Vol.3 N.3 - Journal of Aerospace Technology and Management by Journal of Aerospace Technology and Management

Vol. 3 N. 3 Sep./Dec. 2011

ISSN 1984-9648 ISSN 2175-9146 (online) www.jatm.com.br

Journal of Aerospace Technology and Management V.3, n. 3, Sep./Dec., 2011

Journal of Aerospace Technology and Management www.jatm.com.br

JATMv3n3capa.indd 1

05.12.11 16:44:31

Correspondence

General Information

All correspondence should be sent to: Journal of Aerospace Technology and Management (JATM) is a techno-scientific publication serialized,

JOURNAL OF AEROSPACE TECHNOLOGY AND MANAGEMENT

edited, and published by the Department of Aerospace Science and Technology – (Departamento de Ciência

Instituto de Aeronáutica e Espaço (IAE)

e Tecnologia Aeroespacial - DCTA). It contains articles that have been selected by an Editorial Committee

Praça Mal. Eduardo Gomes, 50- Vila das Acácias

composed of researchers and technologists from the scientific community. The journal is published every

CEP 12228-901

four months, and its main objective is to provide an archival form of presenting scientific and technological

São José dos Campos/ São Paulo/Brazil

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

Contact

research and generate a greater global exchange of knowledge.

Phone: (55)12-3947-5122/5008 E-mail: editor@jatm.com.br

JATM is added/indexed in the following databases; SCOPUS - Elsevier; CAS - Chemical Abstracts Service;

Order your copy (for free): secretary@jatm.com.br

DOAJ - Directory of Open Access Journals; J-GATE - The e-journal gateway from global literature; LIVRE -

Web: http://www.jatm.com.br

Portal to Free Access Journals; GOOGLE SCHOLAR; SUMÁRIOS.ORG - Summaries of Brazilian Journals; EZB- Electronic Journals Library; ULRICHSWEB - Ulrich´s Periodicals Directory and is under analysis in other major indexing databases. JATM is affiliated to ABEC - Brazilian Association of Scientific Editors and all published articles contain

Published by: Departamento de Ciência e Tecnologia Aeroespacial Distributed by: Institute of Aeronautics and Space

DOI numbers attributed by CROSSREF.

Desktop publishing and printing: Zeppelini Editorial Edition: 750

Financial support:

São José dos Campos, SP, Brazil ISSN 1984-9648

Historical Note: JATM was created in 2009 after the initiative of the director of the Aeronautics and Space Journal of Aerospace Technology and Management

Institute, Brigadeiro Engenheiro Francisco Carlos Melo Pantoja. In order to reach the goal of becoming a

Vol. 3, n.3 (sep./dec. 2011) – São José dos Campos: Zeppelini Editorial, 2011

journal that could represent knowledge in science and aerospace technology, JATM searched for partnerships

Four monthly issue

with other institutions in the same field from the beginning. From September 2011, it has been edited by the Department of Science and Aerospace Technology, and it also started to be financially supported by Fundação Conrado Wessel.

1. Aerospace sciences 2. Technologies 3. Aerospace engineering

The copyright on all published material belongs to CDU:629.73

JATMv3n3capa.indd 2

Departamento de Ciência e Tecnologia Aeroespacial (DCTA).

05.12.11 16:44:57

Journal of Aerospace Technology and Management J. Aerosp. Technol. Manag. Vol. 3, No. 3, Sep. - Dec. 2011

Editor in Chief

Executive Editor

Francisco Cristovão Lourenço de Melo Institute of Aeronautics and Space São José dos Campos - Brazil editor@jatm.com.br

Ana Marlene Freitas de Morais Institute of Aeronautics and Space São José dos Campos - Brazil secretary@jatm.com.br

ASSOCIATE EDITORS Adriana Medeiros Gama - Institute of Aeronautics and Space - São José dos Campos - Brazil Ana Cristina Avelar - Institute of Aeronautics and Space - São José dos Campos - Brazil André Fenili - Universidade Federal do ABC- São Paulo - Brazil Angelo Pássaro - Institute for Advanced Studies - São José dos Campos - Brazil Antonio Fernando Bertachini - National Institute for Space Research - São José dos Campos - Brazil Antonio Pascoal Del’Arco Jr. - Institute of Aeronautics and Space - São José dos Campos - Brazil Carlos de Moura Neto - Technological Institute of Aeronautics - São José dos Campos - Brazil Cynthia C. Martins Junqueira - Institute of Aeronautics and Space - São José dos Campos - Brazil Eduardo Morgado Belo - University of São Paulo - São Carlos - Brazil Elizabeth da Costa Mattos - Institute of Aeronautics and Space - São José dos Campos - Brazil Flaminio Levy Neto - Federal University of Brasília - Brasília - Brazil Gilberto Fisch - Institute of Aeronautics and Space - São José dos Campos - Brazil João Luiz F. Azevedo - Institute of Aeronautics and Space - São José dos Campos - Brazil José Márcio Machado - Univ. Estadual Paulista - São José do Rio Preto - Brazil José Roberto de França Arruda - State Universiy of Campinas- Campinas - Brazil Marco Antônio Sala Minucci - Institute for Advanced Studies - São José dos Campos - Brazil Marcos Pinotti Barbosa - Federal University of Minas Gerais - Belo Horizonte - Brazil Mischel Carmen N. Belderrain - Technological Institute of Aeronautics - São José dos Campos - Brazil Paulo Tadeu de Melo Lourenção - Embraer - São José dos Campos - Brazil Valder Steffen Junior - Federal University of Uberlândia - Uberlândia - Brazil Waldemar de Castro Leite - Institute of Aeronautics and Space - São José dos Campos - Brazil

Editorial Production Glauco da Silva Helena Prado A. Silva Janaina Pardi Moreira Mônica Elizabeth Rocha de Oliveira

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 233-236, Sep. - Dec., 2011

233

Editorial Board

Editorial Board Acir Mércio Loredo Souza - Federal University of Rio Grande do Sul - Porto Alegre - Brazil Adam S. Cumming - Defence Science and Technology Laboratory - Fort Halstead - UK Adrian R. Wittwer - National University of the Northeast - Resistencia - Argentine Alain Azoulay - Superior School of Eletricity - Paris - France Alexandre Queiroz Bracarense - Federal University of Minas Gerais - Belo Horizonte - Brazil Antonio Henriques de Araujo Jr - State University of Rio de Janeiro - Rio de Janeiro - Brazil Antonio Sérgio Bezerra Sombra - Federal University of Ceará - Fortaleza- Brazil Bert Pluymers - Catolic University of Leuven - Leuven - Belgium Carlos Eduardo S. Cesnik - University of Michigan - Ann Arbor - USA Carlos Henrique Marchi - Federal University of Paraná - Curitiba - Brazil Charles Casemiro Cavalcante - Federal University of Ceará - Fortaleza - Brazil Cosme Roberto Moreira da Silva - University of Brasília - Brasília - Brazil Edson Aparecida de A. Querido Oliveira - University of Taubaté - Taubaté - Brazil Edson Cocchieri Botelho - Univ. Estadual Paulista - Guaratinguetá - Brazil Fabrice Burel - National Institute of Applied Sciences - Lion - France Fernando Luiz Bastian - Federal University of Rio de Janeiro - Rio de Janeiro - Brazil Francisco Souza - Federal University of Uberlândia - Uberlândia - Brazil Frederic Plourde - Superior National School of Mechanics and Aerotechnics - Poitiers - France Gerson Marinucci - Institute for Nuclear and Energy Research São Paulo - Brazil Gilson da Silva - National Industrial Property Institute - Rio de Janeiro - Brazil Hazin Ali Al Quresh - Federal University of Santa Catarina - Florianópolis - Brazil Hugo P. Simão - Princeton University - Princeton - USA João Amato Neto - University of São Paulo - São Paulo - Brazil Joern Sesterhenn - University of Munich - Munich - Germany Johannes Quaas - Max Planck Institute for Meteorology - Hamburg - Germany John Cater - The University of Auckland - Auckland - New Zealand José Alberto Cuminato - São Carlos School of Engineering - São Carlos - Brazil José Ângelo Gregolin - Federal University of São Carlos - São Carlos - Brazil José Atílio Fritz Rocco - Technological Institute of Aeronautics - São José dos Campos - Brazil José Carlos Góis - University of Coimbra - Coimbra - Portugal José Leandro Andrade Campos - University of Coimbra - Coimbra - Portugal José Maria Fonte Ferreira - University of Aveiro - Aveiro - Portugal José Rubens G. Carneiro - Pontifícia Univers. Católica de Minas Gerais - Belo Horizonte - Brazil Juno Gallego - Univ. Estadual Paulista - Ilha Solteira - Brazil Ligia M. Souto Vieira - Technological Institute of Aeronautics - São José dos Campos - Brazil Luis Fernando Figueira da Silva - Pontifical Catholic University - Rio de Janeiro - Brazil Luiz Antonio Pessan - Federal University of São Carlos - São Carlos - Brazil Márcia Barbosa Henriques Mantelli - University of Santa Catarina - Florianópolis - Brazil Maurizio Ferrante - Federal University of São Carlos - São Carlos - Brazil Michael Gaster - University of London - London - UK Mirabel Cerqueira Resende - Institute of Aeronautics and Space - São José dos Campos - Brazil Nicolau A.S. Rodrigues - Institute for Advanced Studies - São José dos Campos - Brazil Paulo Celso Greco - São Carlos School of Engineering - São Carlos - Brazil Paulo Varoto - São Carlos School of Engineering - São Carlos - Brazil Rita de Cássia L. Dutra - Institute of Aeronautics and Space - São José dos Campos - Brazil Roberto Costa Lima - Naval Research Institute - Rio de Janeiro - Brazil Samuel Machado Leal da Silva - Army Technological Center - Rio de Janeiro - Brazil Selma Shin Shimizu Melnikoff - University of São Paulo - São Paulo - Brazil Tessaleno Devezas - University of Beira Interior - Covilha - Portugal Ulrich Teipel - University of Nuremberg - Nuremberg - Germany Vassilis Theofilis - Polytechnic University of Madrid - Madrid - Spain Wim P. C. de Klerk - TNO Defence - Rijswijk - The Netherlands 234

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 233-236, Sep. - Dec., 2011

ISSN 1984-9648 ISSN 2175-9146 (online)

Journal of Aerospace Technology and Management Vol. 03, No. 03, Sep. - Dec., 2011

CONTENTS Editorial 237

Aeronautical technology in Brazil: a long way to go Mauro Kern Technical Papers

239

Finite element procedure for stress amplification factor recovering in a representative volume of composite materials Paulo Cesar Plaisant Junior, Flávio Luiz de Silva Bussamra, Francisco Kioshi Arakaki

251

15:1 Resonance effects on the orbital motion of artificial satellites Jorge Kennety S. Formiga, Rodolpho Vilhena de Moraes

259

Rate control system algorithm developed in state space for models with parameter uncertainties Adilson Jesus Teixeira

269

Assessment of the synthesis routes conditions for obtaining ammonium dinitramide by the FT-IR José Irineu Sampaio de Oliveira, Márcio Yuji Nagamachi, Milton Faria Diniz, Elizabeth da Costa Mattos, Rita de Cássia Lazzarini Dutra

279

Radar absorbing materials based on titanium thin film obtained by sputtering technique Viviane Lilian Soethe, Evandro Luis Nohara, Luis César Fontana, Mirabel Cerqueira Rezende

287

Numerical evaluation of an air-to-air missile radar cross section signature at X-band Marcelo Bender Perotoni, Luiz Alberto Andrade

295

Sensitivity analysis of airport noise using computer simulation Flavio Maldonado Bentes, Jules Ghislain Slama

301

Lightning risk warnings based on atmospheric electric field measurements in Brazil Marco Antonio da Silva Ferro, Jorge Yamasaki, Douglas Roberto M. Pimentel, Kleber Pinheiro Naccarato, Marcelo Magalhães Fares Saba

311

Experimental results from the sounding vehicle Sonda III test campaign in the Pilot Transonic Wind Tunnel João Batista P. Falcão Filho, Maria Luísa Collucci C. Reis, Algacyr Morgenstern Jr.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 233-236, Sep. - Dec., 2011

235

COMUNICATIONS 325

VSB-30 sounding rocket: history of flight performance Alexandre Garcia, Sidney Servulo Cunha Yamanaka, Alexandre Nogueira Barbosa, Francisco Carlos Parquet Bizarria, Wolfgang Jung, Frank Scheuerpflug

331

Preliminary development plan of the ALR, the laser rangefinder for the ASTER deep space mission to the 2001 SN263 asteroid Antonio Gil Vicente de Brum, Annibal Hetem Jr., Israel da Silveira Rêgo, Cayo Prado Fernandes Francisco, Andre Fenili, Fernando Madeira, Flavio Caldas da Cruz, Marcelo Assafin Thesis abstract

339

Modeling and simulation of a capacitive microaccelerometer Janderson Rocha Rodrigues

339

Production of TiN coatings by EB-PVD on titanium alloys obtained via powder metallurgy Eduardo Tavares Galvani

340

Synthesis of polyaniline on a pilot scale for the processing of microwave absorbers with silicone matrix Joseane Mercia da Rocha Pimentel Gonçalves

341

AD HOC REFEREES

343 Instructions to the Authors

236

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 233-236, Sep. - Dec., 2011

Mauro Kern*

Executive Vice President, Engineering and Technology EMBRAER mkern@embraer.com.br

Editorial Aeronautical technology in Brazil: a long way to go

The Brazilian aeronautical industry today enjoys a respectable reputation and is one of the all-time success stories of the country. If we take into account the number of airplane manufacturers that disappeared in the last 40 years – either in mergers/acquisitions or simply that have gone bankrupt –, it is amazing to see a Brazilian company as one of the largest in this industry, increasing its relative importance year after year. There are only a few technologically advanced sectors in Brazil that are considered as world-class, and the aeronautical industry is certainly one of them. The Brazilian aeronautical industry – Embraer in particular – is the consequence of a strategic vision and long, sustained effort. Right after the Second World War, the vision of the aeronautical technology as a powerful lever for the country development became strong in the minds of some Brazilian military officers. An aeronautical industry in Brazil could only be envisaged upon the solid foundations of education, knowledge, and technology. Therefore, the first initiatives were the creation of the Centro Técnico de Aeronáutica (CTA) and Instituto Tecnológico de Aeronáutica (ITA) in the late 1940s, to lay the ground for the development and manufacture of the future airplanes. It was in the CTA that the Bandeirante was conceived and designed by ITA engineers in the 1960s, being the precursor of a very successful genealogy of airplanes to be designed and built by Embraer. If we look back, we can easily conclude that Embraer always developed and brought to the market the right airplane at the right time. And it was not by chance, the company has been ready, technology and business wise, as opportunities arise. The EMB-110 Bandeirante was ready in the 1970’s by the time deregulation happened in the USA. With deregulation, a great number of regional airlines was created and a great number of regional airplanes was needed. The 19-seater, nonpressurized EMB-110 Bandeirante certification to the strict North-American and European requirements was quite an accomplishment at the time, and a very important step towards real global operations. In the early 1980’s, with the development of the regional aviation around the world, a larger, faster, and more comfortable airplane was required. The 30-seater, pressurized EMB-120 Brasilia was a bold initiative, bringing several innovative features to this market and being certified to large airplanes FAR Part 25 standards. In the 1990’s, another step change was demanded by the continuing growth of regional airlines. This time, the 50-seat jet ERJ-145 was the right airplane at the right time, when pilot union agreements with the airlines – the scope-clauses – boosted the demand for this category of airplanes. The ERJ-145 was developed during very difficult times for Embraer, right before its privatization, and other innovations in the business model – like risk-sharing partnerships – were brought to the market. In the 2000’s, the 70 to 120 seat E-Jets marked the beginning of a new era, bringing comfort and performance standards of larger jets, thus blurring the line between regional and mainline aviations.

*Mauro Kern was born in 1961 in Porto Alegre, Brazil. He received a Mechanical Engineering degree from the Federal University of Rio Grande do Sul (Universidade Federal do Rio Grande do Sul – UFRGS) in 1982 and attended several extension courses in Business Administration. He started his career at Embraer in 1982 as a Systems Engineer, being assigned to the development of the landing gear system for the AMX military aircraft. In 1984, he joined EDE (Embraer´s Equipment Division specialized in landing gear and hydraulic equipment), leading engineering, marketing, sales, customer support and program management activities for over a decade. In 1999, Mauro joined the EMBRAER 170/190 Program, acting in the Program Management Office. He played a key role in the development of the EMBRAER 190, as Chief Engineer and Program Manager. In July 2005, he was assigned to the position of Vice President, Airline Market Programs, where his responsibilities included product development, planning, customer support, technical and commercial negotiations with suppliers, program management and support to the sales campaigns, especially those involving the E-Jets family. In April 2007, Mauro took the responsibility for one of Embraer’s most important business units, as Executive Vice President, Airline Market. In April 2010, he was assigned to the position of Executive Vice-President, New Programs, Airline Market, with the responsibility to develop strategies and bring new commercial airplanes to the market, thus ensuring longevity and growth to this business. In April 2011, he was assigned to the position of Executive Vice-President, Engineering and Technology.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 237-238, Sep. - Dec., 2011

237

Mauro Kern

In more recent years, with the strategic determination to become also a key player in the executive jets business, new products have been developed and are changing the shape of this market. The Phenom 100 and 300, the Legacy 600 and 650 and the Lineage 1000 have made a huge difference in their respective segments, attracting strong interest from very demanding customers. Every new airplane has been a complete new story, with long development cycles permeated by the evolution of technology and the urge to bring innovative products to the market. The new aircraft under development, like the Legacy 500 and the KC-390, and the future developments for the commercial aviation market will bring even greater challenges to Embraer. The technology intensive nature of the aviation industry poses unique challenges to the aircraft manufacturers. The new technology development cycles are long, typically around 15 years from a new discovery or invention to the maturity level necessary for their application and certification on an airplane. No single company can afford the huge expenses necessary to be the state-of-the-art in this industry. Global partnerships, including manufacturers, research centers and universities, with substantial government support, are more necessary for any company to remain competitive. An additional challenge is posed to the aviation industry these days: global warming. Despite the fact that air transport is responsible today for only 2 to 3% of green-house gas emissions, the growth of air travel demand, especially in developing countries, will make it increase its relative contribution to alarming levels in the next decades. Emissions are intimately linked with fuel burn, which is the number one cost element of the airlines. With the competitive nature of the airline business, the quest for efficiency has made airplane and engine manufacturers strive for developing and adding new technologies at every new generation of products. In fact, very few industry segments have seen such an important efficiency improvement â&#x20AC;&#x201C; airplanes today burn around 70% less fuel than they did 40 years ago. The next generation aircraft will incorporate more and more technology to cope with this environmental challenge, such as biofuels, new and lighter materials, advanced flight controls and navigation systems, just to name a few. In summary, education, knowledge, and technology formed the foundation for the creation of the aeronautical industry in Brazil more than 60 years ago. They will keep being key factors to ensure competitiveness and growth of this industry for decades to come.

238

J. Aerosp.Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.3, No.3, pp. 237-238, Sep. - Dec., 2011

doi: 10.5028/jatm.2011. 03033911 Paulo Cesar Plaisant Junior

EMBRAER São José dos Campos/SP – Brazil paulo.plaisant@embraer.com.br

Flávio Luiz de Silva Bussamra*

Instituto Tecnológico de Aeronáutica São José dos Campos/SP – Brazil flaviobu@ita.br

Francisco Kioshi Arakaki

EMBRAER São José dos Campos/SP – Brazil francisco.arakaki@embraer.com.br *author for correspondence

Finite element procedure for stress amplification factor recovering in a representative volume of composite materials Abstract: Finite element models are proposed to the micromechanical analysis of a representative volume of composite materials. A detailed description of the meshes, boundary conditions, and loadings are presented. An illustrative application is given to evaluate stress amplification factors within a representative volume of the unidirectional carbon fiber composite plate. The results are discussed and compared to the numerical findings. Keywords: Micromechanics, Finite element analysis, Composites, Stress amplification factor, Microstress distribution.

INTRODUCTION By analyzing history, it is possible to see the importance of material science applied to Aeronautical Engineering and, in that scenario, the composite materials emerged. As this type of material became more recognized, new branches of research came out. One of these branches is the micromechanics, which is the theory that this study focuses on. Since composite materials play an important role in the modern industry, it is necessary a better understanding of them. Particularly in aeronautical industry, metal alloys have been replaced by composite materials. The best example is the latest Boeing aircraft, 787: 50% of its structure is made of composite materials and 20% of aluminum. Its predecessor has 12% of composite materials and 50% of aluminum (Boeing, 2010). The analysis of composite materials follows a macro, meso, or micromechanical approach. Macromechanics analyzes a laminated plate as a homogeneous anisotropic equivalent plate. In the mesomechanical approach, a laminate is modeled as a stacking sequence of homogeneous layers and interlaminar interfaces (Ladevèze et al., 2005), therefore the prediction of complex behavior, as delamination, can be assessed (Allix, Ladevèze and Corigliano, 1995). The micromechanical analysis goes down to the constituent properties. The object of study on the micromechanical analysis is the representative volume element (RVE), or unit cell, which is the smallest cell capable of representing the overall response of the unidirectional ply to mechanical and thermal loading (Jin et al., 2008). Figure 1 illustrates an example of RVE. Received: 06/09/11 Accepted: 20/10/11

Unidirectional ply

Representative Volume Element

Figure 1. Representative volume element in a unidirectional fiber composite ply.

Jin et al. (2008) show the use of three-dimensional finite element models for obtaining stress distribution on composites. Micromechanical finite element models provide data to obtain the micro-stresses at the matrix/ fiber interface, and great benefits can be reached when a micromechanical approach is considered. Micromechanics of failure gives a more precise way of composite failure prediction. With the micro-stresses, it is possible to determine the failure initiation in the unidirectional ply (Ha, Huang and Jin, 2008a; Tay et al., 2008; Gotsis, Chamis and Minnetyan, 1998). The material lifetime forecast can be obtained with the use of micromechanics of failure associated with the Accelerated Testing Method (ATM) and Evolution of Damage (Sihn and Park, 2008; Ha, Huang and Jin, 2008b). The micromechanical theory considers not only the mechanical loads, but also environmental factors, such as thermal loads, as a result of temperature variation and moisture (Hyer and Waas, 2000; Fiedler, Hojo and Ochiai, 2002).

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 239-250, Sep. - Dec., 2011

239

Plaisant Junior, P.C., Bussamra F.L.S., Arakaki F.K.

The influence of fiber arrangements on mechanical behavior of laminated pates is discussed by Hojo et al. (2009) and Ha, Huang and Jin (2008c). Considerations about the fabrication process are discussed by Aghdam and Khojeh (2003). Studies with reinforcements other than fibers, such as particles, are exposed by Zhu, Cai and Tu (2009). Different types of composites, like bulk metallic glasses (Dragoi et al., 2001), metallic matrix composites (Chaboche, Kruch and Pottier, 1998), and smart composites, which include piezoelectric composites, shape memory alloy (SMA) fiber composites, and piezoresistive composites (Taya, 1999), are also analyzed by micromechanics. Liang, Lee and Suaris (2006) present the results of a comparison between micromechanical finite elements modeling and mechanical testing. RVE use in order to represent the composite material is discussed by Sun e Vaidya (1995), and a study of boundary conditions for the unit cell is shown by Xia et al. (2003). From the aeronautical industry to dentistry, a great variety of products can be benefited from a micromechanical study of a composite material. Sakaguchi, Wiltbank and Murchison (2003) show micromechanical studies to predict composite elastic modulus and polymerization shrinkage for dental materials. The micromechanical theory has direct application for the aeronautical industry. Tsai (2008) points out the use of micromechanics to:

material, mostly because of the dissimilarity on physical properties of the materials. For example, epoxy matrixes present a lower young modulus than the carbon fiber. When a load is applied to a composite material, due to this stiffness difference, the matrix and the fiber tend to show different stresses, resulting in stress concentrations. Therefore, when a unit load is applied to the material, the stresses within the representative volume are no longer unitary, as shown in Fig. 2. σ≠1 σ≠1

σ≠1

fiber

σ≠1

matrix

Figure 2. Differences between micro and macro stresses.

•

predict macro mechanical properties (stiffness constants, expansion coefficients);

Jin et al. (2008) show that there are amplification factors which relates a uniformly distributed unit load (σ, the macro mechanical load) and the internal micromechanical stresses σ, expressed in Eq. 1:

•

control the deformation from mechanical and thermal loads;

ı "M ı A T

•

predict a successive ply failure after the first ply failure and;

•

adjust empirical data by using micromechanical data.

The objective of this paper is to present and to discuss a methodology in order to obtain the stress amplification factors derived from mechanical and thermal loads in a RVE, with the use of two and three-dimensional finite elements models. Also, this paper aims at accessing stress amplification factors in an orthotropic unidirectional ply with epoxy matrix, carbon fiber, and a perfectly bonded matrix/fiber interface, with 60% of fiber volume fraction. The materials remain in the linear elastic domain. The finite element models are analyzed with the commercial software MSC/NASTRAN, version 70.0.6 (MSC, 2011). STRESS AMPLIFICATION FACTORS In a micromechanical level, there is a difference between the applied and actual stresses within the 240

(1)

where, M and A are matrices that collect the mechanical and thermal stress amplification factors, respectively, and ∆T is the increase of the room temperature. Considering all stress components, referred to the material coordinate system xyz (the same as 123), Eq. 1 can be expanded by Eq. 2 (Jin et al., 2008): ı1 = ıxx

M11

ı2 = ıyy ı3 = ızz ı4 = ıyz

M12

M13

M14

M 21 M 22

M 23

M 24

M 31

M 32

M 33

M 34

M 41 M 42

M 43

M 44

ı5 = ızx

M 55

M 56

ı6 = ıxy

M 65

M 66

ı1 ı2 ı3 + ı4 ı5 ı6

A1 A2 A3

¨T

A4 A5 A6

(2)

M can be found by applying unidirectional mechanical loads, one at a time. For instance, if a uniformly distributed unit load is applied at x direction, with no thermal load, Eq. 2 simplifies to Eq. 3:

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 239-250, Sep. - Dec., 2011

Finite element procedure for stress amplification factor recovering in a representative volume of composite materials

Äą1

M11

M12

M13

M14

Äą2

M 21 M 22

M 23

M 24

Äą3

M 31

Äą4

M 32

M 33

M 34

M 41 M 42

M 43

M 44

Äą5

M 55

M 56

Äą6

M 65

M 66

1 0 0 0 0 0

(3)

Äą

The resolution of the linear system (Eq. 3) yields Eq. 4: Äą1

M 11

Äą2 Äą3 Äą Äą Äą

4 5 6

M 21

M 31

(4)

M 41 0 0

Therefore, the stress amplification factors M11, M21, M31 and M41 will be actually the micromechanical stresses Ď&#x192;1, Ď&#x192;2, Ď&#x192;3 and Ď&#x192;4, respectively. The same procedure can be applied to all directions. Consequently, the methodology consists of the application of uniformly distributed unit loads to the representative volume model. Thus, the resulting stress at a specific direction gives the corresponding stress amplification factor. Since the stresses at the representative volume vary at each point, the stress amplification factor is not constant. Materials The RVE of a composite material with 60% of fiber volume fraction, subjected to an uniformly distributed load Ď&#x192;2, is showed in Fig. 3. The fiber is represented as

a solid cylinder. The mechanical properties of the matrix and fiber are listed in Table 1, where Eij are Youngâ&#x20AC;&#x2122;s moduli, Î˝ij are Poissonâ&#x20AC;&#x2122;s ratios, Gij are shear moduli, and Îą the thermal expansion coefficient, referred to the material coordinate system xyz (the same as 123). Table 1. Mechanical properties of the representative volume element materials (Think Composites, 2011).

