advances in the current investigation. The introduction should not include data or conclusions from the work being reported. A final sentence summarizing the novel finding to be presented is permissible. Methodology: The authors are free to use any structure in this section to fit the objectives of the work, they could also rename it (e.g. Numerical analysis, Case study, and so on), and in some cases it may be advisable to omit it. Clear and sufficient information to permit the study to be repeated by others should be briefly given. Standard techniques need only to be referenced. Previously published methods may be briefly described following the reference. Results: This section should be a concise account of the new information that was discovered, with the least personal judgment. Do not repeat in text all the data in the tables and illustrations, but briefly describe what these data comprise. Discussion: The discussion should include the significance of the new information and relevance of the new findings in light of existing knowledge. Only unavoidable citations should be included. Citations to review articles are not encouraged in this section. In some cases, it may be advisable to merge with the previous section (“Results and Discussion”). Acknowledgements: This section should be short, concise, and restricted to acknowledgements that are necessary. The financial support received for the elaboration of the manuscript must be declared in this item. References: Acceptable references include journal articles, numbered papers, books, and submitted articles, if the journal is identified. References must be restricted to directly relevant published works, papers, or abstracts that have been accepted for publication. References from private communications, dissertations, thesis, published conference proceedings, and preprints from conferences should be avoided. Self-citation should be limited to a minimum. Authors are responsible for the accuracy and completeness of their references. References in text: The references should be mentioned in the text by giving the last name of the author(s) and the year of publication. Either use “Recent work (Smith and Farias, 1997)” or “Recently Smith and Farias (1997)”. With three or more names, use the form “Smith et al. (1997)”. If two or more references have the same identification, distinguish them by appending “a”, “b”, etc., to the year of publication. Reference list: References should be listed in alphabetical order, according to the last name of the first author, at the end of the article. Only citations that appear in the text should be referenced. Unpublished papers, unless accepted for publication, should not be cited. Work that is accepted for publication should be referred to as “In press”. It is recommended that each reference contains the digital object identifier number (DOI). References retrieved from the Internet should be cited by the last name of the author(s) and the year of publication, or n.d., if not available, followed by the date of access. Standards should be cited in text by the acronym of entity followed by the number, and they do not need to appear in the reference list. Some examples of references are as the following ones: Alves, M. B., Morais, A. M. F., 2009, “The management of knowledge and technologies in a Space Program”, Journal of Aerospace Technology and Management, Vol. 1, No 2, pp. 265-272. doi:10.5028/jatm.2009.0102265272 Bordalo, S.N. et al., 1989, “The Development of Zonal Models for Turbulence”, Proceedings of the 10th Brazilian Congress of Mechanical Engineering, Vol. 1, Rio de Janeiro, Brazil, pp. 41-44. Coimbra, A. L., 1978, “Lessons of Continuum Mechanics”, Ed. Edgard Blücher, São Paulo, Brazil, 428p. Clark, J. A., 1986, “Private Communication”, University of Michigan, Ann Harbor.

122

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 120-123, Jan. - Mar., 2012

EMBRAPA, 1999, “Politics of R&D”, Retrieved in May 8, 2010, from http://www.embrapa.br/publicacoes/institucionais/polPD.pdf. Silva, L. H. M., 1988, “New Integral Formulation for Problems in Mechanics” (In Portuguese), Ph.D. Thesis, Federal University of Santa Catarina, Florianópolis, S.C., Brazil, 223p. Sparrow, E. M., 1980a, “Forced Convection Heat Transfer in a Duct Having Spanwise-Periodic Rectangular Protuberances”, Numerical Heat Transfer, Vol. 3, pp. 149-167. Sparrow, E. M., 1980b, “Fluid-to-Fluid Conjugate Heat Transfer for a Vertical Pipe-Internal and External Natural Convection”, ASME Journal of Heat Transfer, Vol.102, pp. 402-407. Tables: Tables should be constructed using the table feature in the word processor or using a spreadsheet program, such as Microsoft Excel. They should be numbered in order of appearance in the text, using Arabic numerals. Each table should have a title and an explanatory legend, if necessary. All tables must be referenced and mentioned in the text as “Table” and succinctly described in the text. Under no circumstances should a table repeat data that are presented in an illustration. Statistical measures of variation (i.e., standard deviation or standard error) should be identified, and decimal places in tabular data should be restricted to those with mathematical and statistical significance. Authors should take notice of the limitations set by the size and layout of the journal. Therefore, large tables should be avoided. Figures: All illustrations, line graphs, charts, schemes, photographs, and graphs should be referred as “Figure” and submitted with good definition. Number figures consecutively using Arabic numerals in order of appearance. References should be made in the text to each figure using the abbreviated form “Fig.”, except if they are mentioned in the beginning of the sentences. Captions should be descriptive and should allow the examination of the figures, without reference to text. The size of the figures (including frame) should be 8 cm (one column) or 17 cm (two columns) wide, with maximal height smaller than 22 cm. Equations: Type them on individual lines, identifying them by Arabic numerals enclosed in parenthesis. References should be made in the text to each equation using the abbreviated form “Eq.”, except in the beginning of the sentences, where the form “Equation” should be used.

J. Aerosp. Technol. Manag., São José dos Campos, Vol.4, No 1, pp. 120-123, Jan. - Mar., 2012

123

Correspondence All correspondence should be sent to: JOURNAL OF AEROSPACE TECHNOLOGY AND MANAGEMENT Instituto de Aeronáutica e Espaço Praça Mal. Eduardo Gomes, 50- Vila das Acácias CEP 12228-901 São José dos Campos/ São Paulo/Brazil

Contact Phone: (55)12-3947- 6493/5122 E-mail: editor@jatm.com.br Web: http://www.jatm.com.br

Published by: Departamento de Ciência e Tecnologia Aeroespacial Distributed by: Instituto de Aeronáutica e Espaço Proofreading, version and standardization: Zeppelini Editorial Editing: TL Publicidade e Assessoria Ltda Printing: Ingrafoto Produções Gráficas Cover: Created by Gus Morais - based on the original by Paulo R. Porphirio Moreira Edition: 750 São José dos Campos, SP, Brazil ISSN 1984-9648

Historical Note: JATM was created in 2009 after the iniciative of the diretor of Instituto de Aeronáutica e Espaço (IAE), Brigadeiro Engenheiro Francisco Carlos Melo Pantoja. In order to reach the goal of becoming a journal that could represent knowledge in Science and aerospace technology, JATM searched for partnerships with others institutions in the same field from the beginning. From September 2011, it has been edited by the Departamento de Ciência e TecnologiaAeroespacial (DCTA), and it also started to be financially supported by Fundação Conrado Wessel.

The copyright on all published material belongs to Departamento de Ciência e Tecnologia Aeroespacial (DCTA)