Carbon fiber E11 (Pa) E22=E33 (Pa) G12=G13 (Pa) G23 (Pa) Î˝12 Î˝13 Îą1 (10-6/ÂşC) Îą2, Îą3 (10-6/ÂşC)

2.35x10 1.80x1011 7.48x109 4.90x109 0.20 0.30 0.0 8.3 11

Epoxy matrix Em (Pa) 3.46x109 Î˝m 0.35 57.6 Îąm (10-6/ÂşC)

Numerical ANALYSiS Three sets of finite element models are presented and discussed. In the first set, solid hexahedral elements are used to model the matrix and the fiber. The unit cell is represented as a cube with nondimensionalized edge length (L = 1). Convergence analysis is performed, and stress amplification factors for direct, shear, and thermal loads are presented and discussed. Three-dimensional finite elements are still used in the second set, but the unit cell is no longer modeled as a cube, saving computing efforts. The last set of tests deals with two-dimensional elements. Although they are unable to yield stress concentration factors in transverse directions, in-plane factors are derived. The finite elements solutions are compared with results from Super Mic-Mac software (Think Composites, 2011). Three-dimensional finite element model

The first set of finite element analysis is applied to a three-dimensional model, with the eight-node hexahedral CHEXA Nastran element for the matrix and fiber modeling. The model, including boundary conditions and loads, are further discussed. Loads and boundary conditions y

Ď&#x192;2 = 1

Figure 3. Representative volume element of a composite material with 60% of fiber volume fraction.

When a uniformly distributed tension load is applied to a RVE, the unit cell is constrained at its faces, as shown in Fig. 4, and the free faces must remain flat, as proposed by Jin et al. (2008), and Xia, Zhang and Ellyin (2003). To keep free faces flat, rigid elements are applied to the model. The Nastran rigid element has one master node and one or more slave nodes. The master is the

J. Aerosp.Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.3, No.3, pp. 239-250, Sep. - Dec., 2011

241

Plaisant Junior, P.C., Bussamra F.L.S., Arakaki F.K.

independent one, and it can receive loads. Each slave node will have the same displacements (in the specified direction) of the master one. Therefore, to keep the right vertical face flat in Fig. 4, a uniformly distributed unit load Ď&#x192;2 is modeled as a force over only one node. This master node is connected in direction y to the free face nodes (slave nodes) by rigid elements, so all the nodes in this face will have the same displacements in direction y. The nodes in this face must be free at x and z directions (not connected to the master node), in order to move according to the Poisson effect. Rigid body modes are constrained at x=0, y=z=-0.5. Figure 5 shows one model with rigid, spring, and hexahedral elements.

When a uniformly distributed shear load is applied to the RVE, the unit cell is constrained at its faces as shown in Fig. 6, and proposed by Ha et al. (2008). The same procedure already explained is applied here. Rigid body modes are constrained.

faces remains flat Free Free faces remains flat

Free move Freeto to move onon z z direction direction Other degrees degrees of of Other freedom constraimed constrained freedom

The same procedure must be applied to the others free faces. As it is impossible to enforce two dissimilar displacements on a node, nodes cannot figure as dependent on two different master nodes. Therefore, nodes at the edges are connected by spring elements with high stiffness coefficients. For instance, for face z=0.5 it is necessary to connect the rigid element to the nodes on the edges parallel to the x-axis. However, these nodes are dependent ones on the rigid elements from faces y=0.5 and y=-0.5. Thus, Nastran spring element Celas2, with high stiffness (say, 1010), is used to connect only the z displacements for that particular face and, then, this face remains flat. With the Celas2 element, it is possible to connect only one of the degrees of freedom. From faces z=0.5 and z=-0.5, it is connected the z displacements. For faces x=0 and x=1, it is connected the x displacements.

Ď&#x192;22= 1 All of freedom freedom Alldegrees degrees of constrained, except constrained, except xx translation and translation andxxrotation rotation

zz xx

Free faces Free faces remain remainflat flat

Free move on y y Free totomove on direction direction Other degrees of Other degrees of freedom constrained freedom constraimed

xxtranslation and translation and xx rotation rotation constrained constrained

Figure 4. Unit cell constraints, subjected to the uniformly distributed unit load Ď&#x192;2.

AllAll degrees degreesof of freedom freedom constrained constrained except except xx translation rotation translationand and x x rotation

Figure 6. Unit cell constraints, subjected to a uniformly distributed shear load Ď&#x192;6 = 1.

Convergence analysis

Spring Elements (CELAS2) Same x displacement for the nodes connected by cach spring element at face x=0 and x=1

Ď&#x192;2 = 1

Rigid Element (RBE2) The face remains flat (same x displacement for the nodes at face x=1)

z x

Rigid Element (RBE2) The face remains flat (same z displacement for the nodes at face z=0.5)

Figure 5. Finite element model with: rigid (RBE2), spring (CELAS2) and hexahedral (CHEXA). 242

11 Ď&#x192;66 =

Rigid Element (RBE2) The face remains flat (same z displacement for the nodes at face z=0.5)

Spring Elements (CELAS2) Same z displacement for the nodes connected by cach spring element at face z=0.5 and z=-0.5

To perform a convergence analysis, a uniformly distributed unit load is applied at direction y. Six finite element meshes are tested (Fig. 7 and Table 2), from a poorly refined 136 element mesh (Mesh 1), to a highly refined 19046 element mesh (Mesh 6). In order to present the results of the convergence analysis, it is necessary to define a point nomenclature. The points are numbered according to Tay et al. (2008), as illustrated in Fig. 8. Points 1, 7 and 4 are the best

J. Aerosp.Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.3, No.3, pp. 239-250, Sep. - Dec., 2011

Finite element procedure for stress amplification factor recovering in a representative volume of composite materials

Mesh 1 136 elements

Mesh 2 566 elements

Mesh 3 1950 elements

Mesh 4 4022 elements

Mesh 5 8230 elements

Mesh 6 19046 elements

Figure 7. Representative volume three-dimensional element meshes. Table 2. Number of elements and degree of freedom (DOF) in three-dimensional models.

Mesh

Element type CHEXA CELAS2

RBE2

Total

DOF

136

286

480

566

1641

1800

144

1950

5813

3840

176

4022

12171

Fiber

8000

224

8230

24999

Matrix

18720

320

19046 57765

z x

suitable to have the results displayed, since they are the points of maximum, minimum, and intermediate σy stresses, respectively. The results of the convergence testing are shown in Figs. 9 and 10. It can be seen that after 4,000 elements (Mesh 4), the major stress σmax at Point 1 starts to converge to the value of 1.5. Above 8,000 elements, the results are in the curve asymptotic portion, therefore the chosen mesh is the number 6. Main stresses at points 1, 4 and 7 are compared with results from the Super Mic-Mac Plus (SMM) software (Think Composites, 2011), which are presented in Table 3. SMM has an extensive database of stress amplification factors resultant from finite element analysis for a wide range of physical properties of fiber and matrix, and different volume fractions. It uses interpolation methods to give the values for specifications, which are not in the database.

Figure 8. Point nomenclature.

The stress distribution σy (micromechanical) at the RVE is shown in Fig. 11. Since the load is σ 2=1, the stress distribution σy will characterize the stress amplification factor (M22=1.55). Stress amplification factors for mechanical loads The stress amplification factors with three-dimensional element meshes are found by applying direct and shear uniformly distributed unit loads at the RVE. As already discussed, the stress contours presented in Figs. 12 and 13 represent the distributions of the stress amplification factors. Table 4 shows all the 36 stress amplification factors at Point 1. These results are compared with solutions by SMM in Tables 5 and 6.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 239-250, Sep. - Dec., 2011

243

Plaisant Junior, P.C., Bussamra F.L.S., Arakaki F.K.

Mesh 1 ımax=1.4125

Mesh 2 ımax=1.4798

Mesh 3 ımax=1.4864

Mesh 4 ımax=1.4922

Mesh 5 ımax=1.4952

Mesh 6 ımax=1.4965

Figure 9. Major stress σmax at the representative volume element.

1,6 Point 1

1,4 1,2 1 ıyy 0,8

Point 4

0,6 0,4 0,2 0

Point 7 0

2000

4000

6000

8000 10000 12000 14000 16000 18000 20000

Figure 10. Stress convergence in the representative volume element. Table 3. Stress recovering for representative volume element subjected to a uniformly distributed load σ2 = 1.

Point 1 4 7 244

3D model (mesh 6) 1.49646 0.80874 0.05139

SMM 1.511233 0.810486 0.050587

Difference (%) 0.98 0.22 -1.59

Stress amplification factors for thermal loads Stress amplification factors for thermal loads are obtained by increasing the room temperature by ∆T=1º C. The finite element analysis with Mesh 6, with the same constraints that were previously discussed, yields the stress contours depicted in Fig. 14. Table 7 presents the differences found with SMM.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 239-250, Sep. - Dec., 2011

Finite element procedure for stress amplification factor recovering in a representative volume of composite materials

1.55 1.457 1.363 1.269 1.176 1.082 0.988 0.895 0.801 0.707 0.614 0.52 0.426 0.332 0.239 0.145 0.0514

z x

Output Set: MSC/NASTRAN Case 1 Contour: Solid Y Normal Stress

Figure 11. Stress amplification factor M22.

σ1=1

M11

σ2=1 1.66

M12

σ3=1

M13

0.0038

ıXX

0.00

M21

0.77

-0.0034

M22

1.55

-0.0034

M23

0.73

ıyy -0.32

M31

0.77

0.05

M32

0.73

-0.30

M33

1.55

ızz -0.32

-0.30

0.05

Figure 12. Stress contours within the RVE, subjected to macro-mechanical tension load.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 239-250, Sep. - Dec., 2011

245

Plaisant Junior, P.C., Bussamra F.L.S., Arakaki F.K.

σ4=1

M44

σ5=1

σ6=1

M45

1.66

M46

ıYZ 0.00

M54

M55

1.55

M56

ıZX 0

M64

0.05

M65

M66

1.55

ıXY 0

0.05

Figure 13. Stress contours within the RVE, subjected to macro-mechanical shear load. Table 4. Stress amplification factors from Nastran at Point 1, with Mesh 6.

Stress σxx σyy σzz σyz σzx σxy

σ1=1

2.43x10 -3.14x10-3 2.89x10-3 5.36x10-8 -2.38x10-10 5.61x10-9 -2

σ2=1

7.74x10 1.50 7.33x10-1 -7.71x10-6 1.90x10-10 -4.69x10-6 -1

σ3=1

-8.00x10 -2.62x10-1 5.14x10-2 3.59x10-6 -1.37x10-9 8.96x10-7

σ23=1

σ31=1

σ12=1

-2

-5

-1.45x10 1.19x10-5 -5.32x10-5 1.19 1.26x10-9 -2.94x10-9

-9

-2.73x10 -1.25x10-9 -1.32x10-9 -1.07x10-12 1.60x10-1 -4.85x10-6

-2.20x10-6 -7.22x10-7 -9.73x10-7 -2.37x10-11 -2.67x10-5 1.59

σ3=1 -8.47x10-2 -2.75x10-1 5.06x10-2 1.98x10-11 -4.81x10-13 3.34x10-13

σ23=1 -1.15x10-11 1.95x10-11 -4.29x10-11 1.21 3.63x10-12 -1.75x10-12

σ31=1 -1.87x10-12 2.34x10-12 5.16x10-13 9.42x10-14 1.59x10-1 9.80x10-12

σ12=1 5.72x10-13 5.10x10-12 1.13x10-12 1.15x10-13 -2.53x10-11 1.59

Table 5. Stress amplification factors from SMM at Point 1.

Stress σxx σyy σzz σyz σzx σxy

σ1=1 2.43x10-2 -3.22x10-3 2.89x10-3 1.10x10-12 -4.05x10-14 2.27x10-15

σ2=1 7.81x10-1 1.51 7.34x10-1 -5.67x10-11 2.24x10-12 -4.80x10-13

Table 6. Difference between stress amplification factors between Nastran model (Mesh 6) and SMM at Point 1.

Stress σxx σyy σzz σyz σzx σxy 246

σ1=1

0.00% 2.62% -0.14% -

σ2=1

0.86% 0.98% 0.16% -

σ3=1

5.65% 4.82% -1.59% -

σ23=1

1.69% -

σ31=1

-0.57% -

σ12=1

-0.15%

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 239-250, Sep. - Dec., 2011

Finite element procedure for stress amplification factor recovering in a representative volume of composite materials

ΔT = 1°C

2.2x105

σxx -2.5x105

2.2x105

z x y Output Set: MSC/NASTRAN Case 1 Contour: Solid Y Normal Stress

1.55 1.457 1.363 1.269 1.176 1.082 0.988 0.895 0.801 0.707 0.614 0.52 0.426 0.332 0.239 0.145 0.0514

Figure 15. Representative volume element in a cutaway view on σy stress contour.

σyy -2.5x105

2.2x105

σzz -2.5x105

Figure 14. Stress contours within the RVE, subjected to thermal load ∆T = 1ºC. Table 7. Stress amplification factors at Point 1 for thermal load.

Stress σxx σyy σzz σyz σzx σxy

Model

SMM

-1.92x105 2.06x105 -1.90x105 -3.63 -1.02x10-5 -5.92x10-1

-1.90x105 2.11x105 -1.90x105 -1.75x10-5 7.89x10-7 -1.20x107

Difference (%) -0.74 2.59 0.17 -

reducing the number of elements in x direction. The cubic finite element model showed in Fig. 5 is reduced on its half and the resulting model is called Model 1/2. Those model simplifications are carried out up to Model 1/32, as presented in Fig. 16. The stress amplification factors found with Models 1 to 1/32 are virtually the same, as long as the mesh stays three-dimensional (solid elements). For example, the difference between the full size model and the one with 1/32 of thickness (Model 1/32) is, in the worst scenario, 0.01%. The next step of this study is to verify the differences between the results of the three-dimensional Model 1/32 and two-dimensional ones. Two different two-dimensional Nastran elements are used: the four-node (linear) quadrilateral membrane CQUAD4 element and the eightnode (parabolic) quadrilateral membrane CQUAD8. Figure 17 shows σy stress contours for the RVE subjected to a uniformly distributed unit load σ 2. The correspondent stress amplification factors M22 are listed in Table 8. Table 9 shows that two-dimensional models fail in calculating accurate in-plane results. Conclusion

Finite element model simplifications It is notable that all the results are constant within the element along the x axis, as illustrated in details in Fig. 15. This suggests that: a courser three-dimensional mesh can be used at x direction or the model does not need necessarily to be three-dimensional. A series of model simplifications is carried out. The first part focuses on three-dimensional simplifications by

In this paper, a finite element procedure is presented to access internal stresses in micromechanical analysis of composites materials, with unidirectional fibers. First, three-dimensional finite element models of a RVE are idealized and modeled in Nastran, with CHEXA (hexahedral), CELAS2 (spring), and rigid (RBE2) elements. Convergence analysis is performed. Stress amplification factors are derived within a RVE under tension, shear, and thermal load. Then, two-dimensional Nastran models are also proposed.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 239-250, Sep. - Dec., 2011

247

Plaisant Junior, P.C., Bussamra F.L.S., Arakaki F.K.

Model 1 Thickness=1 Model 1/2 Thickness=0.5 Model 1/4 Thickness=0.25 Model 1/8 Thickness=0.125 Model 1/16 Thickness=0.0625

z x

Model 1/32 Thickness=0.03125

Output Set: MSC/NASTRAN Case 1 Contour: Solid Y Normal Stress 1.55

1.457

1.363

1.269

1.176

1.082

0.988

0.895

0.801

0.707

0.614

0.52

0.426

0.332

0.239

0.145

0.0514

Figure 16. Three-dimensional models for RVE, subjected to macro-mechanical unit load σ2(σy contours).

1.55

1.589

1.595

1.457

1.496

1.502

1.269

1.311

1.316

0.988

1.033

1.036

0.426

0.477

0.0514

0.107

0.105

(a)

(b)

(c)

Figure 17. Stress amplification factors M22 with: (a) three-dimensional Model 1/32 (2984 DOF); (b) linear membrane elements (1464 DOF); (c) parabolic membrane elements (4368 DOF). Table 8. Stress amplification factors with parabolic membrane elements.

Stress σyy σzz σyz

σ1=1

σ2=1

σ3=1

1.545112 0.480766 -

0.152274 0.104816 -

1.287447

The presented two-dimensional models fail in providing accurate in-plane stress amplification factors. Good estimation for internal stress amplification factors is achieved with three-dimensional models. Good results 248

Table 9. Differences between stress amplification factors found with Model 1/32 and with parabolic membrane elements.

Stress σyy σzz σyz

σ1=1

σ2=1

σ3=1

3.25% 34.38% -

41.92% 103.97% -

8.17%

can be derived with a single transverse layer of solid finite elements, an important feature for nonlinear analyses. The good performance of the presented threedimensional models shows that good estimates for stress

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 239-250, Sep. - Dec., 2011

Finite element procedure for stress amplification factor recovering in a representative volume of composite materials

amplification factors can be derived for other than the presented composite (with another volume fraction, or for bidirectional fibers composite), and also to access strain amplification factors. REFERENCES Allix, O., Ladevèze P., Corigliano, A., 1995, “Damage analysis of interlaminar fracture specimens”, Composite Structures, Vol. 31, No. 1, p. 61-74. doi:10.1016/02638223(95)00002-X Aghdam, M., Khojeh, A., 2003, “More on the effects of thermal residual and hydrostatic stresses on yelding behavior of unidirectional composites”, Composite Structures, Vol. 62., No. 3-4, p. 285-90. Boeing. 787 Dreamliner – Program Fact Sheet. Retrieved in 2010 August 2, from: http://www.boeing.com/ commercial/787family/programfacts.html. Chaboche, J.L., Kruch, S., Pottier, T., 1998, “Micromechanics versus Macromechanics: a combined approach for metal matrix composites constitutive modeling”, European Journal of Mechanics, Vol. 17, No. 6., p. 885-908. Dragoi, D., et al., 2011, “Investigation of thermal residual stresses in tungsten-fiber/bulk metallic glass matrix composites”, Scripta Materialia, Vol. 45, No. 2, p. 245-52. Fiedler, B., Hojo, M., Ochiai, S., 2002, “The influence of thermal residual stresses on the transverse strength of CRFP using FEM. Composites Part A: Applied Science and Manufacturing”, Vol. 33., No. 10, p. 1323-6. Gotsis, P.K., Chamis, C.C., Minnetyan, L., 1998, “Prediction of Composite laminate fracture: micromechanics and progressive fracture”, Composites Science and Technology, Vol. 58, No. 7, p. 1137-49. Ha, S.K., Huang, Y., Jin, K.K., 2008a, “Effects of Fiber Arrangement on Mechanical Behavior of Unidirectional Composites”, Journal of Composites Materials, Vol. 42, No. 18, p. 1851-71. Ha, S.K., Huang, Y., Jin, K.K., 2008b, “Life Prediction of Composites using MMF and ATM”. In: Tsai, S., 2008, “Strength and Life of Composites”, Stanford, Department of Aeronautics & Astronautics of Stanford University. Ha, S.K., Huang, Y., Jin, K.K., 2008c, “Micro-Mechanics of Failure (MMF) for Continuous Fiber Reinforced

Composites”, Journal of Composites Materials, Vol. 42, No. 18, p. 1873-95. Hojo, M., et al., 2009, “Effect of fiber array irregularities on microscopic interfacial normal stress states of transversely loaded UD-CFRP from viewpoint of failure initiation”, Composites Science and Technology, Vol. 69, No. 11-2, p. 1726-34. Hyer, M.W., Waas, A.M., 2000, “Micromechanics of Linear Elastic Continuous Fiber Composites”, In: Kelly, A., Zweben, C., 2000, “Comprehensive Composite Materials”, Elsevier Science Publishers. Jin, K.K., et al., 2008, “Distribution of micro stresses and interfacial tractions in unidirectional composites”, Journal of Composite Materials, Vol. 42, No. 18. Ladevèze, P., et al., 2005, “Micro and meso computational damage modellings for delamination prediction”, ICF XI - 11th International Conference on Fracture. Liang, Z., Lee, H.K., Suaris, W., 2006, “Micromechanicsbased constitutive modeling for unidirectional laminated composites”, International Journal of Solids and Structures, Vol. 43, No. 18-9, p. 5674-89. Msc Software, 2011, “MD Nastran: Integrated, Multidiscipline CAE Solution”, Retrieved in 2011 March 23, from http://www.mscsoftware.com/Products/ CAE-Tools/MD-Nastran.aspx. Sagushi, R., Wiltbank, B., Murchison, C., 2004, “Prediction of composite elastic modulus and polymerization shrinkage by computational micromechanics”, Dental Materials, Vol. 20, No. 4, p. 397-401. Sihn, S., Park, J. W., 2008, “An Integrated Design Tool for Failure and Life Prediction of Composites”, Journal of Composites Materials, Vol. 42, No. 18, p. 1967-88. Sun, C. T., Vaidya, R. S., 1996, “Prediction of Composite Properties from a Representative Volume Element”, Composites Science and Technology, Vol. 56, No. 2, p. 171-9. Tay, T.E., et al., 2008, “Progressive Failure Analysis of Composites”, Journal of Composites Materials, Vol. 42, No. 18, p. 1921-66. Taya, M., 1999, “Micromechanics modeling of smart composites. Composites Part A: applied science and manufacturing”, Vol. 30, No. 4, p. 531-6.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 239-250, Sep. - Dec., 2011

249

Plaisant Junior, P.C., Bussamra F.L.S., Arakaki F.K.

Think Composites, 2011, “Super Mic-Mac – a new preliminary design tool for composites”, Retrieved in 2011 March 23, from http://www.thinkcomposites.com/ index_eng.php.

Xia, Z., Zhang, Y., Ellyin, F., 2003, “A unified boundary conditions for representative volume elements of composites and applications”, International Journal of Solids and Structures, Vol. 40, No. 8, p. 1907-21.

Tsai, S., 2008, “Strength and Life of Composites”, 1st ed, Stanford: Department of Aeronautics & Astronautics of Stanford University.

Zhu, L.J., Cai, W.Z., Tu, S.T., 2009, “Computational Micromechanics of Particle reinforced Composites: Effect of Complex Three-dimensional Microstructures”, Journal of Computers, Vol. 4, No. 12, p. 1237-42.

250

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 239-250, Sep. - Dec., 2011

doi: 10.5028/jatm.2011.03032011

Jorge Kennety S. Formiga*

15:1 Resonance effects on the orbital motion of artificial satellites

Rodolpho Vilhena de Moraes

Abstract: The motion of an artificial satellite is studied considering geopotential perturbations and resonances between the frequencies of the mean orbital motion and the Earth rotational motion. The behavior of the satellite motion is analyzed in the neighborhood of the resonances 15:1. A suitable sequence of canonical transformations reduces the system of differential equations describing the orbital motion to an integrable kernel. The phase space of the resulting system is studied taking into account that one resonant angle is fixed. Simulations are presented showing the variations of the semi-major axis of artificial satellites due to the resonance effects. Keywords: Resonance, Artificial satellites, Celestial mechanics.

FATEC - College of Technology São José dos Campos/SP – Brazil jkennety@yahoo.com.br

Federal University of São Paulo São José dos Campos/SP – Brazil rodolpho@gmail.com

*author for correspondence

INTRODUCTION The problem of resonance effects on orbital motion of satellites falls under a more categorical problem in astrodynamics, which is known as the one of zero divisors. The influence of resonances on the orbital and translational motion of artificial satellites has been extensively discussed in the literature under several aspects. In fact, for instance, it has been considered the (Formiga, 2005): resonance of the rotation motion of a planet with the translational motion of a satellite (Lima Jr., 1998; Formiga, 2005); sun-synchronous resonance (Hughes, 1980); spin-orbit resonance (Beleskii, 1975; Vilhena De Moraes e Silva, 1990); resonances between the frequencies of the satellite rotational motion (Hamill and blitizer, 1974); and resonance including solar radiation pressure perturbation (Ferraz Mello, 1979). Generally, the problem involves many degrees of freedom because there can be several zero divisors, but although the problem is still analytically unsolved, it has received considerable attention in the literature from an analytical standpoint. It justifies the great attention that has been given in literature to the study of resonant orbits characterizing the dynamics of these satellites, as can be seen in recent published papers (Deleflie, et al., 2011; Anselmo and Pardini, 2009; Chao and Gick, 2004; Rossi, 2008). In this paper, the type of resonance considered is the commensurability between the frequencies of the satellite mean orbital motion and the Earth rotational one. Such case of resonance occurs frequently in real cases. In fact, in a survey from a sample of 1818 artificial satellites, chosen in a random choice from the NORAD 2-line elements (Celestrak, 2004), about 85% of them are orbiting near Received: 23/05/11 Accepted: 07/08/11

some resonance’s region. In our study, it satellites in the neighbourhood of the 15:1 resonance (Fig. 1), or satellites with an orbital period of about 1.6 hours will be considered. In our choice, the characteristic of the 356 satellites under this condition can be seen in Table 1 (Formiga, 2005).

Satéllite R

Planet Figure 1. Orbital resonance. Table 1. Orbital characteristics for 15:1 resonance.

Orbital characteristic for 15:1 resonance e≤0,1 e i≤5° e≤0,1 e i≥70° e≤0,1 e 55°<i<70° e≤0,1 e 5°<i<55°

Number of satellites 1 290 41 26

The system of differential equations describing the orbital motion of an artificial satellite under the influence of perturbations due to the geopotential, is described here in a canonical form. In order to study the effects of resonances, a suitable sequence of canonical transformations was performed reducing the system of differential equations to an integrable kernel (Lima Jr., 1998). This system is integrated numerically, and simulations can show the behaviour of motion in the neighbourhood of the exact resonance. Some

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 251-258, Sep. - Dec., 2011

251

Formiga, J.K.S., Moraes, R.V.

results are presented for the 15:1 resonances, where graphics representing the phase space and the time variation of some Keplerian elements are exhibited.

R mpq 2L2

(4)

with THE CONSIDERED POTENTIAL Using Hansenâ&#x20AC;&#x2122;s coefficients, the geopotential can be written as in Eq. 1 (OsĂłrio, 1973): R U" 2a Hq (

1),(

"2 m"0 p"0 q"

2 p)

R ae J m F mp (i) a a

(1)

a, e, I, M, ÎŠ, Ď&#x2030; are the Keplerian elements; E t is the sidereal time; Ď&#x2030;E is the Earthâ&#x20AC;&#x2122;s angular speed; J m are coefficients depending on the Earthâ&#x20AC;&#x2122;s mass distribution; F mp(i) represents the inclination functions; and Hq( +1),( -2p)(e) are the Hansenâ&#x20AC;&#x2122;s coefficients.

m) (

(2)

where, l m is the corresponding reference longitude of semi-major axis of symmetry for the harmonic ( ,m). Considering perturbations due to geopotential and the classical Delaunay variables, the Lagrange equations describing the motion can be expressed in canonical forms and a Hamiltonian formalism can be used. The Delaunay canonical variables are given by Eq. 3: L + a , G + a(1 e2 ) ,

H + a(1 e 2 ) cosi Â´

2 p)

1),((

(L, G)x

(5)

where,

m) (

m(h

2 p)g

(6)

METHODOLOGY The following procedure, including a sequence of canonical transformations, enables us to analyse the influence of the resonance upon the orbital elements (Lima Jr., 1998):

Y G L X L x Â´ g h y g h

(3)

M, g , h

Z H G

z h

(7)

Thus, the new Hamiltonian is H (X,Y, Z, , x, y, z, )

2X 2

R' mpq

(8)

where, R' mpq " Hq

where,

( 1),(

"2 m"0 p"0 q" 2 p)

R 2 ae J m F mp (X,Y, Z ) X 2 2

(9)

(X,Y )cos O mpq (x, y, z, )

b) a reduced Hamiltonian, containing only secular and periodic terms containing a commensurability between the frequencies of the motion, is constructed. c) A new Hamiltonian considering a given resonance is constructed. Thus, if n stands for the orbital mean motion, the resonance condition can be expressed by

Â´

, g,M h are coordinates; and L, G, H are the conjugated moment. Extending the phase space, where a new variable Ó¨, conjugated to Î&#x2DC;(t), is introduced to eliminate the explicit time dependence, the Hamiltonian F=F (L, G, H, Î&#x2DC;, Â´, g,M h, Ó¨) of the corresponding dynamical system is 252

xF mp (L, G, H )H q (

J mx

a) canonical variables (X, Y, Z, Î&#x2DC;, x, y, z, Ó¨) related with the Delaunay variables are introduced by the canonical transformation described as follows:

The argument is m(

l"2 m"0 p"0 q"

O mpq ( Â´, g, h, ) " q ' (

where:

2 p)

R 2 R ae L2 L2

cos O mpq ( ', g, h, )

(e)cos O mpq (M, , , )

O mpq (M, , , ) " qM (

R mpq "

qn m

(10)

where, q and m are integers. We will denote by Îą=q/m the commensurability of the resonance.

J. Aerosp.Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.3, No.3, pp. 251-258, Sep. - Dec., 2011

15:1 Resonance effects on the orbital motion of artificial satellites

The Hamilton is simplified as Hr "

R2 2X 2

j"1

(l 2 p m ) X1

B2 j,0, j,0 (X,Y,Z)

(11)

m) (X,Y, Z) cos O mp(

B mp(

"2 m"2 p"0

A first integral C2 given by

) (

" m( x 2 p)z m

(12)

m) , 2

(13)

where

F2 j,0, j H 0 (2 j 1),2 j

Y2 (k m )X1 m Y1 2

R 2 B mp( m) (X,Y, Z ) " 2 2 ae J m X F mp (X,Y, Z ) H m( 1), ( 2 p) (X,Y )

B2 j,0, j,0 (X,Y, Z)

+ 2 j 2 2 j ae J 2 j,0 X 4 j 2

(14)

are secular terms obtained from the conditions q=m=0 e =2p. Equation 14 shows all the resonant terms with tesseral harmonics concerned in the resonance Îą.

d) Using the first integral C1, the following Mathieu transformation (second canonical transformation) is introduced reducing the order of the system. X1 " X

x1 " x " (1

Y1 " Y Z! " (1

)X Y Z

Hc "

2 X 12

E 1

Blmp ,( =2 p=0

j "1 m)

e 1

B(2 p k)mp(

cos

mp,( m) ( x1 , y1 ,

B2 j,0, j,0 (X1, C1,C2 )

(15)

m) (x1,

ONE RESONANT ANGLE The influence of the resonance on the orbital motion can be analyzed integrating a system of differential equations, where the reduced Hamiltonian is obtained from the one with secular and resonant terms: B(2 p k)mp(

sin O(2 p k)mp(

B2,2 j ,0, j ,0 ( X1 , Y1 , C )

j"1

m) (X1, C1, C2 )

z1 " z

A frequency Îą is selected and we consider a new Hamiltonian system containing just secular and resonant terms

R2 2X12

dX1 " dt

1 "

y1 " y z

Fixing a value for one resonant angle, we can consider as short period terms all periodic terms different from the fixed one.

Hc "

cosO(2 p k)mp(

can be obtained for the new Hamiltonian system with the Hamiltonian given by Hr

1 y1 m

Introducing the coefficients k= -2p, with Âłâ&#x2030;Ľ2, sâ&#x2030;¤pâ&#x2030;¤â&#x2C6;&#x17E;, where s is value minimum by p to sÂłâ&#x2030;Ľ0 and k depends on the frequency chosen, we can determine a new dynamical system as function of Hc=Hc(X1,Î&#x2DC;1,x1,Î¸1), where Hc is

p"s

) X Y Z C1

k m ) m

is performed and A new Hamiltonian is defined as critical Hamiltonian (Hc), with secular terms in the X1 variable and the resonant terms with critical frequencies.

A first integral given by (1

x2 x1 (

X 2 X1

2 p m )y

m (

can be found for this new Hamiltonian system. e) A third canonical transformation given by

with O mp(

m Y1 C 2

m) (X1, C1, C2 )

p"s m) (x1,

and dO(2 p k)mp( dt

m m

R2 m X13

B2 j,0, j,0 (X1, C1, C2 ) X1

j"1

B(2 p k)mp( p"s

cos O(2 p k)mp(

m) (X1, C1, C2 )

m) (x1,

J. Aerosp.Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.3, No.3, pp. 251-258, Sep. - Dec., 2011

(16)

253

Formiga, J.K.S., Moraes, R.V.

The analysis of the effects of the 15:1 resonance will be done here considering the tesseral J15,15, and the zonal harmonic J2 with k=13. The dynamical systems (Eq.16) is:

3(C1 C2 )X1.sin

and

(17)

1 C 15X1 . arccos 1 13X1 C2 2

17 4,15020466 10 25.J15,15.a15 e .R

4(C2

26C2 X1 168X12 )cos

14(13C1 15C2 )X1 (C22

The secular and resonance terms are, respectively, Eqs.Â 18 and 19: R17 15 ae J15,15 X132

(18)

F15,15,1 (X1, C1, C2 )H1 16,13 (X1, C1 )

3X1 )31 X116 1

(C1 15X1 )2 (C2 13X1 )2

1 C 15X1 . arccos 1 13X1 C2 2

sin

1 C 15X1 . arccos 1 13X1 C2 2

26C2 X1 168X12 )

B cosO 15,15,1,1

(21)

15,15,1,1

" x1 15

1 15 15,15

(22)

Finally, we can compute the time variations for the Keplerian elements: a, e, i, through the inverse transformations:

X12 +

e 1

kX1 m

i cos

X12

mX1 C1 kX1 C2

(23)

where

and B2,0,1,0 (X1, C1, C2 ) " H 0 3,2 (X1, C2 )

R4 2 ae J 2,0 F2,0,1 (X1, C1, C2 ) X6

(19)

the Eqs. 20 and 21:

17 .J15,15.a15 e .R

1 e2

(24)

with C1 and C2 as integration constants. Results

dX1,15,15,1,1 4,15020466 " dt 4X18 (C2 3X1 )29 X14 25

+a k

C1 +a 1 e 2 cos i 1/ C2

Considering Eq. 18 and Eq. 19 we have explicit functions relating B(2p+k)mp(ÎŹm)(X1, C1, C2) and Flmpq(X1, C1, C2) from

1 C 15X1 cos .arcos 1 13X1 C2 2

254

1 C 15X1 . arccos 1 13X1 C2 2

4 6. cos

The critical angle is

B15,15,1,1 (X1, C1, C2 ) "

(C1 15X1 )2 (C2 13X1 )2

.(16C24 1040C23 X1 24333C22 X12 245609C2 X13 908544X14 )

B15,15,1,1 (X1, C1, C2 ) X1 cos O15,15,1,1 (x1,

3X1 )5 X14 1

4(C2

(C1 15X1 )2 1 C 15X1 cos . arccos 1 13X1 C2 2 (C2 13X1 )2

B2,0,1,0 (X1, C1, C2 ) X1

3.J 2 ae2 R 4

(C2 15X1 )2 (C22 X12

dO15,15,1,1 R 2 " 3 15 dt X1

9C2 X1 18X12 ) 1

(C22

dX1,15,15,1,1 " B15,15,1,1 (X1, C1, C2 ) dt sin O15, 15,11 (x1, 1 )

dO15,15,1,1 R 2 " 3 dt X1

(C2 28

3X1 )2 X12

sin O15,15,1,1

(20)

In this section we present numerical results considering some initial conditions arbitrarily chosen. Several harmonics can also be considered and as an example, for the 15:1 resonance, it was considered the simultaneous influence of the harmonics J2 e J15,15. Let us consider the case e=0.019, i=87Â°, Ď&#x2022;*=0 (critical angle) and the harmonics J2 and J15,15, where numerical values for the Harmonics coefficients are given by

J. Aerosp.Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.3, No.3, pp. 251-258, Sep. - Dec., 2011

15:1 Resonance effects on the orbital motion of artificial satellites

JGM-3 (Tapley et al.,1996). The phase space for the dynamical system (18)–(19) is presented by Fig. 2. This space is topologically equivalent to that of the simple pendulum. The behavior of the motion is presented here in the neighborhood of the separatrix (a about 6930 km). Figure 2 represents the temporary variation of the semimajor axis in the neighborhood of the 15:1 resonance. For one value considered for the semi-major axes, distinct behaviors for their temporary variations can be observed. The central circle represents a new region after an abrupt variation due to the resonance effect. The more the satellites approach the region, which we defined as a resonant one, the more the variations increase. It is remarkable the oscillation in the region between 1,420

Figure 2. Semi-axis versus critical angle: e=0.019, i=87º e ϕ*=0.

Figure 3. Variation of semi-major axis in the time: e=0.01, i=4°, ϕ*=0°.

and 1,440 days, which characterizes paths for which the effect of the resonance is maximum for the case, where e=0.01 and i=4°. This effect can be seen in Figs. 3 and 4 with more details. A new region of libration can be seen in Figs. 5 and 6 when satellites are outside the neighborhood of the resonance. The stabilization of the orbit that appears after a period of 1,420 days after a maximum is related with discontinuity produced by the resonance. By Figs. 7 and 8, we can see that the amplitude of variation of the semi major axis do not change in neighborhoods of the resonance. This libration motion remains for a long time. In both cases, in the resonance

Figure 4. Amplification: ∆a versus time: e=0.019, i=87º e ϕ*=0.

Figure 5. Variation of critical angle in the time: e=0.019, i=87º e ϕ*=0.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 251-258, Sep. - Dec., 2011

255

Formiga, J.K.S., Moraes, R.V.

region, we observed a region sensible to abrupt changes of orbital elements and the possible existence of chaotic region.

a=6930 km 1,5 1,0

(rad)

0,5 0,0 -0,5 -1,0 -1,5 1400 1410 1420 1430 1440 1450 1460 1470 1480 1490 1500 t (days)

Figure 6. Amplification: Variation of critical angle in the time: e=0.019, i=87º e ϕ*=0.

Several other initial conditions were considered and, as a sample, Table 2 contains the amplitude and period of the variations of orbital elements for hypothetical satellites considering low eccentricity, small and high inclinations, and influence of the harmonics J 20 and J 15,15. As it can be observed, the 15:1 resonance can produce a variation of more than 100 m in the semi-major axis and this must be considered in practice when orbital elements are used in precise measurements. CONCLUSIONS A sequence of canonical transformations enabled us to analyze the influence of the resonance on the orbital elements of artificial satellites with mean motion commensurable with the rotation. In this paper, an integrable kernel was found for the dynamical system describing the motion of an artificial satellite under the influence of the geopotential, and considering resonance between frequencies of the mean orbital motion and the Earth rotational motion. The theory, valid for any type of resonance p/q (p = mean orbital motion and q = Earth rotational motion), was applied to the dynamical behavior of a critical angle associated with the 15:1 resonance considering some initial conditions.

Figure 7. Variation of semi-major axis in the time: e=0.01, i=4°, ϕ*= 0°.

a=6932,39 km

This paper provides a good approach for long-period orbital evolution studies for satellites orbiting in regions where the influence of the resonance is more pronounced.

0,12 0,10

(km)

0,08 0,06

ACKNOWLEDGEMENTS

0,04 0,02 0,00 -0,02

The motion near the region of the exact resonance is extremely sensitive to small alterations considered. This can be an indicative that these regions are chaotic.

100

t (dias)

The authors also wish to express their thanks to CAPES (Federal Agency for Post-Graduate Education - Brazil), FAPESP (São Paulo Research Foundation), and INPE (National Institute for Space Research, Brazil) for contributing and supporting this research.

Figure 8. Amplification: ∆a in the time: e=0.01, i=4°, ϕ*= 0°.

256

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 251-258, Sep. - Dec., 2011

15:1 Resonance effects on the orbital motion of artificial satellites

Table 2. Maximum oscillation 15:1 resonance considering J2+J15,15.

Δa (m)

Δe

0.019

4°

120

0.006

ao–2,39

0.019

4°

120

0.0065

16.9x10-4

ao–6,39

0.019

4°

118

0.0062

11.1x10-4

ao+2,53

0.019

4°

100

0.0058

14.8x10-3

0.019

55°

125

0.0068

2.9x10-3

ao–2,39

0.019

55°

110

0. 0063

4x10-3

ao–6,39

0.019

55°

0.0055

3.5x10-3

ao+2,53

0.019

55°

125

0.0029

2.8x10-3

0.019

63,4°

120

0.0068

5.15x10-3

ao–2,39

0.019

63,4°

100

0.0058

4.4x10-3

ao–6,39

0.019

63,4°

0.0048

3.2x10-3

ao+2,53

0.019

63,4°

100

0.0067

5.7x10-3

0.019

87°

125

0.0057

5.72x10-3

ao–2,39

0.019

87°

120

0.0065

7.16x10-3

ao–6,39

0.019

87°

102

0.006

5.7x10-3

ao+4,61

0.019

87°

0.0055

5.73x10-3

ao=6932,39 km

REFERENCES Anselmo, L., Pardini, C., 2009, “Dynamical evolution of high area-to-mass ratio debris released into GPS orbits”, Advances in Space Research, Vol. 43, No. 10, p. 1491-1508. Beletskii, V. V., Resonance Phenomena at Rotations of Artificial and Natural Celestial Bodies. In: Giacaglia, G.E.O. “Satellites dynamics”. Berlin: Verlang, 1975. Celestrak, 2004, “Apresenta os elementos 2-line do NORAD”, Retrieved in August 10, 2004, from http:// www.celestrak.com. Chao, C.C., Gick, R.A., 2004, “Long-term evolution of navigation satellite orbits: GPS/GLONASS/GALILEO”, Advances in Space Research, Vol. 34, p. 1221-1226. Ferraz Melo, S., 1979, Periodic orbits in a region of instability created by independent small divisors. In: Nagoz, Y. E., Ferraz Melo, S. “Natural and artificial satellite motion”. Austin: University of Texas Press, p. 283-292. Florent, D., Alessandro, R., Christophe, P., Gilles, M., François, B., 2011, “Semi-analytical investigations of the long term evolution of the eccentricity of Galileo and GPS-like orbits”, Advances in Space Research, Vol. 47, No. 5, p. 811-821.

Δimáx(°) 2x10-4

Formiga, J. K. S., 2005, “Estudo de ressonâncias no movimento orbital de satélites artificiais”. 133f. Dissertação (Mestrado em Física) – Faculdade de Engenharia do Campus de Guaratinguetá, Universidade Estadual Paulista, Guaratinguetá. Hamill, P. J., Blitzer, L., 1974, “Spin-orbit coupling: a unified theory of orbital and rotational resonance”, Celestial mechanics, Vol. 9, p. 127-146. Hugues, S., 1980, “Earth satellite Orbits with Resonant Lunisolar Perturbations”. Resonances dependent only inclination. Proceedings of the Royal society of London serie A. London, Vol. 372, No. 1745, p. 243-264. Lima, Jr. P. H. C. L., 1998, “Sistemas ressonantes a altas excentricidades no movimento de satélites artificiais”. Tese (Doutorado), Instituto tecnológico de aeronáutica, São José dos Campos. Osório, J. P., 1973, “Perturbações de órbitas de satélites no estudo do campo gravitacional terrestre”. Porto: Imprensa Portuguesa. Rossi, A, 2008, “Resonant dynamics of medium earth orbits: space debris issues”, Celestial Mechanics and Dynamical. Astronomy, Vol. 100, p. 267-286.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 251-258, Sep. - Dec., 2011

257

Tapley, B. D., Watkins, M. M., Ries, J. C., Davis, G. W., Eanes, R. J., et alz, 1996, “The Joint Gravity Model 3”, Journal of Geophysical Research, Vol. 101, No. B12, p. 28029-28050.

258

Vilhena de Moraes, R., Silva, P. A. F., 1990, “Influence of the resonance in gravity- gradient stabilized satellite”. Celestial Mechanics and Dynamical Astronomy, Vol. 47, p. 225-243.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 251-258, Sep. - Dec., 2011

doi: 10.5028/jatm.2011. 03033511

Adilson Jesus Teixeira*

Institute of Aeronautics and Space São José dos Campos/SP – Brazil adilsonajt@iae.cta.br *author for correspondence

Rate control system algorithm developed in state space for models with parameter uncertainties Abstract: researching in weightlessness above the atmosphere needs a payload to carry the experiments. To achieve the weightlessness, the payload uses a rate control system (RCS) in order to reduce the centripetal acceleration within the payload. The rate control system normally has actuators that supply a constant force when they are turned on. The development of an algorithm control for this rate control system will be based on the minimum-time problem method in the state space to overcome the payload and actuators dynamics uncertainties of the parameters. This control algorithm uses the initial conditions of optimal trajectories to create intermediate points or to adjust existing points of a switching function. It associated with inequality constraint will form a decision function to turn on or off the actuators. This decision function, for linear time-invariant systems in state space, needs only to test the payload state variables instead of spent effort in solving differential equations and it will be tuned in real time to the payload dynamic. It will be shown, through simulations, the results obtained for some cases of parameters uncertainties that the rate control system algorithm reduced the payload centripetal acceleration below µg level and keep this way with no limit cycle. Keywords: Rate control, Off-on control, Time optimal control, State space, Bang-bang control.

INTRODUCTION The microgravity environment to perform experiments can be obtained in several ways. One of them uses a sounding rocket that carries a payload out of atmosphere influence and the experiments are performed during the payload ballistic phase. After the payload separation, from the sounding rocket, the payload needs to have its angular velocity reduced in order to minimize the centripetal acceleration of the embedded equipment. To achieve centripetal acceleration to microgravity level (µg), the payload needs to have a rate control system (RCS). The RCS for this purpose normally uses a cold gas subsystem (CGS), which is comprised of on-off actuators. This subsystem has three sets of solenoid valve thrusters, mounted in the service module skin of the payload, and each set can supply a constant torque that changes the angular velocity in one of the main payload axis. Many publications of studies that employ on/off control, which is also named as bang-bang control, can be found. The bang-bang control has been investigated for several decades and been applied in the most different areas. Received: 04/08/11 Accepted: 06/10/11

In Udrişte (2008), it is studied the controllability, observability and bang-bang properties of multi-time completely integrable autonomous linear systems described by partial differential equations (PDE). In O’Brien (2006), it is described a bang-bang control algorithm developed for a double integrator plant that can be extended to higher order plants with two integrators. This approach is studied to be used in a steering controller for an autonomous ground vehicle. Another application is the study for developing a bang-bang algorithm for a path tracking control to a differential wheeled mobile robot, where it shows that the bang-bang algorithm offers better results in accurate and time execution (Nitulescu, 2005). The switching-time computation for a bang-bang control law is developed by Lucas and Kaya (2001) to compute the switching times applied to a nonlinear system with one and after for two control inputs. Ettl and Pfänder (2009) describes the RCS used in several programs for weightlessness research in Europe and in Brazil. The aim of the RCS is to reduce the angular rates of the payload significantly above the atmosphere in order to minimize the centripetal accelerations to a level lower than 10 μg. This article describes the principle of a RCS. Remain publications in the references will be mentioned later in the development of the control algorithm.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 259-268, Sep. - Dec., 2011

259

Teixeira A.J.

A common problem to be solved for a RCS with a on/off control system is to obtain the instant and the duration of each control pulse to be applied to the actuators in order to reduce the payload angular velocity as close as possible to zero in minimum time, and to keep with no cycle limit amplitude. However, nonlinearities and parameters uncertainties of the system to be controlled can affect the desired performance of a designed control system.

MATHEMATIC MODELS

A basic RCS block diagram is shown in Fig. 1:

x'(t) = Ax(t) + Bu(t) y(t) = Cx(t)

A mathematic model can be a set of mathematic equations, linear or nonlinear, that approximately represents a specific behavior of a real system. A payload mathematical model can be represented by a LTI system in state space as shown in Eq. 1:

(1)

where: SU

Control

Actuators

Payload

Figure 1. Rate control system block diagram.

where: SU: represents the sensor unit that measures the angular velocities and accelerations of the payload at its three main axes; Control: represents the control unit that can be programmed with the control algorithm to turn on and off the actuators; and Actuators: represents the actuators units used to change the payload angular velocity. One method to develop a control system for on/off actuators can be found in minimum-time problem, in optimal control theory. This method permits to calculate the optimal time to turn on and off the actuators and, consequently, to get the optimal payload trajectory in the state space. Based on this, it will be developed a modified method that uses the initial conditions for an optimal trajectory to create or adjust a switching function. Minimum-time problem has a good performance when the parameters of a dynamic system are well known, but this performance is affected when there are parameter uncertainties. Therefore, the switching function must be created or adjusted to these uncertainties in real time. In this paper, it will be shown the performance and how to adjust the RCS control algorithm for the following parameter uncertainties: payload mass, payload inertial momentum, actuator torque level, and/or actuator response delay. This control algorithm also has the advantage that it does not require great effort in processing.

260

A ∈ ℜn x n is the dynamic matrix; B ∈ ℜn x p is the control matrix; C ∈ ℜp x r is the output matrix; x(t) ∈ ℜn is the state vector; u(t) ∈ ℜr is the control vector; and y(t) ∈ ℜp is the output vector. To develop the equations, terms that indicate function of t are going to be omitted. In order to show the development of the RCS algorithm and its performance, it was used the payload and actuator mathematical models for the main X axis. Payload mathematical model The payload mathematical model, according to Cornelisse, Schöyer and Wakker (1979), is represented here by a linear time invariant system of a rigid body, symmetric around body X axis. For microgravity experiments, the payload is controlled only during its ballistic trajectory phase above the atmosphere when its influence can be neglected; therefore, no external perturbation is considered. Based on these simplifications, the payload model can be represented by Eq. 2: ¬ lx ¼ ½ 2Fa ® xx ¾

[ p ' ] = [0 ] p + I

(2)

where: p: roll angular velocity (“roll-rate”), in [rad/s]; lx: arm of the actuator force, in [m]; Ixx: inertial momentum around body X axis, in [Kg m2]; Fa: force of each actuator, in [N], perpendicular to the body X axis. The actuator force is multiplied by two because there are two actuators for roll control, one at each opposite side of payload skin, to avoid body translation.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 259-268, Sep. - Dec., 2011

Rate control system algorithm developed in state space for models with parameter uncertainties

To design the RCS it was considered that the payload mathematical model parameters have the values given by Eq. 3:

can be incorporated into the model. Considering model uncertainties structured, the LTI model can be represented by Eq. 5:

lx= 0.5 m Ixx= 30 kg m2

x ' = A + ) A x + B + ) B [u + ) u]

(3)

y(t) = C + ) C x + s a

Actuator mathematical model

where:

The actuator mathematical model is represented here by an on/off second order system, with the following parameters: steady state force of 2 N, rise time (tr) of 50 ms, at 10% of steady state error, and damping ratio of 1.0; and an on/off input control ua. The hysteresis and the electromechanical delay that may affect the actuator response are not considered. Based on these simplifications, the actuator model can be represented by the LTI of Eq.Â 4 (Ogata, 1982):

â&#x2C6;&#x2020;A, â&#x2C6;&#x2020;B and â&#x2C6;&#x2020;C denotes plant parameter uncertainties, â&#x2C6;&#x2020;u is the input parameter uncertainty, sa is the sensor additive noise measurement.

ÂŹx 'a 1 Âź Â Â˝= ÂÂŽx 'a 2 Â˝Âž

ÂŹ 0 1 Âź ÂŹx a1 Â Â˝Â ÂŽ-6006.25 -155 Âž ÂÂŽ x a 2

ÂŹ x a1 Fa = ÂŹÂŽ 1 0 ÂźÂž Â ÂÂŽ x a 2

Âź ÂŹ 0 Âź Â˝+Â Â˝ ua 6006.25 Â˝Âž ÂŽ Âž

Âź Â˝ Â˝Âž

(4)

where:

xa: actuator state vector; ua:Â input control: value 1 (Actuator ON) and value 0 (Actuator OFF); and Fa: actuator force, in [N]. The actuator response to unitary pulse width command of 0.3 s is shown in Fig. 2. The input signal is represented by a black dash line and the actuator force response by the blue solid line.

RCS DESIGN To explore the performance of the CGS, it is necessary to control the actuator during its steady state and its transient response. The problem is how to calculate the moment and pulse width to be applied to the actuator for these completely different behaviors. There are many methods to design a control law for an on/off actuator. Bryson and Ho (1997) presented a design, using the minimum-time problem method in state space to get the optimal trajectory. As there are parameter uncertainties in the mathematical models, it was used this concept to develop a control algorithm that the control law could self adjust to these parameters in real time. Some generic cases of phase plane approach can be found in Ogata (1982) and of state space approach can be found in Takahashi, Rabins and Auslander (1972). Switching function design The first approach of designing a rate control system in state space for a system, which has parameters uncertainties, is presented in Teixeira (2009). Here it is presented an improvement of this approach in order to get better performance and to reduce the limit cycle.

2 1.5 Force [N]

(5)

0.5 0 0

0.05 0.1

0.15 0.2

0.25 0.3 Time [s]

0.35 0.4

0.45 0.5

Figure 2. Actuator response to the unitary pulse.

Model parameter uncertainties If uncertainties about the model or unmeasured inputs to the process are structured (Frisk, 1996), that is, it is known how they enter at the system dynamics; this information

Choosing the state variable p and pâ&#x20AC;&#x2122; for the state space, the optimal trajectory is that one, which takes the payload state from an initial state to a final one, considered here the origin of state space, in minimum time. As can be seen in Fig. 2, after turning off the actuator there is a residual thrust that affects the payload state. Therefore, for the system control purpose this residual thrust need to be considered when getting optimal trajectory. Considering that the actuator is turned off during its positive or negative steady state, it can be obtained two optimal trajectories from these respectives payload mathematical model states towards the state space origin. These optimal trajectories are shown in Fig. 3.

J. Aerosp.Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.3, No.3, pp. 259-268, Sep. - Dec., 2011

261

Teixeira A.J.

as those used for the payload and actuator mathematical models. Therefore, the switching function presented in Fig. 5 will not keep the performance designed for the RCS. Thus, this algorithm must be adjusted to these unknown values.

0.06 p’ [rad/s/s]

0.04 0.02 0 -0.02 -0.04 -0.06 -0.08 -2

-1.5

-1

-0.5

0 0.5 p [rad/s]

1.5

2 x 10-3

Figure 3. Optimal trajectory that moves the payload mathematical model state vector to the state space origin.

These optimal trajectories could be used as a decision for the control algorithm only if the actuator was turned off from its steady state, but as it is desired to explore the CGS performance, the RCS control algorithm needs to operate also during the actuator transient response, therefore, it is necessary to build the optimal trajectories for various pulse width command during the actuator transient response. Figure 4 shows the angular acceleration of the payload mathematical model, in time, when the actuator is turned on and off during its transient response.

p’ [N]

0.06 0.04 0.02 0 -0.02

-0.06

0.06

-0.08 -2

0.05 0.04

-1.5

-1

-0.5

0 p [rad/s]

0.03

0.5

1.5

2 x 10-3

Figure 5. Switching function to control the payload when its actuator is commanded during its transient response.

0.02 0.01 0.05 0.1

0.15 0.2 0.25 0.3 0.35 0.4 Time [s]

0.45 0.5

Figure 4. Angular acceleration of the payload mathematical model for three different pulse width commands during the actuator transient response.

Through several simulations, with various pulse width for the actuator operating in its transient response, and rebuilding the respective optimal trajectories, it is possible to get the initial condition, where the actuator should be turned off. By interpolating all initial conditions, it is possible to obtain one curve that will split the state space where the actuator must be turned on or off. This curve will be called switching function and it is shown in Fig. 5. Control law design To design the control law, it must be considered that the payload and actuator dynamic parameters are not the same 262

0.08

-0.04

0.08 0.07

The dynamic behavior of the real payload will be only obtained during the moment that it receives any input command, therefore, the switching function must be created or adjusted in real time. The easiest way to create it will be through linear segments connecting each point obtained for the switching function. Therefore, to begin this process the switching function is initially represented by two linear segments, one between the points in state space (‑pMax, p´Max) and (0, 0) and another between points (0, 0) and (pMax, ‑ p´Max). These segments are the initial conditions for the optimal trajectory from Fig. 5. Additional points shall be added or adjusted after each measure of the state vector error, which will be explained later.

p’ [rad/s2]

0.08

After several simulations, it was observed that the final conditions of the switching function for the RCS algorithm needed to be changed a little bit and had to to be added some inequality constraints (Bryson et al., 1997) to compensate the simulation step and to reduce the limit cycle amplitude, when angular velocity of the payload mathematical model is close to zero. These inequality constraints and the switching function compound a decision function that will generate the conditions to turn on or off the actuator. Using the state space approach, the control system algorithm only needs to test the payload state vector against the decision function instead of spending a lot of processing in integrating and calculating the width and the instant that each pulse should be applied to the actuators. To describe the decision function it is shown, in Fig. 6 only the considerations of control for negative angular velocity.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 259-268, Sep. - Dec., 2011

Rate control system algorithm developed in state space for models with parameter uncertainties

pâ&#x20AC;&#x2122; pâ&#x20AC;&#x2122;Max Transition A OFF ON

Transition B pâ&#x20AC;&#x2122;Min ON/ OFF

-pMax

OFF p -pâ&#x20AC;&#x2122;Min

-2pMin -pMin

pMin

Figure 6. Decision function for negative angular velocity control.

where:

â&#x20AC;˘

pMax: maximum negative angular velocity value of the switching function; -2pMin: twice the desired increment for the negative angular velocity of the switching function; â&#x20AC;&#x2018;pMin and pMin: angular velocity precisions according to microgravity level specified; pâ&#x20AC;&#x2122;Max: maximum positive angular acceleration value of the switching function; and â&#x20AC;&#x2018;pâ&#x20AC;&#x2122;Min and pâ&#x20AC;&#x2122;Min : angular accelerations to measure the state vector error related to the state space origin.

Note: the best control law to generate the pulses for the blue area is under study. A good response was obtained for a pulse width equals to one step integration (1 ms) and pulse period equals to the actuator falling time (50 ms). By this way the payload state vector moves toward the origin of the state space.

The decision function for negative angular velocity that controls the positive actuator is compounded by four areas and two transitions described below: â&#x20AC;˘ â&#x20AC;˘

â&#x20AC;˘

â&#x20AC;˘ â&#x20AC;˘

Red Area: the payload angular velocity must be reduced. The positive actuator must be turned on. Blue Area: close to the specified state space area for low gravity level (green area). The control algorithm for this area will generate command pulses to the actuator to drive the payload state vectors to the state space origin. Green Area: corresponds to the state space area around the state space origin to get the desired low gravity level. At this area, the algorithm control must keep the actuator turned off. White area: corresponds to the payload state vector where the actuator must be turned off. Transition A: payload state vector when the control algorithm generates the command to turn off the actuator. The correspondent angular velocity of the switching function will be able to be adjusted.

Transition B: payload state vector when the angular acceleration reaches the range from â&#x20AC;&#x201C;pâ&#x20AC;&#x2122;Min to pâ&#x20AC;&#x2122;Min. The payload angular velocity measured will be used to adjust the value of the angular velocity stored at the switching function when occurs the Transition A.

The control law must use the following tests, according to each area: 1. Red Area: Actuator ON

p < S (p') U p < -p f

Max

(6)

2. Blue Area: Actuator ON/OFF (Pulse Commands) (7) 3. Green Area: Actuator OFF

(8)

4. White Area: Actuator OFF remain (p, pâ&#x20AC;&#x2122;) states while pâ&#x20AC;&#x2122; > pâ&#x20AC;&#x2122;Min

(9)

where: Sf(pâ&#x20AC;&#x2122;) is the switching function in function of pâ&#x20AC;&#x2122;.

J. Aerosp.Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.3, No.3, pp. 259-268, Sep. - Dec., 2011

263

Teixeira A.J.

To adjust or create a switching function segment, according to the payload dynamic, the RCS algorithm needs to process the following steps: â&#x20AC;˘

It was also considered the noise at sensors measures in the simulations. This noise was modeled as a color noise with Gaussian distribution, given by Eq. 11, which was added to the true sensor measure:

detect the Transition A;

f x | Č? ,Äą =

2 2 e 2Äą - x-Č?

â&#x20AC;˘

store the payload state vector (pA, pâ&#x20AC;&#x2122;A);

â&#x20AC;˘

detect the Transition B;

where:

â&#x20AC;˘

obtain the state vector error that is the payload state vector at transition B;

â&#x20AC;˘

if the state vector error is outside the blue and green areas do the correction of the switching function:

Îź=0 is the noise mean; Ď&#x192;av=2eâ&#x20AC;&#x2018;5Â rad/s is the noise standard deviation for angular velocity; Ď&#x192;aa=2eâ&#x20AC;&#x2018;5Â rad/s2 is the noise standard deviation for angular acceleration.

If there is not a segment for (pA, pâ&#x20AC;&#x2122;A), then create it; else correct the stored value of the angular velocity of the switching function through the Eq. 10: Sf (pâ&#x20AC;&#x2122;A) = pA â&#x20AC;&#x201C; pb

RCS ALGORITHM PERFORMANCE The RCS algorithm performance was verified through simulation to evaluate its behavior and the limit cycle amplitude of the payload angular velocity. To do so, it was used an angular velocity range from (â&#x20AC;&#x2018;1.0Â toÂ +1.0)Â eâ&#x20AC;&#x2018;4rad/s as the final payload angular velocity. These values represent a range from (â&#x20AC;&#x2018;5Â toÂ +5)Â e-9 m/s2 that is lower than the needed one to get the microgravity condition for the parameters values of payload mathematical model given by Eq. 3. This velocity angular range was used to test the limit cycle amplitude that would be generated by the RCS algorithm. The initial values used for simulation were:

(11)

The simulations were performed using: Simulink version 6.3, MatLab, version R14 SP3, Dormand-Prince solver and 1 ms integration step. The RCS performance is shown in the following cases. Case I In case I, the payload and actuators mathematical models have the nominal parameters specified for the RCS design given by Eq. 3 and 4. FigureÂ 7 shows a state space with the payload trajectory toward the state space origin, where: black solid line represents the payload state vector trajectory; red dashed line represents the maximum angular velocity for the switching function; solid red lines represent the switching function; blue solid lines represent the limit area for the on/off control; and green dashed lines represent the range specified for the angular velocity and acceleration near the vector state space origin. The abscissa axis is used for angular velocity in [rad/s] and the ordinate axis is used for angular acceleration [rad/s2].

0.08 0.06

Initial payload roll angular velocity: p0=0.005Â rad/s (initial value used only for graphical purposes). Maximum values for the switching function, obtained from Fig. 5:

0.04 pâ&#x20AC;&#x2122; [rad/s2]

â&#x20AC;˘

(10)

Regarding to the decision function for the control of the positive angular velocity, appropriate considerations shall be made in order to develop the whole algorithm control.

â&#x20AC;˘

Äą 2Ę&#x152;

0.02 0 -0,02 -0,04

|pMax| = 1.7eâ&#x20AC;&#x2018;3 rad/s and |pâ&#x20AC;&#x2122;Max| = 7.0e-2 rad/s2. â&#x20AC;˘

Angular and acceleration steps for the switching function:

|pMin| = 1.0e-4 rad/s and |pâ&#x20AC;&#x2122;Min| = 4.1e-3 rad/s2. 264

-0,06 -0,08 -2

-1

1 2 p [rad/s]

5 x 10-3

FigureÂ 7. Payload matethematical model trajectory for nominal parameters.

J. Aerosp.Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.3, No.3, pp. 259-268, Sep. - Dec., 2011

Rate control system algorithm developed in state space for models with parameter uncertainties

It can be seen in Fig. 7 that the RCS algorithm turn the negative actuator on and after some time turn it off in the optimal time to drove the payload state vector close to the state space origin in minimum time.

p [rad/s]

Control

Figure 8 shows the actuator control signal and the respective payload angular velocity in time representation. First graphic represents the control signal to the actuator, where positive values correspond to the commands for positive actuator and negative values correspond to the commands for the negative one. Second graphic corresponds to the payload angular velocity and the third graphic is the magnified payload angular velocity in the specified desired range for the design (-pMin < p, pMin). 1 0 -1 0

0.05

x10-3 4 2 0 0 0.05

0.1

0.15

0.2

0.25

0.3

0.35 0.4

0.45 0.5

get the desired low gravity level, the RCS control algorithm created one segment for the switching function, according to the error measured during the transition B and generated another command to the negative actuator to drive the payload state vector to the origin of the state space. Comparing Fig. 9 to Fig. 7, it can be seen that the changes at the switching function were due to the changes of the payload and actuator dynamic parameters. Figure 10 shows the time response of the actuator control signal and the respective payload angular velocity. First graphic shows the control signal to the negative actuator. Second graphic shows the behavior of the payload angular velocity and the third one is a magnified view of the angular velocity graphic to show that the payload angular velocity was reduced to the desired angular velocity range and it was maintained within this range with no limit cycle.

0.08 0.1

0.15

0.2

0.25

0.3

0.35 0.4

0.06

0.45 0.5

0.04

2 1 0 -1 -2 0

0.05

0.1

0.15

0.2

0.25

0.3

0.35 0.4

0.45 0.5

p’ [rad/s2]

p [rad/s]

x10-4

0.02 0 -0.02 -0.04

Time [s]

-0.06

Figure 8. Actuator control signal and payload mathematical model angular velocity for nominal parameters.

-1

1 2 p [rad/s]

5 x10-3

Figure 9. Payload matethematical model state vector trajectory for case II parameters.

Control

The payload mathematical model angular velocity was reduced to the desired range at minimum time with no limit cycle. It can be seen a small pulse width command to the positive actuator. This command was generated due to the noise at the angular velocity measure.

-0.08 -2

Case II

1 0 -1 0

0.05

0.1

0.15

0.2

0.25

0.3

0.35 0.4

0.45 0.5

0.05

0.1

0.15

0.2

0.25

0.3

0.35 0.4

0.45 0.5

x10-4 2 1 0 -1 -2 0 0.05

0.1

0.15

0.2

0.25

0.3

0.35 0.4

0.45 0.5

Figure 9 shows the payload state vector trajectory. It can be noted that the negative actuator was initially turned off before the optimal time, related to the new payload and actuator dynamic characteristics. As the payload state vector did not reach the area specified in the state space to

p [rad/s]

Case II shows the RCS algorithm performance considering the following parameters changes to the mathematical models: mass is 20% heavier, which increases the inertial momentum (Ixx = 36 kg m2); the actuator is faster, its rise time was 30% reduced (tr = 35 ms), and the actuator force was 25% reduced (Fa = 1.5 N).

p [rad/s]

x10-3 4 2 0 0

Time [s]

Figure 10. Actuator control signal and plant angular velocity for case II parameters.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 259-268, Sep. - Dec., 2011

265

Teixeira A.J.

Case III

Figure 11 shows the payload state vector trajectory. It can be noted that the negative actuator was turned off initially after its optimal time, related to the new payload and actuator dynamic characteristics. As the payload state vector did not reach the low gravity level specified in the state space, the RCS control algorithm changed the maximum positive angular velocity of the switching function from (1.7 to 3.1) e‑3rad/s, according to the error measured during the transition B. After that, the RCS control algorithm generated commands and created segments to the switching function.

0.06 0.04 p’ [rad/s2]

0.02 0 -0.02 -0.04 -0.06 -0.08 -1

p [rad/s]

5 x10-3

Figure 11. Payload matethematical model state vector trajectory for case III parameters.

Figure 12 shows the time response of the actuator control signal and the respective payload angular velocity. First graphic shows the control signal to the negative actuator. Second graphic shows the behavior of the payload angular velocity and the third one is a magnified view of angular velocity graphic to show that the payload mathematical model angular velocity was reduced to the desired angular velocity range and it was maintained within this range with no cycle limit.

266

Control

0 -1 0

0.1

0.15

0.2

0.25

0.3

0.35 0.4

0.45 0.5

x10-3 4 2 0 -2 0 0.05

0.1

0.15

0.2

0.25

0.3

0.35 0.4

0.45 0.5

x10-4 2 1 0 -1 -2 0 0.05

0.1

0.15

0.2

0.25

0.3

0.35 0.4

0.45 0.5

p [rad/s]

0.05

Time [s]

Figure 12. Actuator control signal and plant angular velocity for case III parameters.

CONCLUSION It was shown that a RCS algorithm developed presents good performance, although there are uncertainties at the payload and actuators dynamic parameters.

0.08

-0.1 -2

p [rad/s]

Case III shows RCS algorithm performance considering the following parameters changes to the mathematical models: the payload mass is 20% lighter, which reduces the inertial momentum (Ixx = 24 kgm2); the actuator is slower, its rise time was increased for 30% (tr = 65 ms); and the actuator force was increased for 25% (Fa = 2.5 N).

Using the state space approach, as the time is implicit; the system control needs only to test the state vector in state space instead to solve partial differential equations. Based on minimum-time problem method for an on/off control system, adding some additional inequality constraints and the capability for the control algorithm to adjust the switching function, it is possible to converge the payload state vector to a feasible desired range, although the payload and actuators have dynamic parameter uncertainties. Note that each point added to the switching function represents the optimal state to turn off the actuator. If this condition happens again, the state trajectory will follow the optimal trajectory. It was shown through simulation that the RCS algorithm could self adjust to parameter uncertainties such as: ±20% of payload mass change, ±30% of actuator rise time, and ±25% of actuator force. REFERENCES Bryson, Jr., Arthur, E., Ho, Yu-Chi, 1997, “Applied Optimal Control – Optimization, Estimation and Control”, Hemisphere Publishing Corporation, USA.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 259-268, Sep. - Dec., 2011

Rate control system algorithm developed in state space for models with parameter uncertainties

Cornelisse, J. W., Schöyer, H. F. R., Wakker, K. F., 1979, “Rocket Propulsion and Spaceflight Dynamics”, Pitman Publishing Limited, London, England. Ettl, J., Pfänder, J., 2009, “Rate Control System for Sounding Rockets”, 19th ESA Symposium on European Rocket and Balloon Programmesand and Related Research, Bad Reichenhall, Germany, (ESA SP-671). Frisk, E., 1996, “Model-Based Fault Diagnosis Applied To A Si-Engine”, Linköping University, Linköping, Sweden. Lucas, S. K., Kaya, C. Y., 2001, “Switching-Time Computation for Bang-Bang Control Laws”, Proceedings of the American Control Conference, Arlington, VA, USA. Nitulescu, M., 2005, “Controlling a Mobile Robot Along Planned Trajectories”, Control Engineering and Applied Informatics, Vol. 7, N. 2, p. 18-24.

O’Brien Jr., R.T., 2006, “Bang-Bang Control for Type-2 Systems”, Proceedings of the 38th Southeastern Symposium on System Theory, Tennessee Technological University, Cookeville, TN, USA. Ogata, K., 1982, “Modern Control Engineering”, Editora Prentice/Hall do Brasil Ltda, Rio de Janeiro, RJ, Brazil. Takahashi, Y., Rabins, M. J., Auslander, D. M., 1972, “Control and Dynamic Systems”, Addison-Wesley Publishing Company, Massachusetts, USA. Teixeira, A. J., 2009, “Rate Control System For Plant Parameter Uncertainties”, Proceedings of the 19th ESA Symposium on European Rocket and Balloon Programmes and Related Research, SP-671 pg. 255-260, Bad Reichenhall, Germany. Udrişte, C., 2008, “Multitime Controllability, Observability and Bang-Bang Principle”, Journal of Optimal Theory Applications, Vol. 139, p. 141-157. DOI 10.1007/s10957008-9430-2.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 259-268, Sep. - Dec., 2011

267

doi: 10.5028/jatm.2011. 03033311

José Irineu Sampaio de Oliveira*

Instituto de Aeronáutica e Espaço São José dos Campos/SP – Brazil josejiso@iae.cta.br

Márcio Yuji Nagamachi

Instituto de Aeronáutica e Espaço São José dos Campos/SP – Brazil marciomyn@iae.cta.br

Milton Faria Diniz

Instituto de Aeronáutica e Espaço São José dos Campos/SP – Brazil miltonmfd@iae.cta.br

Elizabeth da Costa Mattos

Instituto de Aeronáutica e Espaço São José dos Campos/SP – Brazil elizabethecm@iae.cta.br

Rita de Cássia Lazzarini Dutra

Instituto de Aeronáutica e Espaço São José dos Campos/SP – Brazil ritarcld@iae.cta.br *author for correspondence

Assessment of the synthesis routes conditions for obtaining ammonium dinitramide by the FT-IR Abstract: Over the last two decades, many routes have been proposed to synthesize ammonium dinitramide (ADN). However, most of them lie in routes in which reactants are too expensive for large-scale production. In this sense, the use of ordinary reactants is of paramount importance in this case. The aim in this synthesis consists on nitrating a starting reactant in a reaction known as nitration. Both the nitrating agent and the starting reactant should preferably be ordinary, narrowing the possibility of having realistic options for them. The most ordinary nitrating agent consists of a mixture of sulfuric and fuming nitric acids. Therefore, the breakthrough must come from the suitable choice of the starting reactant. However, so far, the only viable reaction relies on the use of sulfamate salts. Even though the process with this kind of salt has been largely commercially developed, only few information are available in the literature to properly address issues emerged from it. In this study, an attempt is made to enlighten some effects on the product caused by modifications in the route conditions. Characterization of the resulting products was confirmed by FT-IR in the region of MIR and NIR. The characteristic bands employed to identify ADN in the region of middle infrared were: 3129 and 1384 cm-1 (NH4+); 1537, 1344, 1209, and 1177 cm-1 (NO2); 1032, 954 cm-1 (N3); 828, 762 and 732 cm-1(NO2). The near infrared analysis pointed out few bands at 5185 and 4672 cm-1 in NH combination bands region. The resulting middle infrared spectrum was compared to the reference found in the literature for this product. The results show excellent agreement with the expected product. Keywords: propellant, oxidizer, energetic material, ADN synthesis, ammonium dinitramide, FT-IR, MIR, NIR.

INTRODUCTION Solid propellants are mainly composed of particulates of energetic materials. The study of these energetic materials is the way to optimize performance, burning behavior, stability, detonation properties, processing characteristics, and mainly their sensitivity (Teipel, 2005). Several compounds can take part in solid propellant composition, and although ammonium perchlorate (AP) is the most cited in the literature among the oxidizers, its downside, however, lies in the fact that it can release a large amount of chloride into the environment (Nagamachi et al., 2009). Received: 02/07/11 Accepted: 03/09/11

Based on the chemical structure, CL-20 or hexanitrohexaazaisowurtizitane (HNIW) and ammonium dinitramide (ADN) are potential components to be used in solid propellant formulation due to their special characteristics. Even though performance is the primarily criterion for the designer to define the convenience of any material, there are other practical considerations, including: availability and price; thermal and mechanical sensitivity (insensitive ammunitions – IM); processibility; compatibility; chemical and thermal stability; mechanical behavior temperature dependence; burning behavior; and pressure exponent (Teipel, 2005). It is also crucial to consider the oxidizing properties (including the oxygen positive balance) in the propellant development

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 269-278, Sep. - Dec., 2011

269

Oliveira J.I.S. et al.

process. ADN has positive oxygen balance of +25.8%, being a promising energetic material as a viable alternative to replace the ammonium perchlorate (AP) as an oxidizer in solid propellants. Its application can be found in civilian rockets and in smoke-low-emitting missiles (Teipel, 2005; Wingborg, 2010). Monopropellant for micro-propulsion systems for space applications are almost based on hydrazine. This compound, however, is highly toxic, volatile, and carcinogenic. Therefore, it demands high costs in safety measures at the whole process. Hence, over the years, there has been a considerable interest in Europe and in the USA in finding possible substitutes for hydrazine. ADN liquid monopropellants seem to be also a potential substitute to replace it whether for being easier to handle or for its specific impulse, 10% higher than for hydrazine (Wingborg, 2005). ADN was cited for the first time by professor Tartakowsky at the Annual Conference (ICT), in 1993 (Nagamachi et al., 2009). Simultaneously, in the USA, it was presented as a new oxidizer for rocket-solid propellants by Pak (Teipel, 2005). ADN has the chemical formula NH4N(NO2)2 (Squeme 1), which is of primary interest in solid-rocket propellants as a potential substitute for AP for keeping high efficiency in systems that already employ AP. The use of ADN in combination with an energetic binder can enhance the specific impulse (Isp) to reach values around 2,600 Ns/kg. If a metal fuel, such as Al or AlH3, is also incorporated to this formulation, the specific impulse can reach even higher, around 2,700 to 2,800 Ns/kg (Teipel, 2005).

_ O

N +

N _

_ O

NH4+ Squeme 1. Chemical structure of ADN (Shaw, 1993).

Among all the possible routes for this synthesis, in this paper, it was focused on the nitration of amidesulphonates. On one hand, the reaction can be carried out in the absence of organic solvents (Squeme 2). However, on the other hand, it must take place in a strong acid medium, which can also lead to 270

KO3SNH2+HNO3→KHSO4+NH4N(NO2)2+H2O Squeme 2. Reaction involved in the route of ADN synthesis, by nitration of amidesulphonates (Teipel, 2005).

the decomposition of the dinitramide. This synthesis is completed by recovering ADN from the reaction medium followed by purification. This process was developed by FOI and Nexplo and now it has been used for commercial production (Teipel, 2005). Several techniques are cited in the literature for characterization of ADN (Christe et al., 1996; Santhosh et al., 2002). Christe et al. (1996) have used Raman spectroscopy and infrared, in the middle infrared (MIR) region, to characterize ADN. The infrared (IR) bands attribution were: 1526 cm-1, attributed to the unsymmetrical stretching (υ a) in phase of group NO 2; 1344 cm-1 of symmetrical stretching (υ s) in phase of group NO2; and 1181 cm-1 (υ s out of phase of group NO 2). Santhosh et al. (2002) have used ultraviolet spectroscopy (UV), IR, and thermal methods and they have observed that the solution of ADN presents strong absorption with maximum at 284 nm under UV, and that IR spectra showed characteristic bands at 3136 cm-1 (υ N-H of NH4+), 1531 cm-1 (υa in phase of group NO2), and 1344 cm-1 (symmetrical stretching (υs) in phase of group NO2). In this paper, ammonium sulfamate and sulfamic acid were used as starting materials for obtaining ADN in different route conditions for these syntheses. The proportion of reactants and purification methods were tested, and the resulting samples were analyzed in the MIR region. For the synthesis in which ammonium sulfamate was used as a starting reactant, the samples were also analyzed by near infrared (NIR) spectroscopy. Thus, it was expected to cover a more ample zone in the spectral region of IR, which is meant for a suitable characterization of the resulting products. EXPERIMENTAL Routes of synthesis for obtaining ADN The routes applied in this study make use of sulfamic acid and ammonium sulfamate as starting reactants. The route conditions for these syntheses were adapted from Vörde and Skifs (2005) and Langlet, Ostmark and Wingborg (1997).

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 269-278, Sep. - Dec., 2011

Assessment of the synthesis routes conditions for obtaining ammonium dinitramide by the FT-IR

Synthesis route 1 (using sulfamic acid) ADN 40F/20 (40 mL HNO3 Fuming/20 mL H2SO4) Nitration For nitration of sulfamic acid, 40 mL of fuming nitric acid (96%) and 20 mL of sulfuric acid (95%) were mixed in a 500 mL 3-neck-round flask. The mixture was cooled down to -37ºC in a cryogenic bath (Ultra Termostato Criostato of Optherm, with controller HAAKE DC3). It was used 40% CaCl2 in water as the refrigerant fluid. Sulfamic acid was added in small portions of 0.5 to 1.0 g (total=5 g) under strong agitation (250 rpm) with a mechanical stirrer, the mixture was left to react for 1.5 hour at the temperature ranging from -30 to -35ºC.

and left inside the round flask. Acetone was added to promote dissolution of ADN. The round flask was manually shaken several times for 15 minutes. The content in the round flask was filtered with a paperfilter funnel, which was transferred to an erlenmeyer and newly cooled in cryostatic bath. White precipitated was formed again, which was finally analyzed by the IR. This cooling and filtering process was repeated for three times. Every filtered collected with acetone was placed in a warm bath and kept at nearly -33ºC under intense vacuum until the whole acetone was vaporized. The left liquid was a yellow solution highly concentrated. This solution was then transferred to a watch glass and left to vaporize at ambient temperature. Synthesis route 2 (using ammonium sulfamate)

Neutralization

Nitration

The reaction was conducted for 1.5 hour, then the whole mixture was transferred to a 1 L erlenmeyer, containing 85g small pieces of ice (deionized water) and equipped with magnetic stirrer. The erlenmeyer, was immersed and surrounded by pieces of ice inside a large flask. The aim of this procedure was to dilute the mixture before adding base, which can prevent it from reacting strongly with the acid mixture. Ammonium hydroxide was slowly dropped with a 200 mL burette, aiming at preventing the temperature inside the mixture from surpassing 0º C. The pH of solution was measured with litmus. Ammonium hydroxide was added until the solution became slightly alkaline. It was used 142.5mL of base for the neutralization.

Nitration has been conducted at temperatures ranging from -35 to -37º C in all the experiments. Concentration of acids, time of reaction, and mass of ammonium sulfamate have been varied, according to Table 1.

Purification The mixture provided from the neutralization step was transferred to a three-neck-round flask, and the liquid was left to vaporize in a bath at 35°C under vacuum (maximum capacity). After 80% of the liquid was vaporized, the round flask was removed and, with the aid of a spatula, the precipitated stuck on the flask was scratched out. This material was slightly ground

Neutralization The resulting mixtures from the nitration were diluted and cooled with 100g of ice, except ADN 40F/20 (40mL HNO3 Fum/20mL H2SO4) with “no dilution”, in which this step was eliminated. The mixtures were neutralized with 70 to 140mL of ammonium hydroxide. The neutralization was carried out by dropping NH4OH straight into the round flask, and immersed in a cryostatic bath at -35ºC. Purification For the synthesis of ADN 38/12 and ADN 30F/30, granular active coal (GAC) was used to extract the final product. It was previously boiled in acetone and it was dried inside an oven at 130°C for six hours. Adsorption of ADN was carried out by pouring the slightly alkaline mixture in a 500 mL florence flask with 20 or 11g of GAC, and shaken

Table 1. Data regarding the nitration – Syntheses routes 2.

Synthesis ADN 38/12 30 F/30 40 F/20 no dilution 35 F/35 F 40 F/20

NH2SO3NH4 (g) 17 5 5 5 5

HNO3 (%) 65 95 95 95 95

H2SO4 (%) 98 98 98 >100(SO3) 98

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 269-278, Sep. - Dec., 2011

Time (minutes) 60 90 30 90 90 271

Oliveira J.I.S. et al.

manually for 10 or 45 minutes, respectively in each case. The mixture was filtered in a Buchner funnel and washed with approximately 300 mL of deionized water (Santhosh et al., 2002). Desorption of ADN was carried out by recirculating acetone through a Soxhlet, in which GAC was covered by a paper-filter. Vacuum and temperature were controlled in order to keep it at 60° or 50° C, respectively, in each case. The rest of the material inside the round flak was analyzed by IR. For the synthesis of ADN 40F/20 with “no dilution”, ADN 35F/35F and ADN 40F/20, the neutralized solution was filtered and the precipitated was washed with acetone. The resulting liquid turned out yellow. A large amount of precipitated lying on the filter was discarded. Water formed during the reaction and acetone in the filtered were removed at 40º C under vacuum. The purification process was repeated for three times, and the same yellow product was obtained each time. This product was left on a watch glass inside a desiccator under vacuum, and then it was analyzed by IR. Analysis by IR Spectrum one spectophotometer of PerkinElmer had been used in this paper, in the region ranging from 4000 to

400 cm-1, resolution 4 cm-1 and gain 1. The samples were prepared according to the trasmission technique, with KBr pellets with concentrations around 0.5: 400 mg. For the NIR analysis, the samples were analyzed with the same equipment in the region ranging from 7800 to 4000 cm-1, under the same conditions. RESULTS AND DISCUSSION MIR analysis of the starting material used in the route 1 Squeme 3 shows the chemical structure of the starting material used in the referred route of synthesis O H 2N

O Squeme 3. Sulfamic acid structure formula (NH2SO3H).

Figure 1 presents the sulfamic acid MIR spectrum, whose bands are at the same wave numbers found in the spectrum from the reference in the literature

53.2 50 45 40 35

30 25 20 15 10 5 0,1 4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 cm

600 400

-1

Figure 1. Sulfamic acid MIR spectrum. 272

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 269-278, Sep. - Dec., 2011

Assessment of the synthesis routes conditions for obtaining ammonium dinitramide by the FT-IR

(SciFinder®, 2011). The characteristic bands are likely attributed (Silverstein and Webster, 2000; Urbanski et al., 1977; Smith, 1979) to their functional groups and they are situated at: 3151 cm-1 (NH); 2455 cm-1 (OH); and 1451, 1306, 1070 cm-1 (S=O). The absorption at 710 and 690 cm-1 can likely be associated with the group S-O. MIR analysis of the resulting product in the route 1 (using sulfamic acid): ADN 40F/20 (40 mL HNO3 Fuming/20 mL H2SO4) The resulting sample was named ADN 40F/20 (40 mL HNO3 Fuming/20 mL H2SO4) – sulfamic acid. Figure 2A shows the MIR bands of the sample. It has been observed more intense bands around 3420, 3130, 1710 and 1380 cm-1. According to Christe et al. (1996) and Santhosh et al. (2002), those bands are characteristics of ADN at: 3136 e 1380 cm-1 (NH4+); 1531, 1344, 1238 and

1181, 828 and 738 cm-1 (NO2); 1025 and 954 cm-1 (N3). However, in the MIR spectrum (Fig. 2A), only one band at 1380 cm-1 can be pointed out, which situates in the region characteristic of Group NH4+ of ADN. The MIR spectrum of the resulting sample is more similar to ammonium nitrate (Fig. 2B), indicating that, according to this route, this product is basically ammonium nitrate, which is likely obtained owing to the degradation during the purification process of the synthesis product. MIR analysis of the starting material of route 2 Squeme 4 shows the chemical structure of the starting material used in this route. Figure 3 shows the MIR spectrum of ammonium sulfamate, whose bands are at the same wave numbers found in the reference (SciFinder ®, 2011). The characteristic bands attributed to their functional groups are situated at: 3288 cm -1 (NH); 3195 and

4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800

600 400

cm-1

Figure 2. (A) MIR spectrum of the product obtained in synthesis 1 by using sulfamic acid: ADN 40F/20 (40 mL HNO3 Fuming/20 mL H2SO4); (B) MIR spectrum of ammonium nitrate – Lot n. 171169 – B. Herzog. J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 269-278, Sep. - Dec., 2011

273

Oliveira J.I.S. et al.

spectrum of the resulting product was more similar to ammonium nitrate, which presents large bands.

O H 2N

MIR analysis of the resulting product in route 2 (using ammonium sulfamate): ADN 30F/30 (30 mL HNO3 Fuming/30 mL H2SO4)

ONH4

O Squeme

4. Ammonium (NH2SO3NH4).

sulfamate

structural

formula

1400 cm -1 (NH 4+). The absorption ranges from 1260 to 1000 cm -1, it is likely associated with the groups S=O e S-O. MIR analysis of the resulting product in route 2 (using ammonium sulfamate): ADN 38/12 (38 mL HNO3/12 mL H2SO4) Like route 1, in route 2 any characteristic bands of ADN were found. Instead, only one band at 1380 cm -1, characteristic of the group NH 4+, was found. The

The resulting product presents bands at 3118 cm-1 (NH4+); 1377 cm-1 (NH4+); 1192 cm-1 (NO2); 1028 cm-1 (N3); 825 cm-1 (NO2) and 763 cm-1 (NO2), pointing out the presence of some characteristic absorptions of ADN, including contamination with ammonium nitrate, according to what was observed in the previously obtained products (Fig. 4). MIR analysis of the resulting product in route 2 (using ammonium sulfamate): ADN 40F/20 (40 mL HNO3 Fuming/20 mL H2SO4) “no dilution” The resulting product presents bands at 3121 cm-1 (NH4+); 1384 cm-1 (NH4+); 1177 cm-1 (NO2) and 825 cm-1 (NO2), indicating the presence of few bands related to ADN besides contamination with other products (Fig. 5).

70.5 65 60 55 50 45 40 %T 35 30 25 20 15 10 5.2 4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800

600 400

cm-1 Figure 3. MIR spectrum of ammonium sulfamate. 274

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 269-278, Sep. - Dec., 2011

Assessment of the synthesis routes conditions for obtaining ammonium dinitramide by the FT-IR

80.7

75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0.0 4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800

600 400

cm-1 Figure 4. MIR spectrum of ADN 30F/30.

75.0 70 65 60 55 50 45 40 %T 35 30 25 20 15 10 5 0.1 4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 cm-1

600 400

Figure 5. MIR spectrum of ADN 40F/20 “no dilution”. J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 269-278, Sep. - Dec., 2011

275

Oliveira J.I.S. et al.

MIR analysis of the resulting product in route 2 (using ammonium sulfamate): ADN 35F/35F (35 mL HNO3 Fuming/35 mL H2SO4 Fuming) The procedure applied for this synthesis did not provide any samples. MIR and NIR analysis of the resulting product in route 2 (using ammonium sulfamate): ADN 40F/20 (40 mL HNO3 Fuming/20 mL H2SO4) Characteristic bands employed in this route to identify ADN in the MIR region were: 3129 cm-1 (NH4+); 1384 cm-1 (NH4+); 1537 cm-1 (NO2); 1344 cm-1 (NO2); 1209 and 1177 cm-1 (NO2); 1032 cm-1 (N3); 954 cm-1 (N3); 828 cm-1 (NO2); 762 cm-1 (NO2); and 732 cm-1 (NO2). The MIR spectrum was compared to one from the literature (Christe et al., 1996) and shows good agreement (Fig. 6).

NIR analysis was conducted in this case since MIR characteristics bands of ADN were observed, which allows a more ample characterization of the resulting product (Fig. 7). Some number of bands at 5185 and 4672 cm-1 were seen. They were situated in the combination region of composed bands that present groups NH (Goddu, 1960). CONCLUSION Analysis conducted by FT-IR spectroscopy has proved a very effective way to characterize ADN, starting reactants, and sub-products. All characteristic bands for these compounds could be found, besides the resulting spectra, which also allow concluding that ADN 40F/20 (40 mL HNO3 Fuming/20 mL H2SO4) route is the most promising to synthesize ADN among the alternatives tested in this paper. The result was only concluded based on IR spectroscopy. However, more researches must be done in order to learn about purity, yield, and how to deal with the large amount of sub-products.

A 954 828 762 732 %T

1344

1032 1177 1430 1384 1209

1537

3129

827 952 761 1343 1531

3129 4000 3600 3200 2800 2400

2000

1800 cm-1

1426

1194

1600 1400

730 1180 1026

1200 1000 800 650

Figure 6. Comparison of IR (MIR) spectra. (A) ADN 40F/20 and (B) ADN reference.

276

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 269-278, Sep. - Dec., 2011

Assessment of the synthesis routes conditions for obtaining ammonium dinitramide by the FT-IR

56.6 54 50 5185

46 %T

42 38 34 30

26.5 78007600

4672 7200

6800

6400

6000 cm-1

5600

5200

4800

4400 4000

Figure 7. Partial NIR spectrum of ADN 40F/20.

REFERENCES Christe, K. O. et al., 1996, “The Dinitramide Anion, N(NO2)2-”, Phillips Laboratory, Propulsion Directorate, Edwards Air Force Base, California.

SciFinder®, American Chemical Society, 2011. Copyright© Bio-Rad Laboratories, Infrared spectral data from the BioRad/Sadtler IR Data Collection, Bio-Rad Laboratories, Philadelphia, PA (US). Spectrum nº BR087064.

Goddu, R., 1960, “Near-Infrared Spectrophotometry”, Advances Analytical Chemistry and Instrumentation, Vol. 1, p. 347-425.

Shaw, R. W., 1993, “Overviews of Recent Research on Energetic Materials – Advanced Series in Physical Chemistry”, Singapore: World Scientific Publishung Co.

Langlet, A., Ostmark, H., Wingborg, N., 1997, “Method of preparing dinitramidic acid and salts thereof”, US Patent 5976483, FOI, Sweden.

Silverstein, R. M., Webster, F.X., 2000, “Identificação Espectrométrica de Compostos Orgânicos”, Rio de Janeiro: Livros Técnicos e Científicos S.A.

Nagamachi, M. Y., Oliveira, J. I. S., Kawamoto, A. M., Dutra, R. C. L., 2009, “ADN – The new oxidizer around the corner for an environmentally friendly smokeless propellant”, Journal of Aerospace Technology and Management, São Paulo, Vol. 01, No. 2, p. 153-160.

Smith, A.L., 1979, “Applied infrared spectroscopy”, New York, John Wiley & Sons.

Santhosh, G. et. al., 2002, “Adsorption of ammonium dinitramide (ADN) from aqueous solutions: 1. Adsorption on powdered activated charcoal”, Journal of Hazardous Materials, India, p. 117-126.

Teipel, U., 2005, “Energetic Materials. Particle Processing and Characterization”, WILEY-VCH Verlag GmbH & Co., Weinheim. Urbanski, J. et al., 1977, “Handbook of Analysis of Synthetic Polymers and Plastics”, John Wiley & Sons, New York.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 269-278, Sep. - Dec., 2011

277

Oliveira J.I.S. et al.

Vörde, C., Skifs, H., 2005, “Method of producing salts of dinitramidic acid”, WO/2005/070823, Sweden. Wingborg, Niklas, et al., 2005, “Characterization and Ignition of ADN-Based Liquid Monopropellants”, Proceedings AIAA/ASME/SAE/ASEE Joint Propulsion, 41th Conference, Arizona.

278

Wingborg, N., et al., 2010, “Development of ADN-based Minimum Smoke Propellants”, Proceedings AIAA/ ASME/SAE/ASEE Joint Propulsion, 46th Conference, Nashville.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 269-278, Sep. - Dec., 2011

doi: 10.5028/jatm.2011. 03030511

Viviane Lilian Soethe*

Technological Institute of Aeronautics São José dos Campos/SP – Brazil vivianes@reitoria.ufsc.br

Evandro Luis Nohara

University of Taubaté Taubaté/SP – Brazil evandro.nohara@unitau.br

Luis César Fontana

University of Santa Catarina State Joinville/SC – Brazil fontana@udesc.br

Mirabel Cerqueira Rezende

Institute of Aeronautics and Space São José dos Campos/SP – Brazil mirabelmcr@iae.cta.br *author for correspondence

Radar absorbing materials based on titanium thin film obtained by sputtering technique Abstract: Titanium thin films with nanometer thicknesses were deposited on polyethylene terephthalate (PET) substrate using the triode magnetron sputtering technique. It was observed that the titanium thin film-polymeric substrate set attenuates the energy of the incident electromagnetic wave in the frequency range of 8 to 12 GHz. This result allows to consider this set as a radar absorbing material, which may be employed in automobile, telecommunication, aerospace, medical, and electroelectronic areas. Results of the reflectivity show that the attenuation depends on the thin film thickness, as a determining factor. Thin films with 25 to 100 nm thickness values show attenuation of the electromagnetic wave energy from around 20 to 50%. Analyses by Rutherford backscattering spectrometry provided information about the thickness of the thin films studied. Hall effect analyses contributed to better understand the influence of the thin film thickness on the electron mobility and consequently on absorption properties. Keywords: Radar absorbing material, Magnetron sputtering, Thin film, Titanium.

INTRODUCTION The technology involving thin film deposition on polymeric substrates is being more widely studied, due to its potential application in different areas. Innovations in industry and academic studies involving thin films uses in microelectronics, optics, solar cells, sensors, and special packing areas are cited (Bregar, 2004; Nie et al., 2007). Recent studies show that nanometer thin films deposited on appropriate substrates present physical characteristics different from the ones presented by the conventional films. For example, Kantal films with thickness from 10 to 200 nm, when used as coating in waveguide internal walls, present efficient performance as electromagnetic radiation absorber – more usually known as radar absorbing materials (RAM), due to the frequency range of application. These films, in the frequency range of 16.3 to 17.5 GHz, present absorption values around 0.8 dB (~17% of attenuation), depending on the coating thickness (Bhat, Datta and Suresh, 1998).

area, technological advances involving RAM depend on the development of materials to attend a wide range of frequencies, in other words, wavelengths varying from m to mm. In this context, RAM development makes more and more important for the control of electromagnetic wave propagation and its harmful effects on living beings and on the equipment. RAM are characterized by converting the energy of electromagnetic wave into thermal energy. Such materials are classified in two types, according to their interactions with the electromagnetic wave: materials with dielectric losses, which interact with the wave electric field, and materials with magnetic losses, which interact with the wave magnetic field. Conventional microwave absorbers have thickness values between mm to cm and weight/ area varying from 1.0 to 20 kg/m2 (Fortunato et al., 2002; Hashsish, 2002; Mikhailovsky, 1999; Nie et al., 2007).

RAM present innumerous applications, as in equipment electromagnetic shielding employed in automotive and aerospace industries and military technology, as well as in electrical and electronic devices and systems for wireless communication (Biscaro, Rezende and Faez, 2008; Bregar, 2004; Folgueras and Rezende, 2007; Hashsish, 2002; Nie et al., 2007; Rezende, Silva and Martin, 2000). Considering the complexity of the aeronautical engineering

The RAM studied in this paper is a metallic thin film with dielectric losses. In this case, when an external electric field is applied, several electric dipoles on the dielectric material (thin film) are formed. These dielectric dipoles are guided by the applied electric field. The interaction between the dipoles and the electric field leads to the formation of aligned dipoles, according to the applied electric field, enabling the material to store potential electric energy (Folgueras and Rezende, 2007).

Received: 17/02/11 Accepted: 25/08/11

Metallic thin films deposited on appropriate polymeric substrates present particular physical characteristics if

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 279-286, Sep. - Dec., 2011

279

Soethe, V.L. et al.

compared to their bulk materials. As mentioned in the literature, metals are excellent reflectors of microwaves, since they tend to keep null the electric field on their surfaces (Mayes, 2006). However, some metals and transition metals may behave as absorbers when reduced to nanometer thickness. As previously cited, Kantal films with thicknesses varying from 10 to 200 nm perform effectively as RAM when used as coating on waveguide internal walls (Bhat, Datta and Suresh, 1998). In comparison to conventional RAM, the nanometer films can present similar electromagnetic wave attenuation performance, but they are lighter. Therefore, thin films are able to interact with the electromagnetic wave, forming electric dipoles. In this particular case, the mechanism of absorption is based on the polarization of the metallic film and losses. Firstly, when the electromagnetic wave reaches the film it becomes polarized by the wave electric field and, consequently, electric current (Eddy currents) is produced due to the induced polarization. After that, the electromagnetic wave energy is changed into heat through the known Joule effect (Balanis, 1989; Nohara, 2003), due to the presence of defects in the crystalline structure of the nanofilm, which confers resistance to the electric current. Thus, the wave attenuation occurs when the thickness values of the metallic layer are smaller than, or at most similar to, the skin depth (δ) value of the metal (Bhat, Datta and Suresh, 1998; Ishii and Yasaka, 2004). The variation of this parameter (δ) depends on the characteristics of the metallic material used in the film production, mainly its electric conductivity, and also on the incident radiation wavelength that interacts with the film (Ohring, 1991). When the metallic layer thickness is adequate, the resulting electric current becomes confined into the film (Salmon, 1993) and losses occur (Bosman, Lau and Gilgenbach, 2003; 2004; Shubin, et al., 2000). The skin depth value can be estimated by Eq. 1, which correlates frequency (f), electric conductivity (σ) (inverse of electric resistivity) and the magnetic permeability of the vacuum (µ = 4π.10-7 H/m) (Kaiser, 2004; Serway, 1998). For thin films with thickness around nanometers, the electric resistance is very different from the bulk material. Thus, the skin depth estimation, related to the electric resistivity, is an important parameter to support the thin film preparation. 1 I" U f RX

where: 280

δ: is the skin depth; f : is the frequency; µ : is the magnetic permeability of the vacuum (µ = 4π.10-7 H/m), and σ: is the electric conductivity. Considering the new tendencies in RAM developments and the academic importance of this subject, the aim of this paper was to study the influence of the thickness and electric properties of titanium thin films on the electromagnetic radiation attenuation in the frequency range from 8 to 12 GHz. MATERIALS AND METHODS Titanium film deposition Titanium thin films were deposited on commercial polyethylene terephthalate (PET) substrates, with 0.1 mm of thickness, using the triode magnetron sputtering (TMS) technique, based on the TMS equipment presented by Fontana and Muzart (1998). The present process differs from the conventional magnetron sputtering, due to the presence of a screen parallel to the target, 2.0 cm from it. This screen is grounded and it improves the confinement of the plasma next to the target, thus providing greater stability and higher rate of ionization. This system makes it possible to work at low gas pressures (2.0 mTorr of argon). As a result, the atoms sputtered interact with the substrate without colliding with the gas atoms of the plasma atmosphere. Figure 1 shows a scheme of this system and Table 1 presents the experimental parameters used in the depositions. Argon

Plasma

Sputtered atoms

power Vacuum Target

(1)

Grid

Samples

Chamber of film deposition ! Figure 1. Scheme of the equipment used for the titanium film deposition (based on Fontana and Muzart, 1998).

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 279-286, Sep. - Dec., 2011

Radar absorbing materials based on titanium thin film obtained by sputtering technique

Table 1.

Experimental parameters used for the titanium thin film depositions.

Voltage Current Pressure Time of Thickness (V) (A) (mTorr) deposition (nm) (s) -320 2.0 2.0 5-90 25-420 Experimental characterization Reflectivity measurements were taken in order to obtain the attenuation values of the incident electromagnetic wave in the titanium thin film. These measurements were obtained in a vector network analyzer (8510, Hewlett-Packard, USA). The present experimental apparatus allows evaluating the attenuation value of the metallic thin film stored on the polymeric substrate in the frequency range of 8 to 12 GHz. The waveguide technique involves a device made with high mechanical precision, where the propagation of electromagnetic wave occurs in a closed system (Nicholson and Ross, 1970). This system basically consists of a waveguide with one terminal to generate the microwave signal and another one to collect the reflected signal that is conducted for spectral analysis (Nohara, 2003). Two different methodologies are used to measure the wave attenuation in this equipment. In the first, it is possible to measure the absorbed energy (Ea) considering the difference between incident (Ei) and transmitted energies (Et). Figure 2a shows this condition schematically. This setup is similar to the RAM characterization in free space condition and it provides information about the intrinsically energy absorbed. In the second methodology, a metal plate (Al plate) is located behind the thin film/substrate set. In this case, it is evaluated the performance of the sample in the electromagnetic energy attenuation, considering the thin film positioned on the metal plate.

Waveguide (metallic)

Surface analyses of the films were performed by the scanning electron microscopy technique (SEM) using an equipment from LEO, model 435 VPI, and thin film samples without special preparation. Rutherford backscattering spectroscopy (RBS) (Bubert et al., 2002) was also employed to complete the surface characterization of the films. For this, a beam of He+ with energy of 2.2 MeV was used. The results were analysed by the RUMP software (Doolittle, 1985). Based on the SEM and RBS analyses the morphology and the thickness of the films were evaluated, respectively. The thickness of the films was also evaluated by perfilometry technique, using an equipment Dektak 3030, and thin film samples were deposited on silicon plates, simultaneously to the thin film deposition on PET substrate. This methodology guarantees the same thickness for the deposited films on two different substrates. The silicon plate has an adhesive tape that is pull out from the silicon substrate before the perfilometry analysis, creating a step between the silicon substrate and the deposited film. This step is measured by a tip moving on the sample. The electric resistivity of the films was evaluated by a four-point resistivity methodology using an equipment Veeco, model FPP5000. The electronic mobility was measured by the Hall effect technique using a Hall system produced by Bio-Rad Company, model HL5500.

(b)

(a)

The Ea is obtained by the difference between Ei and the reflected energy (Er). Figure 2b shows this apparatus schematically. This process simulates the absorber material put on a reflected surface, representing a real situation, for example, the RAM application to aeronautical fuselage. Both methods, schematically presented in Fig. 2, were used to characterize the prepared films.

Ea Et

sample Metallic thin film

Waveguide (metallic)

Tottaly refletive plate

sample Metallic thin film

Figure 2. Waveguide settings used for the reflectivity measurements of the titanium films: (a) evaluation of the intrinsic absorption; (b) reflectivity with metallic plate (adapted from Nohara, 2003). Ea: absorbed energy; Er: reflected energy; Ei: incident energy; Et: transmitted energy.

J. Aerosp.Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.3, No.3, pp. 279-286, Sep. - Dec., 2011

281

Soethe, V.L. et al.

RESULTS AND DISCUSSION The perfilometry measurements show that the titanium thin films were deposited at 3.6 nm/s on both polymeric and silicon substrates. This parameter, when determined by RBS analyses, shows values close to 5.0 nm/s. Besides the differences observed, these values present the same magnitude. However, considering that the RBS is a technique more precise, the thin film thickness values were calculated based on the deposition rate of 5.0 nm/s. The RBS analyses also show that the thin films are homogeneous, that is, without impurities. These results are attributed

to the adequate parameters used in the deposition process, how high the deposition rate and low pressure (vacuum) were during the process. SEM analyses of titanium thin films with 25, 45 and 150 nm of thickness show similar morphologic aspects for the analysed surfaces (Fig. 3). In the used magnification, Fig. 3 shows surfaces with large uniformity, continuous structure, and absence of imperfections. Figure 4 shows the average of the reflectivity measurements in the frequency range of 8 to 12 GHz

Figure 3. Scanning electron microscopy technique of titanium thin film with different thickness values: (a) 25 nm; (b) 45 nm, and (c) 150 nm.

282

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 279-286, Sep. - Dec., 2011

Radar absorbing materials based on titanium thin film obtained by sputtering technique

Attenuation (%)

Ti thin films deposited by TMS

crystalline structure of the nanofilm as the thickness decreases, which favors the losses in the metallic film, as cited in the literature (Bosman, Lau and Gilgenbach, 2003; 2004; Shubin et al., 2000). y=a.xb

(2)

where: y: represents attenuation (%); x: represents thickness (nm) and a and b: are constants (240.9 and -0.62, respectively).

100

200

300

400

500

Thickness (nm) Figure 4. Relationship between microwave attenuation and titanium thin film thickness.

for the titanium thin films with different thicknesses. These results were obtained by using the waveguide technique, according to schedule in Fig. 2b, which shows that the slender film (25 nm) presents attenuation values near 45% in the frequency range analysed and thicker films (>100 nm) behave as less efficient microwave absorbers. The curve observed in Fig. 4 can be represented by a polynomial function, as shown in Eq. 2, where a and b are constants, equals to 240.9 and -0.62, respectively. Based on such equation, it is possible to estimate that titanium films with thickness values near 4.0 nm attenuate nearby 100% of the incident electromagnetic wave. This behavior is explained by the increase of defects in the Table 2.

From the electric resistivity measurements of the titanium thin films by using the four-point resistivity technique and the thickness of the films, the skin depth parameters were calculated according to Eq. 1 for 8 and 12 GHz. The skin depth calculated, the electric resistivity (ρ), and the superficial resistance (R) are presented in Table 2. Table 2 shows that the electric resistivity decreases as the thickness increases. When correlating these results with the curve in Fig. 3, it is verified that larger attenuation of electromagnetic wave energy is associated with the thinner films. This behavior is explained by the thickness increase to be related to the lower electric resistivity, which diminishes the losses and consequently the attenuation. Thus, films with smaller thickness provide more effective microwave absorbers, in agreement with Machlin (1998). This behavior is better observed in Table 3, which presents the thickness in function of the average absorption. Table 2 also shows that the thickness of the films is much smaller than the calculated skin depth.

Skin depth values of titanium thin films with different thicknesses.

Time of deposition (s)

Thickness x 109 (m)

Surface resistance (Ω)

ρ x 106 (Ω.m)

δ x 105 (m) (f=8 GHz)

δ x 105 (m) (f=12 GHz)

5±1

25±5

117

132

6.47

5.28

10±1

50±5

28.0

6.34

1.41

1.15

15±1

75±5

19.7

6.69

1.45

1.18

20±1

100±5

6.51

2.53

0.89

0.73

Table 3.

Microwave absorption and thickness of titanium thin films.

Thickness (nm) 25±5

Average absorption (%) 44.3

Table 4.

Electronic mobility of titanium thin films with different thicknesses.

Thickness (nm) 25±5

Electronic mobility (µe) (cm2/V-s) 0.708

50±5

33.0

50±5

1.11

75±5

40.1

75±5

6.94

100±5

17.8

100±5

17.3

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 279-286, Sep. - Dec., 2011

283

Soethe, V.L. et al.

These results show that titanium thin films behave more efficiently as microwave absorbers when there is thickness around 25 nm. This behavior is associated with the Eddy current formation on the film surface and it is also the lowest electronic mobility that favors the losses. Figure 5 shows the reflectivity measurements of the titanium thin films with different thicknesses, obtained in different deposition times. The curves showed in Fig. 5a were obtained by using the setup depicted in Fig. 2a, while Fig. 4b presents the curves obtained in the setup showed in Fig. 2b. Figure 5a shows the behavior of thin film with 25 nm, where absorption values between 40 to 50% in all frequency range (8-12 GHz) are observed. Samples with larger thicknesses present lower absorption values, but nearly constant in all frequency range evaluated. Curves of samples prepared with thicknesses larger than 150 nm are not presented because they show similar behaviors, with absorption values between 5 and 10%. Figure 5b presents the reflectivity mesurements using an aluminum plate (100% reflector) under the thin films

(a) 70

Ti films (pure) Analysis without metallic plate 25 nm 50 nm 150 nm

Attenuation (%)

60 50 40 20

(setup in Fig. 2b). These curves show the maximum absorption value and the frequency where this event occurs. These data confirm the dependence of microwave attenuation with the thin film thickness, and they also present a slight variation of the maximum attenuation with the frequency (8.6-9.0 GHz), according to the thin film thickness. CONCLUSION The TMS proved itself as a suitable technique to obtain titanium thin films, which perform efficiently as RAM. Reflectivity results from the waveguide technique (8-12 GHz) of the processed metallic thin films, with thickness range from 25 to 100 nm, show microwave attenuation values around 50% in broadband (8-12 GHz). These attenuation results are attributed to the electric resistivity and electronic mobility of the titanium thin films. The reflectivity results show that the maximum attenuation value (around 50%) occurs for thinner films (25 nm), which also present lower resistivity and electronic mobility that favor the losses. ACKNOWLEDGMENTS The authors acknowledge the financial support of Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), process no. 05-01258-05; Financiadora de Estudos e Projetos (FINEP), process no. 1757-03; and National Counsel of Technological and Scientific Development (CNPq), processes no. 305478/2009-5 and 311396/2006-2; and also UDESC for providing the TMS equipment; Laboratório de Materiais e Feixes Iônicos (LAMFI) for the RBS analyses; Laboratório de Sistemas Integráveis (LSI)

(b)

Ti films (pure) Analysis with metallic plate

25 nm 50 nm 150 nm

15 Attenuation (%)

The electric behavior of the titanium thin films was also evaluated by Hall effect technique. Table 4 shows the electronic mobility of the studied films and it is observed that this property increases as the thickness increases, meaning that the electron mobility is favored in thicker films. This behavior suggests that thicker films begin to show the characteristics of the bulk metal, i.e., a reflector material. For thinner films, the electron mobility is diminished favoring the losses as heat by Joule effect, as mentioned by Nohara (2003).

20 8

10 11 Frequency (GHz)

Figure 5. Attenuation curves of titanium thin films deposited at different times: (a) without metal plate (intrinsic absorption); (b) with metal plate (reflectivity measurements).

284

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 279-286, Sep. - Dec., 2011

Radar absorbing materials based on titanium thin film obtained by sputtering technique

from USP/SP for the electric resistivity and perfilometry analyses; and Laboratório de Semicondutores (LabSem) of PUC/RJ for the Hall effect analyses. References Balanis, C.A., 1989, “Antenna Theory: analysis and design”. John Wiley Sons, New York, USA. Bhat, K.S., Datta, S.K. and Suresh, C., 1998, “Electrical and microwave characterization of kanthal thin films: temperature and size effect”, Thin Solid Films, Vol. 332, No. 1-2, pp. 220-224. doi:10.1016/S00406090(98)01103-1. Biscaro, R.S., Rezende, M.C. and Faez, R., 2008, “Influence of doped polyaniline on the interaction of Pu/PAni blends and on its microwave absorption properties”, Polymers for Advanced Technologies, Vol. 19, No. 2, pp.151-158. doi: 10.1002/pat.990. Bosman, H., Lau, Y.Y., Gilgenbach, R.M., 2003, “Microwave Absorption in a Thin Film”, Applied Physics Letters, Vol. 82, No. 9, p.1353. doi:10.1063/1.1556969. Bosman, H., Lau, Y.Y., Gilgenbach, R.M., 2004, “Power absorption by thin films on microwave windows”, IEEE Transactions on Plasma Science, Vol. 32, No. 3, pp. 1292-1297. doi:10.1109/TPS.2004.827579. Bregar, V.B., 2004, “Advantages of Ferromagnetic Nanoparticle Composites in Microwave Absorbers”, IEEE Transactions on Magnetics, Vol. 40, No. 3, pp. 1679-1684. doi: 10.1109/TMAG.2004.826622. Bubert, H., Jenett, H., 2002, “Surface and Thin film Analysis”, Institute of Spectrochemistry and Applied Spectroscopy (ISAS), Wiley-VCH, Germany, 336p. Doolittle, L.R., 1985, “Nuclear Instrument Method – Rump simulation code”, B9, pp. 344-351. Folgueras, L.C., Rezende, M.C., 2007, “Hybrid multilayer structures for use as microwave absorbing material” Proceedings of the SBMO/IEEE MTT-S International Microwave & Optoelectronics Conference, Brazil, pp. 483-487. Fontana, L.C., Muzart, J.L.R., 1998, “Characteristics of triode magnetron sputtering: the morphology of deposited titanium films”, Surface and Coatings Technology, Vol. 107, No. 1, pp. 24-30. doi:10.1016/ S0257-8972(98)00576-3.

Fortunato, E., Nunes, P., Costa, P.D., Brida, D., Ferreira, I. and Martins, R., 2002, “Characterization of aluminium doped zinc oxide thin films deposited on polymeric substrates”, Vacuum, Vol. 64, pp. 233-236. doi:10.1016/S0042-207X(01)00319-0. Hashsish, E.A., 2002, “Design of wideband thin layer planar absorber”, Journal of Electromagnetic Waves and Applications, Vol. 16, No. 2, pp. 227-241. doi: 10.1109/APS.2011.5997137. Ishii, N., Yasaka, Y., 2004, U.S.Patent No 6823816. Available at: www.patents.com/us-6823816.html. Kaiser, K.L., 2004, “Electromagnetic Compatibility Handbook”, CRC Press, Boca Raton, USA. Machlin, E.S., 1998, “Materials Science in Microelectronics”, Elsevier, 2a ed., New York, Vol. 2, pp. 1-70. Mayes, E., 2006, U.S. Patent No 6986942. Available at: www.patents.com/us-6986942.html. Mikhailovsky, L.K., 1999, “Solution of actual problems of electromagnetic compatibility by means of Spin (NonCurrent) Electronics and Non-phase Electrodynamics”, Proceedings of the VIII International Conference on Spin Electronics – Section of International Conference on Gyromagnetic Electronics and Electrodynamics, Vol. 1316, Moscow Region, Fisanovka, Rússia, pp. 327-349. Nicolson, A. M., Ross, G. F., 1970, “Measurement of the Intrinsic Properties of Materials by Time Domain Techniques”, Instrumentantion and Measurement, Vol. 19, pp.377-382. doi: 10.1109/TIM.1970.4313932 . Nie, Y., et al., 2007, “The electromagnetic characteristics and design of mechanically alloyed Fe-Co particles for electromagnetic-wave absorber”, Journal of Magnetism and magnetic materials, Vol. 310, pp.13-16. doi:10.1016/j.jmmm.2006.07.021 Nohara, E.L., 2003, “Materiais Absorvedores de Radiação Eletromagnética (8-12 GHz) Obtidos pela Combinação de Compósitos Avançados Dielétricos e Revestimentos Magnéticos”. Ph.D. Thesis, Technological Institute of Aeronautics, São José dos Campos, S.P., Brazil. Ohring, M., 1991, “The Materials Science of Thin Films”, Stevens Institute of Technology, Departament of Materials Science and Enginnering, Hoboken, New Jersey, Academic Press, San Diego, pp. 531.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 279-286, Sep. - Dec., 2011

285

Soethe, V.L. et al.

Rezende, M.C., Silva, F.S. and Martin, I.M., 2000, “Materiais absorvedores de radiação eletromagnética”, Spectrum, Vol. 2, pp. 17-20. Salmon, L.G., 1993, “Evaluation of thin film MCM materials for high-speed applications”, IEEE Transactions on Components, Hybrids, and Manufacturing Technology, Vol. 16, No. 4, pp. 388-391. doi: 10.1109/33.237934.

286

Serway, R.A., 1998, “Principles of Physics”, Saunders College Texas, Fort Worth, London. Shubin, V.A., et al., 2000, “Local electric and magnetic fields in semicontinuous metal films: beyond the quasistatic approximation”, Physical Review B, Vol. 62, No. 16, pp. 11230-11244. doi: 10.1103/ PhysRevB.62.11230.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 279-286, Sep. - Dec., 2011

doi: 10.5028/jatm.2011. 03034111

Marcelo Bender Perotoni*

Technological Institute of Aeronautics São José dos Campos/SP – Brazi marcelo.perotoni@ufabc.edu.br

Luiz Alberto Andrade

Institute of Aeronautics and Space São José dos Campos/SP – Brazil andradelaa@iae.cta.br *author for correspondence

Numerical evaluation of an airto-air missile radar cross section signature at X-band Abstract: The remote detection of a vehicle requires that some kinds of its emissions are tracked and detected. Usually, electromagnetic emissions are used in the form of radar (electromagnetic waves in the range of radiofrequency and microwaves). Different types of antennas are used as sensors, tailored to the signal frequency band and its polarization, as well as to the target distance (higher gain antennas used for low amplitude signals). For the specific case of radars, the use of computational methods to address the electromagnetic signature (spatial pattern of the scattered energy from the object) has become widespread, given the high costs and complex equipment associated with these respective measurements. Therefore, the use of computer simulation is ideally suited for creating a realistic database of targets and its respective signatures. The same computer-created signatures database can also be used for the thermal range, enabling a complete technology solution for the signature and design of stealth vehicles, with reduced emissions. Keywords: Radar cross section, Electromagnetic scattering, Computational modeling.

INTRODUCTION Radars have become a fundamental tool in the areas of defense and homeland security, since the Second World War (Grant, 2010). Since its inception, several new tools were added, namely ultra wide band (UWB), processing algorithms, digital signal processors, and so on (Kouemou, 2009; Skolnik, 1981). Although modern techniques rely heavily on signal processing (software) for increasing the detection capabilities, the hardware is still a crucial issue. The reason is that the frequencies used by radars normally require the use of microwave instrumentation, which presents higher costs and complexity when compared to low-frequency circuitry and equipment. The spatial waveform shape of the return radar signal (echo) is the vehicle radar cross section (RCS) signature. Every object has a specific signature, which helps to identify what kind of structure is under analysis (for instance, determining whether it is a friend or foe). Figure 1 depicts the radar signature of a generic glider at 0.5 GHz, for a frontal incidence, taken from the examples supplied with the CST Microwave Studio® (CST, 2010) package. The electromagnetic energy from the incoming wave develops currents along the metallic surface, which then re-radiate in several directions. That is the field which hits the receiver radar, located on the ground

Received: 13/09/11 Accepted: 08/10/11

90 60

120

150

180

210

0 30

330

300

240 270

Figure 1. Simulated result for a bistatic RCS of a glider (length 14.6 m, wingspan 13.75 m). The incoming wave is represented by the arrow (500 MHz). The scale is shown in dBsm (CST, 2010).

or embarked somewhere. From Fig. 1, it is possible to see that some directions have a higher energy density than others. The receiver signal can be captured and processed on the same position where it was transmitted; this is the so-called monostatic radar signature. Signatures are called bistatic when the receivers can be spread in other directions. In

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 287-294, Sep. - Dec., 2011

287

Perotoni M.B., Andrade L.A.

the case of Fig. 1, a bistatic scenario simulation is shown. If a receiver was placed on the same position of the transmitter (180º), it would be seen a target with 1.6 dBsm (or 1.45 m2). The unit used is that of a surface (square meters), so that it has a relation to an analogous physical area, which scatters the same energy. On the other hand, if the receiver is positioned at the 0 angle, the target will be detected with a RCS of around 25 dBsm (316 m2). The measurement of the RCS signatures from real targets (aircraft, tanks, vessels) is a complex and costly task. For the case of an aircraft, it requires its placement on an adequate area, which is normally wide (comparable to the aircraft size). In addition, the microwave instrumentation has to be able to illuminate the object with enough energy in order that the returned signal can be discriminated against the environment noise floor. On top of that, measurements done on the ground do not represent a true environment, since during real flights there is no ground plane. In view of these complexities, computer simulations have been used to predict and analyze radar signatures. For instance, the design of stealth vehicles (i.e. vehicles whose RCS signatures are very low when compared to their physical size) relied on the computer analysis to get a geometric shape able to scatter the incoming wave, in such a way the receiver signal is as small as possible (Grant, 2010).

Any metallic object illuminated by an incident electromagnetic wave develops along its surface electric currents, which in turn re-radiate. The unknown to be determined is the current density J(r), which is found as the solution of an integral equation. It is written as a matrix equation, after the MoM discretization (Davidson, 2005), in which MoM stands for method of moments. The solution is achieved in an iterative approach, by methods such as conjugate gradient, which uses approximately N2 operations per iterations, with N equals to the number of unknowns. The problem of a metallic object subjected to an incident electric field Ei(t) is represented by the electric field integral equation (EFIE) (Davidson, 2005): t µsG(r , r’) J(r’)dS’ = 4Ui tEi(r) kM

(1)

where, t represents an unit tangent vector on the surface S; k the wave number; J(r’) the current density unknown; η the medium impedance; and i the imaginary term.

This article presents a short overview of the numerical methods used in the microwave analysis. Monostatic and bistatic simulated signatures of a real short range, air-to-air missile are presented. Comparisons with measurements are also shown.

The primed r variable regards the source variable and the unprimed r is the observation point variable. G is the Green Function representing the problem, given by Eq. 2 (Davidson, 2005):

METHODS

eik|r - r’| G(r , r’) = (1- 12 r) |r - r’| k

Numerical analysis

(2)

To solve the microwave range scattering problems, an appropriate solution of the Maxwell Equations is sought, subjected to the particular boundary conditions. The numerical solution of those equations involves a previous step, the discretization, where the object and its surrounding volume are sliced into small elements (forming the electromagnetic mesh). Then, the Maxwell equations are applied to each of those small elements, whose fields/currents/voltages are determined.

Analogously, a magnetic field integral equation (MFIE) can also be written. When there are resonances due to enclosed cavities, numerical problems arise. The solution is the use of a combination of both MFIE and EFIE, named Combined Field Integral Equation. That is the reason to avoid hollow structures; the missile or aircraft model is better simulated as made of a solid piece of metal, without hollow parts in their interiors. These equations can be numerically solved by the MoM, which basically finds the solution by a matrix inversion.

Volume meshes are commonly used when the object is electrically small, like most antennas. However, the computation of large-scale models using volume mesh methods becomes intractable with even moderate hardware. The reason is that the meshing of the hollow part of a missile and the air area around it can be neglected, since the external shell is the main responsible for the scattering. For that kind of application, surface mesh is used instead; only the external 2D surface (sheet) is meshed.

The Multi-Level Fast Multipole Method (MLFMM) (Song et al., 1997) is used to further reduce the problem complexity, by making the MoM matrix sparse. It is achieved by the reduction of the coupling to only nearby elements through the use of small cubic volumes. Then, the problem has its number of operations reduced to Nlog(N), enabling the computation of large scale problems. Figure 2 illustrates the idea by showing the coupling scheme difference between the MoM and the MLFMM.

288

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 287-294, Sep. - Dec., 2011

Numerical evaluation of an air-to-air missile radar cross section signature at X-band

Figure 2. The left picture shows that in the MoM all elements are allowed to couple to each other, generating a very dense matrix. The MLFMM, on the right, allows only certain elements to couple, resulting in a sparse matrix.

It is interesting to stress that some methods, like the Finite-Difference-Time-Domain (FDTD), rely on mathematical operations that fundamentally are simple, requiring subtractions and sums, but at expenses of large and constant memory accesses. On the other hand, MoM and MLFMM require the inversion of a large matrix, mathematically and computationally much more processor (CPU) intensive (Munteanu, Timm and Weiland, 2010). As the frequency increases even further, turning the electrical size of the problem too large, then the alternative is the use of asymptotic methods (Geometrical Optics – GO). A ray tracing scheme computes the incoming and reflected rays (Shooting and Bouncing Rays – SBR). Evidently, there is a tradeoff between the precision and the computational effort between the two approaches, MoM/MLFMM, and asymptotic. It can be stated that, as a general rule, the accuracy increases as we move from Asymptotic to MLFMM and to MoM, whereas the simulation speed that requires the random-access memory (RAM) decreases. The SBR method launches a dense grid of finite rays that hit the object, and later the multiple reflections are computed, taking into consideration the geometry of the problem. This method (Pike and Sabatier, 2002) extends the Physical Optics (PO) by defining the surface currents developed on the structure in terms of the computed fields. Therefore, it is also possible to map the surface current density, which is important for identifying hot spots in the aircraft. The solver (numerical method used) was the I-solver (Integral Equation), based on the MoM/MLFMM method. The other solver that was employed was the A-solver (Asymptotic), based on the Shooting and Bouncing Rays method.

RESULTS AND DISCUSSION Bistatic simulations The missile here analyzed is named Piranha, which was developed by a joint program between the Brazilian Air Force and the Navy. It is a short range, air-to-air unit, with an infrared seeker (Coelho, 2007). Figure 3 shows the model and its main dimensions. For 10 GHz, its electrical size is 95 λ long and 22 λ wide.

0.66 m 2.85 m

Figure 3. Picture showing the missile with its main dimensions.

The 3D complete missile model was imported from a mechanical computer aided design file (CAD) (Catia, 1998) into the workspace of the electromagnetic solver. The missile material is considered as being made out of a perfect electric conductor (without losses). The boundary conditions are set to open space. Figure 4 depicts the surface mesh obtained along the missile surface. A good quality mesh (i.e. with homogeneous elements, showing good aspect ratio and with similar sizes) helps getting a better and faster simulation. The aspect ratio plays for the surface mesh a vital role, meaning that the ratio between the biggest and smallest component of the structure directly impacts on the mesh quality. The ideal situation is when the aspect ratio is close to one (largest and smallest dimensions are similar).

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 287-294, Sep. - Dec., 2011

289

Perotoni M.B., Andrade L.A.

the mechanical model contains a small detail, which is approximately smaller than one tenth of the wavelength (relative to the incoming plane wave), it can be removed without gross loss of accuracy. The difficulties associated with the mesh around a small screw or bolt (smaller than one tenth of wavelength), for instance, do not pay off in terms of final accuracy. It is therefore simpler if it is eliminated. The first study regards the bistatic RCS response to a X-band, 10 GHz signal (frequency where most onboard radars operate). For that, the incoming plane wave is assumed to be incident right on the frontal side of the missile (90º in the Fig. 5), with the electric field aligned to the missile longitudinal axis, as Fig. 5 suggests. The results of the bistatic RCS is also shown in Fig. 5. It can be seen that the frontal RCS has a value of -9.6 dBSm. The RCS considers the absolute power received with the co and cross polarizations included. It means that the absolute power involves both vertical and horizontal polarizations – actually the square root of both squared components. A real world measurement will have to count on antennas able to receive both polarizations. The simulation used the MoM/MLFMM Method, with first order elements. They are triangles with straight sides, and they enable a faster simulation in comparison to higher order elements.

Figure 4. Surface mesh for the object. Zoomed in is a detailed area near the stabilizers.

Surface cells are computationally more demanding than hexahedral cells (used by time domain/finite-difference time-domain (FDTD) or transmission line method (TLM) solvers) (Munteanu, Timm and Weiland, 2010). It means that with hexahedral cells the structure is represented by the use of small bricks (the mesh cells), whereas for the surface mesh, small planar triangles have to cover the structure surface, adjusted to complex details so that they accurately represent fine details. Figure 4 shows the surface mesh for the object. It can be seen that the mesh elements are homogeneous (they have similar sizes), even in the region of the stabilizers. It greatly improves the convergence of the problem.

Another point of interest is the lateral incidence. Figure 6 shows the excitation (the electric field here is orthogonal to the missile axis), and the respective result. For this situation, the lateral RCS has a value of 24 dBsm.

Frequently, the mechanical model needs refinements in order to enable a functioning surface mesh. Refinements here usually refers to simplifications. For instance, if

90 60

120

150

180

-25

-10

330

210 300

240 270

Figure 5. Computed RCS, units (dBsm). The scenario involves a bistatic response to a frontal excitation (90º in the figure). 290

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 287-294, Sep. - Dec., 2011

Numerical evaluation of an air-to-air missile radar cross section signature at X-band

90 60

120

150

0 18.91 29.46 40

180

330

210 300

240 270

Figure 6. Computed RCS, units (dBsm). The scenario involves a bistatic response to a lateral excitation (0Âş in the figure).

The results presented in Figs. 5 and 6 show that the scattered power is higher to the situation where the missile is illuminated laterally (-9.6 for frontal versus 24 dBsm for the lateral case). It is intuitive to see that the physical area that intersects the incoming wave is larger for the lateral case, justifying the difference. A lateral illumination of an incoming missile is however preferred for an earlier incoming missile detection. For the case of an onboard radar which detected an incoming missile, few seconds are left for the detection and the evasion maneuver. Thus, few decibels of difference in the received signal (related to the RCS parameter dBsm) can enlarge the period between the

detection and evasion, increasing the survivability rate of the plane under attack. Another result of the simulation is the identification of the hot spots, namely the particular points on the surface that concentrate the higher currents when illuminated by the plane wave. These currents are responsible for scattering the energy back to the source. Therefore, if the goal is to minimize the RCS towards a stealth vehicle design, those hot spots need to be identified and eliminated. The alternative to eliminate or diminish the current density on hot spots is by means of a geometric reshape or by using Radiation

A/m

0.01 0.00838 0.0071 0.006 0.00504 0.00422 0.0035 0.00289 0.00236 0.0019 0.0015 0.00116 0.000866 0.000611 0.000391 0.000201 0

Figure 7. Current distribution caused by a frontal (left) and lateral (right) incidence. The red areas are those where higher amplitudes are developed due to the incoming plane wave. Since the lateral incidence has a higher overall RCS than the frontal case, it also develops currents with larger amplitudes. J. Aerosp.Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.3, No.3, pp. 287-294, Sep. - Dec., 2011

291

Perotoni M.B., Andrade L.A.

the missile (180º), in 90 steps. For each angle, the RCS is computed in that single direction only. The results also present a comparison between the MoM/MLFMM and the asymptotic solver.

Absorbing Materials (RAM) (Grant, 2010). In Peixoto et al. (2011), measurements are presented, which show that by covering a missile with RAM, the overall RCS is lowered. Figure 7 depicts the hot spots (visualized as red areas) for the frontal (Fig. 5) and lateral (Fig. 6) incidences. The incident plane wave has amplitude of 1 V/m. Since the frontal incidence has an overall lower RCS (the intersection area is much smaller than that of the lateral case), the developed currents are also of lower amplitude.

A comparison between the results of both techniques is shown in Table 1 alongside with some experimental results (Peixoto et al., 2011). It is worth mentioning that the measurement setup showed a dynamic range limitation, i.e. too high-noise floor (Peixoto et al., 2011). Therefore, only the higher energy peaks of RCS were detected, like for the angles of 180 and 90 in Table 1. It justifies the differences seen for other angles, like 0. Another difference relies on the fact that the warhead is not metallic (it contains the infrared seeker and other systems, so it needs to be transparent), whereas the computer model is completely metallic. It imposes a severe difference especially for the 0º incidence. Other significant difference is the fact that the measurement was done in an outdoor facility, with the presence of the ground, which for 10 GHz

Monostatic simulations A monostatic scenario involves the rotating of the transmitter around the target, i.e., the transmitter and the receiver are located in the same point. Computationally, it is a more challenging task, since for every position the electromagnetic environment is different, generating a different system matrix, too. Figure 8 shows that the incident wave is swept from the frontal direction (equivalent to 0º) to the rear side of

90 60

120

AsymptoticSolver MoM-MLFMM

150

180

-60

-40

-20

0 20

Figure 8. Illustration of the monostatic range of simulation and results for both MoM/MLFMM and asymptotic solvers. The asymptotic solver used was the A-solver in CST Microwave Studio® (CST, 2010), using medium precision.

292

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 287-294, Sep. - Dec., 2011

Numerical evaluation of an air-to-air missile radar cross section signature at X-band

Table 1.

Comparison between the performances with two different solvers and measurements.

Solvers/ RCS (dBsm) Angle

MLFMM Asymptotic

Experimental (Peixoto et al., 2011) -22.0

Table 2.

Comparison between the performances of the two different solvers.

Comparison

Asymptotic

MLFMM

Simulation time

4 minutes

9 hours

Peak RAM used

48 Mbytes

2.8 GBytes

0º

-9.62

-3.73

30º

-15.7

-23.5

CONCLUSIONS

60º

-8.22

-14.2

90º

18.4

17.3

120º

-8.14

-22.7

150º

-14.6

-16.0

180º

-3.59

1.59

A study concerning the X-band RCS signature of a real air-to-air missile is presented. Two numerical techniques to perform the computer simulation are shown, alongside with results compared to measurements performed in an outdoor facility. The results showed a reasonable similarity with measurements, considering that the real world measurement setup and the missile were not completely similar to the virtual representation. Since RCS measurements require a complex and sophisticated setup, which is not always available, prediction techniques based on simulation can be implemented in order to complement the real world measurements. The requirements for computer prediction are the software package and a moderately equipped hardware, alongside with mechanical models of the objects that are free from unnecessary details, but they are also accurate in terms of dimensions and shapes.

17.0

0.00

may behave like a kind of ground plane, whereas the simulation was done in a perfect non-reflective environment (Peixoto et al., 2011). Comparisons with measurements involve a difficult task, regarding an accurate object model (not easily available with correct materials and geometrical details), as well as a correct representation of the measurement setup. Therefore, computer simulations should be faced as a complement to measurements, enabling a somewhat easier and less costly alternative. The faster simulation of the asymptotic solver results in some angles showing larger differences compared with the I-solver (for instance 60º and 120º). Further refinements in the asymptotic solver, like requiring higher precision, might better approximate those results in regard of comparisons with the MLFMM. It is usually assumed that both solvers show similar results for regions where a high RCS value is present (main lobes), whereas minor lobes or nulls may present considerable differences. The use of more than one solver comes into play whenever it is necessary to cross-check results from within virtual simulations, i.e., when measurements are not available. If two different numerical methods with two different mesh types give results that are similar (though not absolutely equal), the user can then achieve a certain degree of confidence on the simulation. A comparison showing the performance in terms of required RAM memory and time is shown in Table 2. The computer used was a Quad Core Opteron, 2.51 GHz, with 64 GBytes RAM. It is noticeable that the asymptotic solver presents an advantage by trading the speed for precision (Sadiku, 2001; Huang and Boyle, 2008).

REFERENCES Catia, 1998, version 5, Retrieved in Oct, 27th 2011, from www.3ds.com. Coelho, L. V., 2007, “Missile Approach Warning System and its application in defense aircraft”, Conference SIGE IX, São José dos Campos, Brazil, in Portuguese. CST Microwave Studio, version 2010, Retrieved in Oct, 27th 2011, from www.cst.com. Davidson, D.B., 2005, “Computational Electromagnetics for RF and Microwave Engineering”, Cambridge University Press, Cambridge, England. Grant, R., 2010, “The radar game: understanding stealth and survivability”, Mitchell Institute Press, Arlington, the US. Huang, Y., Boyle, K., 2008, “Antennas: from theory to practice”, Ed. Wiley, West Sussex, the UK. Kouemou, G., 2009, “Radar Technology”, Ed. In-Teh, Vukovar, Croatia. Munteanu I., Timm M., Weiland T., 2010, “It’s about time”, IEEE Microwave Magazine, Vol. 11, No. 2, p. 60-9.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 287-294, Sep. - Dec., 2011

293

Perotoni M.B., Andrade L.A.

Peixoto, G.G., Alves, M.A., Orlando, A.J.F., Rezende, M.C., 2011, “Measurements in an Outdoor Facility and Numerical Simulation of the Radar Cross Section of Targets at 10 GHz”, Journal of Aerospace Technology and Management, São José dos Campos, Vol. 3, No. 1, p. 73-8. Pike, E.R., Sabatier P.C., “Scattering”, 2002, Ed. Academic Press, ISBN 0-12-613760-9, London, the UK.

294

Sadiku, M.N., 2001, “Numerical Techniques Electromagnetics”, 2nd Edition, CRC Press.

Skolnik, M.I., 1981, “Introduction to Radar Systems”, 2nd Edition, Ed. Mc Graw Hill. Song, J., Lu, C.C., Chu, W.C., 1997, “Multilevel Fast Multipole Algorithmfor Electromagnetic Scattering by Large Complex Objects”, IEEE Transactions on Antennas and Propagation, Vol. 45, No. 10, p.1488-93.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 287-294, Sep. - Dec., 2011

doi: 10.5028/jatm.2011. 03033011

Flavio Maldonado Bentes*

Fundação Jorge Duprat Figueiredo de Segurança e Medicina do Trabalho Rio de Janeiro/RJ – Brazil flavio.bentes@ufrj.br

Jules Ghislain Slama

Instituto Alberto Luiz Coimbra de PósGraduação e Pesquisa em Engenharia Rio de Janeiro/RJ – Brazil jules@mecanica.ufrj.br *author for correspondence

Sensitivity analysis of airport noise using computer simulation Abstract: This paper presents the method to analyze the sensitivity of airport noise using computer simulation with the aid of Integrated Noise Model 7.0. The technique serves to support the selection of alternatives to better control aircraft noise, since it helps identify which areas of the noise curves experienced greater variation from changes in aircraft movements at a particular airport. Keywords: Sensitivity analysis, airport noise, computer simulation.

LIST OF SYMBOLS ANAC: National Civil Aviation Agency SCxi: Sensitivity cofficient of the movement variable SCxi’: Sensitivity coefficient of the movement variable without 10% of the aircrafts dB: Decibel dB(A): Decibel, according to the A ponderation curve DNL: Day-night average noise level ΔΦ: Variation in the area of the noise curve FAA: Federal Aviation Administration Φ: Area of the noise curve INM: Integrated Noise Model LAeq: Equivalent sound pressure level LAeqD: Day equivalent sound pressure level LAeqN: Night equivalent sound pressure level RBAC: Brazilian Regulation for Civil Aviation SEL: Sound exposure level Sxi: Sensitivity to a movement xi SBRF: Guararapes International Airport (Recife/PE – Brazil) xi: Movement Variable for a group of aircrafts

Introduction With the global growth of aerial navigation, airport authorities have become more concerned about issues related to aircraft noise. For Infraero (2010), to navigate

Received: 25/06/11 Accepted: 09/09/11

means to safely conduct a watercraft or an aircraft from one point to another, which is a complex guidance process that enables long journeys with the goal to reach a specific place safely. The safety aspect should also include issues related to sound emission, once they can cause not only discomfort, but also damage to those who are continuously exposed to this type of noise. It is possible to say that the study of airport noise is really relevant worldwide, especially regarding issues related to aircraft noise. As to this aspect, the study concerning the sensitivity analysis is significantly helpful, since it allows identifying which areas of the noise curves have varied more from the changes in the aircraft movements at a specific airport. Airport noise is usually a result of discreet events, such as landings and take-offs. There are different noise sources in airports, coming from land operations involving aircraft fueling, movements and maintenance, however, landing and take-off operations are considered as the main noise sources of an airport. According to Morais, Slama and Mansur (2008), airport noise is a result of a sound field with intermittent temporal characteristics. The noise coming from the aircraft movements is directly related to the procedures of the aircrafts on the ground, be it before take-off or after landing. The study concerning airport noise embraces different fields of knowledge, from physics to mechanical engineering, especially focusing on the acoustic phenomenon and issues concerning the environment. The sensitivity analysis of airport noise is a method that uses acoustics software to simulate scenarios, with the objective to help control airport noise. Together with the guidelines of the balanced approach established by the International Civil Aviation Organization (2004), the technique contributes with a

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 295-300, Sep. - Dec., 2011

295

Bentes F.M., Slama J.G.

better analysis of the variations in the areas exposed to the levels of noise coming from the aircrafts. The numerical simulations use the Integrated Noise Model 7.0 (INM), which was created by the Federal Aviation Administration (2009) and enables the appearance of airport noise curves. INM requires the description of different airport parameters, such as runways, trajectories, fleet, route, airport coordinates, runway thresholds and noise curves starting from the choice of discomfort metrics. METRICS FOR AIRPORT NOISE There are different types of metrics to assess airport noise. Basically, noise metrics represents the energetic average of sound pressure levels in a definite period of time. According to the Brazilian Regulation for Civil Aviation 161 (2011), in order to determine noise curves, calculations should be made with software that uses appropriate methods with the day-night average level (DNL). This study presents a summary of some existing metrics: equivalent sound pressure level – LAeq, sound exposure level – SEL, and day-night average level – DNL). Equivalent sound pressure level - LAeq Noise levels can usually vary during a definite period of time. For Gerges (2000), the damaging effects of noise depend not only on its level, but also on how long it lasts. It is possible to say that LAeq is a constant sound pressure level that is equal to the variable noise levels during the measuring period, in terms of acoustic energy. As a consequence, LAeq represents the average sound level resulting from the integration throughout a period of time that can be defined with the logarithmic sum of all sound levels. LAeq can be divided between day and night. LAeqD is the day equivalent sound pressure level and represents the average sound energy calculated during daytime, from 7 to 22h, with a total of 15 hours. LAeqD is determined by Eq. 1. L A (t) 22 ¬ 1 ¼ (t) 10 LAeqD = 10log 10 dt ½ µ 7 ® 54000 ¾

(1)

LAeqN is the night equivalent sound pressure level, and represents the average sound energy calculated during the night, from 22h to 7h, with a total of 9 hours. LAeqN is determined by Eq. 2. L A (t) 7 ¬ 1 ¼ (t) 10 LAeqN = 10log 10 dt ½ µ 22 ® 32400 ¾

296

(2)

Sound exposure level – SEL SEL represents the total noise energy produced from an event. It is possible to say that SEL represents a logarithmic expression of the acoustic energy of the event, once it exceeds a specific type of noise, as if it had happened within a second. Thus, SEL is obtained by the sum of all sound pressure levels in one unit of time, inside the analyzed interval. Since SEL is a logarithmic expression regarding sound exposure in time, it can be used to compare the noise energy of events that last for different periods. The mathematical formulation to express the definition of SEL is demonstrated in Eq. 3: ¬1 SEL = 10log ® T0

t T t

¼ (t) dt ½ P02 ½¾

P 2A

(3)

Day-Night Average Sound Level – DNL DNL is commonly used to define the level of exposure to airport noise, and it also corresponds to the average sound energy caused by all airport events in a period of 24 hours. Ten dB (A) are added to the noise level for sound levels that occur during the night, from 22h to 7h of the next day, due to the higher sensitivity and disturbances caused by noise at night. According to the Code of Federal Regulations 14 CFR 150 (2004), DNL combines the sound energy of all aircraft operations from events that occur during daytime at an average noise exposure for that day. It is possible to say that the calculation of DNL is similar to LAeq, except that DNL adds 10 dB (A) to the night sound and is calculated in a period of 24 hours. According to Bistafa (2006), the relation between them is obtained with LAeq of every hour of each day. The average energy sum of the day and night, with extra 10 dB (A), results in the DNL. Eq. 4 mathematically defines DNL. LA(t) ¬ 22 LA(t) ¼¿ 7 ¯ 1 10 10 DNL = 10log ° 10 dt + 10 dt½ À (4) µ µ 22 ½¾ Á ± 3600.24 ® 7 10

DNL is usually used to define the areas of the noise curve, and has functions such as quantifying the cumulative noise exposure, considering events taking place during the day and the night. In Brazil, because of a recommendation by RBAC 161 (2011), DNL is used to calculate airport noise curves. The Code of Federal Regulations 14 CFR 150 (2004) also emphasizes that DNL has a penalty for night events since they cause more discomfort. We can say that DNL will identify the events that cause higher noise levels.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 295-300, Sep. - Dec., 2011

Sensitivity analysis of airport noise using computer simulation

METHODS AND DATA With the use of INM 7.0, the variations of the area of the noise curve (Î&#x201D;Ď&#x2020;) will be studied with DNL, as established by RBAC 161 (2011), in relation with variations of airport movements. The values of sensitivity coefficients are calculated after the elaboration of noise curves with INM, over the individual variation of each parameter, with other fixed parameters. Thus, it is possible to say that K = (x 1 , x 2 ,..., xn ) , in which the variable xn corresponds to the aircraft movements, during the daytime or the night. Considering the Ď&#x2020; variation when x1, x2,..., xn varies to x1 + â&#x2C6;&#x2020;x1, x2 + â&#x2C6;&#x2020;x2,â&#x20AC;Ś, xn + â&#x2C6;&#x2020;xn, as demonstrated in Eq. 5: Â¨K x1 x2,...,xn " K x1+Â¨ x1,x2+Â¨ x2,..., xn+Â¨ xn - K x1 x2,...,xn)

(5)

Thus, Î&#x201D;Ď&#x2020; can be described as demonstrated in Eq. 6: Â¨ K x1,x2,...,xn "

Â&#x2DC;K Â&#x2DC;K Â&#x2DC;K Â¨ x Â¨ x Â¨ x Â&#x2DC;x1 1 Â&#x2DC;x2 2 Â&#x2DC;x2 n

(6)

Therefore, it is possible to obtain the relative variation, which is equivalent to Eq. 7:

xi Â¨K K Â¨[i

(10)

The sensitivity coefficients can be expressed for x1 movements (group A, daytime), x2 (group A, night), x3 (group B, daytime), and x4 (group B, night). Equation 11 represents the sensitivity coefficient for a determinate xi movement: CSxi ~ -10

Ki - K 0 K0

(11)

Considering a logarithmic relation, it is possible to relate the logarithm of the area of the noise curve and the logarithm in relation to the movements multiplied by their respective sensitivity coefficients, which will result in Eq. 12:

+ CSxnlogexn

+ cte

(7)

The xi motion sensitivity can be defined by Eq. 8: (8)

Replacing Eq. 8 in Eq. 7, we come to Eq. 9: Â¨ K x1,x2,...,xn Â¨ x Â¨ x Â¨ x " Sx1 1 Sx2 2 Sxn n K x1,x2,...,xn x1 x2 xn

CSxi ~

logeK x1 x2,...,xn " CSx1logex1 + CSx2logex2 + ...

Â¨ K x1,x2,...,xn x1 Â&#x2DC;K Â¨ x1 " K x1,x2,...,xn K x1,x2,...,xn Â&#x2DC;x1 x1 xn Â&#x2DC;K Â¨ xn K x1,x2,...,xn Â&#x2DC;xn xn

xi Â&#x2DC;K Sxi= K x ,x ,...,x Â&#x2DC;x1 1 2 n

The values of sensitivity coefficients are defined from the determination of areas of noise curve using INM for the variation of each xi movement. Thus, Ď&#x2020; values were determined for x1, x2, x3, (â&#x20AC;Ś), xn in the initial situation and after the parameter variation for x1 + Î&#x201D;x1,x2 + Î&#x201D;x2,x3 + Î&#x201D;x3,(â&#x20AC;Ś), xn+ Î&#x201D;xn. Therefore, the sensitivity coefficient for xi will be demonstrated in Eq. 10:

(12)

The sensitivity analysis was conducted with computational numerical analysis in Guararapes International Airport â&#x20AC;&#x201C; Recife/PE, Brazil (SBRF). Information concerning the flights was gathered via online airline schedule provided by the National Civil Aviation Agency (ANAC), from Brazil (2011). Table 1 presents aircraft movements by period. RESULTS

(9)

The areas of noise curves were calculated for all the aircrafts. Afterwards, the areas of noise curves in groups A and B were calculated both for daytime (D) and night (N)

Table 1.Aircraft movements by period.

Movements Period D x1 x2

x3 x4

D N

Aircrafts A318, A319, A320, A321, A332, A343, B733, B734, B737, B738, B744, B752, B762, B763, E190, F100 A318, A319, A320, A321, A332, A343, B733, B734, B737, B738, B744, B752, B762, B763, E190, F100 AT72, B722, L410 AT72, B722, L410

J. Aerosp.Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.3, No.3, pp. 295-300, Sep. - Dec., 2011

Group A A B B 297

Bentes F.M., Slama J.G.

movements. Table 2 presents the values of areas of noise curves calculated for the respective groups of aircrafts. Calculation was conducted with DNL, as recommended by RBAC 161 (2011), for different noise curves, and the object of analysis was Guararapes International Airport – Recife/PE, Brazil (SBRF). Table 3 presents the values of areas of noise curves calculated after 10% of the aircrafts had been removed for all movements. Simulations were conducted with the same metrics, resulting in DNL 55, 60, 65, 70, 75, 80 and 85 dB(A) noise curves, with their respective characteristics. Table 4 presents the sensitivity coefficient values before (SCxi) and after (CSxi’) 10% of the aircrafts were removed.

Sensitivity variations are more noticeable for bigger changes in noise curve areas, which were calculated and are demonstrated in Tables 2 and 3. From the analysis conducted after obtaining the sensitivity coefficients, it is possible to imply there will be a higher variation in the noise curve areas for the movement variables x3, x4 (Table 2) and x3’, x4’ (Table 3). The higher the variation of the area of noise curves, the bigger the reduction of the noise, since the area of the noise curve will decrease. For movements x1, x2, x1’, x2’, especially x2, x2’, lower values of sensitivity coefficients were calculated in almost all the curves, which shows a lower variation as to noise curves for the respective movements and the maintenance of higher noise levels, close to the initial condition.

Table 2. Values of noise curve areas with all the aircrafts.

DNL (dB(A)) 55 60 65 70 75 80 85

All the aircrafts 47,609 19,034 7,030 2,569 1,022 0.25 0.1

Noise curve areas with all the aircrafts (km²) Group A Group A Group B Daytime (x1) Night (x2) Daytime (x3) 11,382 30,998 3,875 4,359 11,753 1,326 1,684 4,582 0.338 0.589 1,756 0.102 0.161 0.689 0.023 0.065 0.173 0 0.013 0.073 0

Group B Night (x4) 1,778 0.527 0.111 0.045 0.014 0.001 0

Table 3. Values of noise curve areas after 10% of the aircrafts were removed.

DNL (dB(A)) 55 60 65 70 75 80 85

Noise curve areas after 10% of the aircrafts were removed (km²) Group A Group A Group B All the aircrafts Daytime (x1’) Night (x2’) Daytime (x3’) 38,983 10,513 28,028 3,685 14,266 4,166 10,699 1,289 5,645 1,581 4,313 0.479 2,025 0.579 1,648 0.165 0.838 0.231 0.653 0.045 0.197 0.086 0.258 0.008 0.085 0.023 0.095 0.001

Group B Night (x4’) 1,617 0.556 0.214 0.087 0.028 0.005 0

Table 4. Values of sensitivity coefficients for different noise curves before and after the aircrafts were removed.

DNL (dB(A)) 55 60 65 70 75 80 85 298

SCx1 7,609 7,710 7,605 9,998 9,998 7,400 8,700

SCxi values SCx2 SCx3 3,489 9,18608 3,825 9,30335 3,482 9,99952 3,165 9,99960 9,993 9,99977 3,080 10,00000 2,700 10,00000

SCx4 9,62654 9,99972 9,99984 9,99982 9,99986 9,96000 10,00000

SCx1’ 7,303 7,080 7,199 9,997 7,243 5,635 7,294

SCxi’ values SCx2’ SCx3’ 2,810 9,05472 2,500 9,09645 2,360 9,99915 1,862 9,99919 2,208 9,46301 3,096 9,59391 1,176 9,88235

SCx4’ 9,58520 9,99961 9,99962 9,99957 9,66587 9,74619 10,00000

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 295-300, Sep. - Dec., 2011

Sensitivity analysis of airport noise using computer simulation

CONCLUSIONS Using the sensitivity analysis by computational numerical simulation enables to identify the variations in the most significant areas of noise curves that should be carefully analyzed by airport authorities. Since the subject of airport noise is really relevant in the international context and due to the expectations as to the growth of the aerial modal, the study of sensitivity analysis can be seen as a tool to help noise control, especially since it enables identifying which areas of noise curve vary the most. Thus, its use makes measurements to control airport noise more effective. ACKNOWLEDGEMENTS To Instituto Alberto Luiz Coimbra, of post-graduation and Research in Engineering of Universidade Federal do Rio de Janeiro and to the study group in Airport Noise, which allowed the performance of this study. REFERENCES ANAC, 2011, “Agência Nacional de Aviação Civil, Horário de Transportes”, from: http://www.anac.gov.br/ hotran. Bistalfa, S. R., 2006, “Acústica aplicada ao controle de ruído”, Ed. Edgard Blücher, São Paulo, Brasil.

Federal Aviation Administration, 2009, “Integrated Noise Model, User’s Guide, Versão 7.0”, FAA, Washington, USA. Code of Federal Regulations 14 CFR 150, 2004, “Noise Compatibility Planning”, Federal Government of the United States, Washington, USA. Gerges, S. N. Y., 2000, “Ruído – Fundamentos e Controles”, 2ª edição, Universidade Federal de Santa Catarina, Florianópolis, Brasil. Infraero, 2010, “Empresa Brasileira de Infra-Estrutura Aeroportuária”. from: http://www.infraero.gov.br/index. php/br/navegacao-aerea.html. International Civil Aviation Organization, 2004, “Final draft of Guidance on the Balanced Approach to Aircraft Noise Management”, Montreal, Canada. Morais, L. R., Slama, J. G., Mansur, W. J., 2008, “Utilização de barreiras acústicas no controle de ruído aeroportuário”, Universidade Federal do Rio de Janeiro, COPPE/PEM /LAVI/ GERA. VII SITRAER, Rio de Janeiro, Brasil. pp. 732-744. Regulamento Brasileiro de Aviação Civil 161 (RBAC 161), 2011, “Planos de zoneamento de ruído de aeródromos – PZR” Texto aprovado por meio da Resolução ANAC nº 202, Secretaria de Aviação Civil, Agência Nacional de Aviação Civil – ANAC, Brasília, Brasil.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 295-300, Sep. - Dec., 2011

299

doi: 10.5028/jatm.2011. 03032511

Marco Antonio da Silva Ferro*

Institute of Aeronautics and Space São José dos Campos/SP – Brazil marcomasf@iae.cta.br

Jorge Yamasaki

Institute of Aeronautics and Space São José dos Campos/SP – Brazil yamasakijy@iae.cta.br

Douglas Roberto M. Pimentel

Institute of Aeronautics and Space São José dos Campos/SP – Brazil douglas.roberto.fis@gmail.com

Kleber Pinheiro Naccarato

National Institute for Space Research São José dos Campos/SP – Brazil kleberp@dge.inpe.br

Marcelo Magalhães Fares Saba

National Institute for Space Research São José dos Campos/SP – Brazil msaba@dge.inpe.br *author for correspondence

Lightning risk warnings based on atmospheric electric field measurements in Brazil Abstract: This paper presents a methodology that employs the electrostatic field variations caused by thundercloud formation or displacement to generate lightning warnings over a region of interest in Southeastern Brazil. These warnings can be used to prevent accidents during hazardous operations, such as the manufacturing, loading, and test of motor-rockets. In these cases, certain equipment may be moved into covered facilities and personnel are required to take shelter. It is also possible to avoid the threat of natural and triggered lightning to launches. The atmospheric electric field database, including the summer seasons of 2007/2008 and 2008/2009 (from November to February), and, for the same period and region, the cloud-to-ground lightning data provided by the Brazilian lightning detection network – BrasilDAT – were used in order to perform a comparative analysis between the lightning warnings and the cloud-toground lightning strikes that effectively occurred inside the area of concern. The analysis was done for three areas surrounding the sensor installation defined as circles with 5, 10 and 15 km of radius to determine the most effective detection range. For each area it was done using several critical electric field thresholds: +/- 0.5; +/- 0.8; +/- 0.9; +/- 1.0; +/- 1.2; and +/- 1.5 kV/m. As a result of the reduction of atmospheric electric field data provided by the sensor installed in area of concern and lightning provided by BrasilDAT, it was possible, for each of the areas of alert proposals, to obtain the following parameters: the number of effective alarms; the number of false alarms; and the number of failure to warning. From the analysis of these parameters, it was possible to conclude that, apparently, the most interesting critical electric field threshold to be used is the level of 0.9 kV/m in association with a distance range of 10 km around the point where the sensor is installed. Keywords: Atmospheric electric field, Electric field-mill, Lightning.

Introduction A recent study showed that more than half of lightning casualties resulted from the first or one of the first few cloud-to-ground (CG) flashes in a storm, and significant numbers of casualties resulted from returning to outdoor activities just before lightning had actually ceased (Lengyel, 2004). Between these times, when the threat of lightning is obvious, there are fewer casualties. Thus, it can be concluded that the initiation and cessation of lightning activity are critically important periods for both patrimonial and human safety. Therefore, great effort has been made to develop methods for accurate lightning occurrence forecast both to make secure the development of critical activities and to protect human life in several outdoor activities. Received: 06/06/11 Accepted: 20/09/11

Several papers have dealt with lightning warning methods developed from CG lightning location systems (LLS) (Murphy and Cummins, 2000; Murphy et al., 2002; Holle et al., 2003). Furthermore, other particular researches combine total lightning with weather radar information in an effort to improve the accuracy of lightning threat alarms (Murphy and Holle, 2005; 2006). Finally, some recent studies presented automated lightning warning systems as a combination of lightning detection information and data from one or more electric field mills (EFMs) (Murphy et al., 2008; Montanya et al., 2008; Beasley et al., 2008; Aranguren et al., 2009). These studies had shown that the EFM measurements were strongly influenced by the characteristics of the place where it was installed, for example: the cloud charge center height in that region, the topography, and so on. Thus, the effectiveness of lightning warning methods using electric field data changes in one region to another. It is important to note that all the studies took place in the Northern hemisphere. Naccarato et al. (2008) showed a preliminary analysis comparing automatic warnings triggered only by EFM and those

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 301-310, Sep. - Dec., 2011

301

Ferro, M.A.S. et al.

based only in CG lightning discharge occurrence in a specific region in Southeastern Brazil. The present study extends the former analysis carried out by Naccarato (2008) for the same region, located in São José dos Campos, state of São Paulo, comparing data from one EFM to the CG lightning data provided by the Brazilian Lightning Detection Network (BrasilDAT) to evaluate how the atmospheric electric field variation data can be used to support the decision-making process of generating a lightning risk warning.

Charge Amplifier Charged Sensor Plate

Shutter Open Eletric Field Charges Sensor Plate

Shutter Closed Eletric Field Blocked

Figure 1. Diagram of operation of an electric field mill.

EFMs are designed to determine the relative strength of the electric field by comparing its level in a known, stable, uncharged, and reference object. When an uncharged sensor plate is exposed to an electric field, it becomes charged. Thus, unlike lightning detection systems, which respond to fast transients in the electromagnetic field generated by lightning, EFMs detect the electrostatic field and relatively slow changes in that field. They detect the presence of charge separation and net charge directly above and in the immediate surroundings of the sensor. Depending on where the charge is located, the effective detection range of an EFM varies from a few kilometers to perhaps as much as 20 km (Murphy et al., 2008). Field changes in the order of a fraction of a second are due to the overall rearrangement of the thundercloud charge distribution, which is produced by a lightning flash, and slower field changes are due to cloud electrification and rearrangement of space charge in the atmosphere. Hence, since the conditions are constantly changing, there is a need to constantly measure the strength of the electric field, which is translated into the need to alternately read the charged state of the sensor plate, discharge it, and read again. This is accomplished by repeatedly exposing the sensor plate to the external electric field to charge it, then shielding the plate to allow it to discharge. The process of exposing (charging) and shielding (discharging) the sensor plate from the electrical field is accomplished by means of a rotary shutter, consisting of a motor-driven, mechanically complementary rotor/ stator pair. As the motor rotates, the shutter alternately opens to allow the external electric field to charge the sensor plate, and then it closes (Fig. 1) in order to shield the sensor plate to discharge or reset, in preparation for the next measurement. The main advantage of the EFM is the protection from the occurrence of the first CG lightning strike.

302

Uncharged Sensor Plate

Rotating Shutter

BASIC CONSIDERATIONS EFM

Charge Amplifier

Electric field meters are typically factory calibrated using a parallel plate method, where a uniform electric field is developed by applying a known voltage between parallel conductive plates. Each EFM is factory calibrated in a parallel plate calibration fixture with the instrument aperture mounted in upward-facing. A linear fit of the calibration data results in a calibration equation expressed as: E = f.V

(1)

The multiplier f is a function of the EFM electrode dimensions and the characteristics of the charge amplifier’s electronic circuit. The manufacturer of the EFM used in this study estimated that the measurement accuracy of f for the instrument calibrated in the parallel plate electric field calibrator is ±1%. However, when monitoring the Earth’s electric field, Eq. 1 is valid only if the instrument aperture is mounted flush with the Earth’s surface and upward-facing. Yet, for permanent outdoor measurements of electric field, a flush-mounted and upward-facing orientation is problematic because of dirt, bird droppings, rain, and so on, collecting on the sense electrodes and fouling the measurement. Consequently, a downward facing and elevated configuration is used for longterm field applications. Inverting the EFM reduces the effective gain, while increasing it height above ground enhances the gain, with respect to an ideal upwardfacing flush-mounted geometry. A site correction factor Csite is necessary to correct f for non flush-mounted configurations. The corrected multiplier becomes as Eq. 2: E = Csite.f.V

(2)

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 301-310, Sep. - Dec., 2011

Lightning risk warnings based on atmospheric electric field measurements in Brazil

In this equation, f is unique for each EFM, yet independent of a given site, whereas Csite is unique for each given site, yet independent of the particular EFM used at the site. Csite is typically determined by using a flush-mounted upward-facing unit in the vicinity of the site that needs correction. An upwardfacing calibration kit was used to hold the EFM in a flush-mounted upward-facing position. The collected data from both units, the upward-facing unit and a downward facing EFM installed in a specific site, are plotted (Fig. 2). A best-fit line computed from the data resulted in the Csite.

The BrasilDAT overall DE was already estimated by a DE model (DEMo) recently developed by the ELAT Group (Naccarato and Pinto Jr., 2008). A large area of the country is covered by an 80% or higher DE network. COMPLEMENTARY OBSERVATIONS Charge separation inside a thunderstorm cloud causes a reversal of the electrostatic polarity and an increase in the

LLS A LLS can locate CG lightning flashes with detection efficiency (DE) higher than 80% and location accuracy lower than 500 m, due to its network of precise sensors that detect the electromagnetic radiation of the lightning channels.

Figure 3. Area covered by BrasilDAT network.

200

EF for downward facing position -10000

-8000

-6000

-4000

-2000

y = 0.110x + 9.917 R2 = 0.813

2000

-200 -400 -600 -800 -1000 -1200

EF upward facing position

Furthermore, based on quality criteria and correction parameters, the central processing unit (responsible for computing the solutions based on the sensor reports) can discriminate more than 90% of the intracloud (IC) discharges from CG lightning strokes. In Brazil, there is a LLS operating since 1998, called BrasilDAT. Nowadays, this network is composed by 36 sensors as shown in Fig 3.

-1400 Figure 2. Determination of Csite. J. Aerosp.Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.3, No.3, pp. 301-310, Sep. - Dec., 2011

303

Ferro, M.A.S. et al.

magnitude of the field measured by an EFM in its vicinity. These characteristics can be used to trigger a warning, given that charge separation has to precede lightning. However, in not all cases does the field reverse polarity at a particular EFM site as a storm develops, and the exact magnitude reached by the field depends strongly on the distance between the EFM and the cloud charge regions (Murphy, 1996). Thus, a warning system using a fixed threshold for the field magnitude may or may not pick up all storms.

installed in São José dos Campos (23º19’48.20”S and 45º48’31.85”W) in the Southeast region of Brazil. This period includes the summer seasons of 2007/2008 and 2008/2009 (from November to February). For the same period and region, it was selected the CG lightning data provided by the BrasilDAT network in order to perform a comparative analysis between the lightning warnings and the CG lightning strikes, which effectively occurred inside the AOC.

A LLS has a greater effective detection range (about hundreds of kilometers) than an EFM (about one or two tens of kilometers). Thus, a LLS is more effective than an EFM to detect cases where a mature thunderstorm moves toward the area of concern (AOC) from elsewhere.

There were multiple thunderstorms that moved over the AOC during the period of the study. Some of them started and developed directly over the AOC. In the others cases a mature thunderstorm moved towards the AOC from elsewhere. Most of them moved from South to the North, following the direction of the valley formed by two chains of mountains named “Serra do Mar” and “Serra da Mantiqueira”, where the city of São José dos Campos is located.

LLS is a complex measurement system that by itself can help to provide a relative accurate CG lightning forecast to a particular area. However, the main disadvantages of a LLS are the very high installation and maintenance costs and it is not able to forecast the first CG lightning strike. Thus, the main reason for using EFMs is their capability of detecting the development of a thunderstorm directly over the AOC (Fig. 4).

Figure 4. Area surrounding the sensor installation site.

OBJEcTIVES The main objective of this study was to define criteria that will be used by an atmospheric electric field monitoring system to provide lightning risk warnings. Since the AOC is located in a region that has good coverage by the BrasilDAT, it is possible to compare the warnings generated by the EFM to the CG lightning data provided by BrasilDAT in order to evaluate the rate of false warnings and/or fail to warning based on some critical field thresholds. METHODOLOGy An atmospheric electric field database from November, 2007 to nowadays was created using an EFM

304

Therefore, the main purpose of this analysis is to simulate lightning warnings based on the atmospheric electric field measurements provided by the EFM and to compare them to the CG lightning data provided by the BrasilDAT network after some time from the beginning of warning. Thus, it is possible to evaluate the performance of the atmospheric electric field monitoring system. In order to simulate EFM automatic warnings, the equipment is assumed to be installed in the center of the AOC. High frequency oscillations due to the rearrangement of cloud charge by lightning are also detected by the EFM during a thunderstorm. These fast field changes have to be removed from the dataset to avoid that the lightning-caused field changes make the field rise above the threshold level used in a warning system. The use of a 60-second average of the EFM measurement values is assumed to be a satisfactory smoothing technique to filter high frequency oscillations. Figure 5 shows an example of this smoothing technique. The warnings will be evaluated based on the following parameters: •

effective alarm (EA) is the warning that was triggered before the CG lightning occurred inside the AOC;

•

lead time (LT) is the time interval between the time of the warning and the occurrence of a CG lightning inside the AOC;

•

failure to warning (FTW), when the first CG lightning strikes the AOC area without a previous warning;

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 301-310, Sep. - Dec., 2011

Lightning risk warnings based on atmospheric electric field measurements in Brazil

Electric Field [kV/m]

10 8

1.5

0.5

-0.5

-2

-1

-4

-1.5

-6

-2 -2.5

5 km - discharges

10 km - discharges

-8

15 km - discharges

Electric Field (kV/m)

-10

Pick Current [kA]

Electric Field X Lightning Discharges 09/03/2008

2.5

Figure 5. Example of the smoothing technique.

•

false alarm (FA) is a warning triggered without the subsequent occurrence of a CG lightning inside the AOC after 45 minutes since the trigger.

The forecast verification will be done based on the contingence table described in Table 1. Table 1. Contingence table.

Observed

Forecast

Yes

Total

Yes

Forecast

FTW

CNW

Not Forecast

Total

Observed

Not Observed

Total

EA: effective alarm; FA: false alarm; FTW: failure to warning; CNW: correctly not warned. Two statistical variables will be used in this study: the probability of detection (POD), which is simply the ratio of the number of successful warnings to the total number of episodes of a CG in the AOC, and the FA rate (FAR), which is the ratio of the number of FA to the total number of alarms (Eqs. 3 and 4). POD = EA / (EA + FTW)

(3)

FAR = FA / (EA + FA)

(4)

They will help evaluating how the atmospheric electric field data could be used to better support the correct threshold level trigger of a lightning risk warning. The analysis will be done for both circular and annular regions. On one hand, the influence of the number of EA, FA and FTW is better denoted using the annular regions ranging from 0 to 5 km; 5 to 10 km, and 10 to 15 km. On the other hand, to calculate the POD and the FAR is better to use areas surrounding the sensor installation defined as circles with 5, 10 and 15 km of radius. The analysis for each area is going to be done using several critical electric field thresholds: +/- 0.5; +/- 0.8; +/- 0.9; +/- 1.0; +/- 1.2; and +/- 1.5 kV/m. The warning is classified as successful if one or more CG lightning strike(s) is(are) detected inside the AOC up to 45 minutes after the warning is triggered. A delay time (DT) of 45 minutes is adopted to alarm extinction. This value was chosen as the midrange of the active phase of a storm cloud found in literature. The count of DT is restarted every time an alert criterion is detected. RESULTS AND DISCUSSION The total number of EA, FA and FTW was analyzed as a function of critical electric field threshold and the distance of sensor installation site. The behavior of the EA, FA and FTW total numbers are resumed in Table 2.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 301-310, Sep. - Dec., 2011

305

Ferro, M.A.S. et al.

Table 2. Total number of EA, FA and FTW.

Total nr of FA Total nr of FTW

0.8

0.9

1.0

1.2

1.5

0 to 5

5 to 10

10 to 15

0 to 5

5 to 10

10 to 15

0 to 5

5 to 10

10 to 15

EA: effective alarm; FA: false alarm; FTW: failure to warning. Figure 6 shows the behavior of the EA, FA and FTW total numbers as a function of the electric field threshold. The results show that the total number of EA decreases, while the electrical field threshold increases. But it shows a steady value between 0.8 and 1.0 kV/m dropping again from this point. There is a significative increase in the total number of EA when the annular region is changed from 0 to 5 km to 5 to 10 km. However, between the annular regions of 5 to 10 km and 10 to 15 km, the total numbers of EA are very close, suggesting that there is no significant gain changing the annular region from 5 to 10 km to 10 to 15 km. The FA behavior shows a smooth oscillation with the smaller number of FA occurring for the threshold of 1.0 kV/m. It is very interesting the significant decrease in the total number of FA when the annular region is changed from 0 to 5 km to 5 to 10 km. But, in the same way that EA, between 5 to 10 km and 10 to 15 km the total numbers of FA are very close, suggesting that there is no significant gain changing the annular region from 5 to 10 km to 10 to 15 km. The total number of FTW increases when the distance range is increased, which shows a smooth oscillation between 0.9 and 1.0 kV/m. Figure 7 shows the behavior of the total number of EA, FA and FTW as a function of the distance from the sensor. It is possible to note that the total number of FA and FTW does not suffer significant modification when the electric field threshold increases. Also, it is possible to note that there is no significant change in the total number of EA when the annular region changes from 5 to 10 km to 10 to 15 km for all electric field thresholds. In general, the

306

Total Nr of EA

0.5

EA as a function of the threshold 100 80 60 40 20 0

56 55 29 0.5

53 44 28 0.8

52 47 28 0.9

48 44

41 38

26 1

19 1.2

34 30 15 1.5

0 to 5km 5 to 10km 10 to 15km

EF thresold [kV/m} EA as a function of the threshold 100 80 60 40 20 0

Total Nr of EA

Total nr of EA

Critical electric field threshold (kV/m)

40 15 13 0.5

41 25

39 21

37 18

42 23

16 0.8

16 0.9

13 1

1.2

34 19 15 1.5

0 to 5km 5 to 10km 10 to 15km

EF thresold [kV/m} EA as a function of the threshold 100 80 60 40 20 0

Total Nr of EA

Distance range (km)

12 0.5

17 0.8

18 1

0.9

77 54 23 1.2

81 59 24 1.5

0 to 5km 5 to 10km 10 to 15km

EF thresold [kV/m}

Figure 6. Total number of EA, FA, and FTW as a function of electric field threshold.

total number of EA continuously decreases when the field threshold increases, after 1.0 kV/m, and for the annular region 5 to 10 km and bigger EA is smaller than FTW. Therefore, from the point of view of total number of events of EA, FA and FTW, apparently the most interesting critical electric field threshold to be used is the level of 0.9 kV/m for the annular region of 5 to 10 km. Though the annular region from 10 to 15 km shows the biggest number of EA events, it also shows a high number of FTW. Taking into account safety, it is better to adopt a configuration that results in a small number of FTW, although resulting in a EA reduction and some increase in the FA number, mainly when the total numbers of EA for the annular region of 5 to 10 km and 10 to 15 km are very close, as shown in Figs. 5 and 6. The ratio of predicted CG flashes versus FTW shows larger values (between 6:1 and 2:1) for all electric field thresholds when the actuation area of the sensor is restricted to a radius of 5 km as shown in Table 3. The ratio shrinks as the sensor’s actuation area increases, reaching 1:1, i.e., for each predicted CG flash there is another that the sensor failed to predict.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 301-310, Sep. - Dec., 2011

Lightning risk warnings based on atmospheric electric field measurements in Brazil

Table 3. Relationship between predicted CG flash versus FTW.

Total Nr of Events

Total Nr of events for 0.5 kV/m 100 80 60 40 20 0

Critical field threshold Distance (kV/m) range (km) 0.5 0.8 0.9 1.0 1.2 1.5

FA FTW EA 0 a 5km

5 a 10km

10 a 15km

(EA + FA)/ FTW

Total Nr of Events

Total Nr of events for 0.8 kV/m 100 80 60 40 20 0

FA FTW EA 0 a 5km

5 a 10km

10 a 15km

Total Nr of Events

Total Nr of events for 0.9 kV/m 100 80 60 40 20 0

FA FTW EA 0 a 5km

5 a 10km

10 a 15km

Total Nr of Events Total Nr of Events

Total Nr of events for 1.2 kV/m 100 80 60 40 20 0

FA FTW EA 0 a 5km

5 a 10km

10 a 15km

Total Nr of Events

Total Nr of events for 1.5 kV/m 100 80 60 40 20 0

FA FTW EA 0 a 5km

5 a 10km

Table 4. Probability of detection.

Radius Critical field threshold (kV/m) (km) 0.5 0.8 0.9 1.0 1.2 1.5

FA FTW EA 5 a 10km

However, the ratio of predicted CG flash versus FTW is an approximated notion of warning effectiveness. The POD results in a more adequate indicator inasmuch as it takes into account the EA instead of the predicted CG flash number, indicating the percentage of lightning events observed, indicated as “yes” in the contingency table (Table 1), which have been properly warned (Eq. 3). The POD as a function of the electric field threshold and sensor’s actuation area is shown in Table 4.

Total Nr of events for 1.0 kV/m

0 a 5km

EA: effective alarm; FA: false alarm; FTW: failure to warning.

10 a 15km

100 80 60 40 20 0

10 a 15km

Figure 7. Total numbers of EA, FA and FTW as a function of the distance range of sensor.

EA/(EA + FTW)

EA: effective alarm; FA: false alarm; FTW: failure to warning. The use of the electric field threshold of 0.9 kV/m together with the area with a radius of 10 km, suggested by previous analyses, leads to a POD value of 60%, bigger than values found in some past studies. Using an electric field threshold of 1.0 kV/m, Aranguren et al. (2009) found 37.5% for POD in Catalonia, Spain, and Murphy et al. (2008) found 34.4% in Florida. Naccarato et al. (2008) in a previous study carried out in the same area of the present paper, analyzing a dataset of 30 days and continuous records and using a threshold of 0.5 kV/m and distance range of 10 km, found that 82.4% of the discharges inside their AOC were correctly warning. Beasley et al. (2008) observed that the electric field magnitude exceeded 1.0 kV/m, inside an area with radius of 10 km around the strike point and ten minutes before the first stroke, in 66.0% of the cases. The fair-weather electric field could present values of some few hundred Volts/m due to some disturbances (e.g., mist, smoke, etc.) and they could influence the number of

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 301-310, Sep. - Dec., 2011

307

Ferro, M.A.S. et al.

FA. FAR indicates the percentage of FA related to the total number of predicted flashes (Eq. 4). FAR as a function of the critical field threshold and sensor’s actuation area is shown in Table 5.

•

Beasley and Murphy (2008) used the data from the network of EFM at the Kennedy Space Center and the adjacent Cape Canaveral Air Force Station that comprises 31 sensors, though Murphy had analyzed the data from only two sensors in the network.

Table 5. False alarm rate.

Radius (km)

FA/(EA + FA)

Critical field threshold (kV/m)

CONCLUSIONS

0.5 0.8 0.9 1.0 1.2 1.5

EA: effective alarm; FA: false alarm. The use of electric field threshold of 0.9 kV/m together with the area with a 10 km radius, suggested by previous analyses, leads to a FAR value of 41%, smaller than the ones obtained by Aranguren et al. (2009) (87.0% - Catalonia region, Spain) and Murphy et al. (2008) (74.1% - Florida, USA) in the previously mentioned studies (Table 6). Although the previous analyses have pointed the threshold of 0.9 kV/m like the more interesting to use, for the purpose of comparing it with previous studies, Table 6 shows the values related to the threshold of 1.0 kV/m. With regards to the information in Table 6, it is important to note that: •

Beasley (2008) used data concerning only the lightning that effectively occurred inside the AOC to analyze the warning system efficiency;

•

Murphy and Naccarato (2008) did not analyze in their studies the LT;

•

Naccarato (2008) did not relate the DT – the warning duration time interval – and used the threshold of 0.5 kV/m to trigger warnings; and

With the purpose of evaluating how the atmospheric electric field variation data can be used to support the decision-making process of generating a lightning risk warning, it was used information from an atmospheric electric field database from November, 2007 to February, 2009 using an EFM in Southeastern Brazil. Since the area of interest lies in a region with excellent coverage of BrasilDAT, it was possible to compare the warnings generated by the proposed system to the CG lightning data provided by BrasilDAT in order to evaluate the rate of false warning and/or fail to warning based on some electric field thresholds. The analysis was carried out for both circular and annular regions. The influence of the numbers of EA, FA and FTW is better denoted using the annular regions of 0 to 5 km, 5 to 10 km and 10 to 15 km. To calculate the POD and the FAR, it is better to use areas surrounding the sensor installation defined as circles with 5, 10 and 15 km of radius. The analysis for each area was done using several electric field thresholds: +/- 0.5; +/- 0.8; +/- 0.9; +/- 1.0; +/- 1.2; e +/- 1.5 kV/m. As a result of the reduction of atmospheric electric field data provided by the sensor installed in AOC and lightning provided by BrasilDAT, it was possible, for each of the areas of alert proposals, to obtain the following parameters: the number of EA; the number of FA; and the number of FTW. From the analysis of these parameters, it was possible to conclude that, apparently, the critical electric field threshold more interesting to be used is the level of 0.9 kV/m in association with a distance range of 10 km around the point where the sensor was installed.

Table 6. Resume comparing studies in the literature.

Sensors used

Threshold (kV/m)

LT (min.)

DT (min.)

POD (%)

FAR (%)

Beasley (2008)

1.0

9.0 - 12.0

10.0

66.0

-x-

Murphy et al. (2008)

1.0

-x-

15.0

34.4

74.1

Naccarato et al. (2008)

0.5

-x-

82.4

17.6

Aranguren et al. (2009)

1.0

6.5

30.0

37.5

87.0

Present study

1.0

13.0

45.0

58.0

41.0

-x-: not analyzed in the study. 308

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 301-310, Sep. - Dec., 2011

Lightning risk warnings based on atmospheric electric field measurements in Brazil

To this electric field threshold, the choice of distance range of 10 km is justified by the fact that, from a security standpoint, it is preferable to have a greater number of FA and less FTW, than otherwise.

data from both lightning detection and location network combined with electric field measurements and meteorological radar.

Values found for the POD (58%) and for the FAR (41%), using the electric field threshold of 1.0 kV/m and area with a radius of 10 km are significantly better than the ones found by Aranguren et al. (2009) in Catalonia. However, they show worse performance than that found by Naccarato et al. (2008) in the same region of this study. It is important to note that, like this paper, both studies used data from only one sensor. Nevertheless, Naccarato et al. (2008) adopted the threshold of 0.5 kV/m to trigger a warning, which resulted in high values of POD as a consequence of a large number of EA and a small number of FTW.

References

The values of POD found by Murphy et al. (2008) and Aranguren et al. (2009) (34.4 and 37.5%, respectively) are smaller than those found herein probably due to the fact that they are using a small DT, which tends to decrease the number of EA diminishing as a consequence the POD. Besides that, it is important to note that, as mentioned in the Introduction, the electric field sensor’s measurements are strongly influenced by the local characteristics of the installation site (cloud center charge height in that region, topography, etc.). Therefore, measurements from sensors installed at sea level (Murphy – Florida) can result in a smaller POD as a consequence of a bigger distance to the center charge of the cloud than the one carried out in higher places (Aranguren – Terrassa, Spain – 300 m above the sea level). This study was carried out in a region located 800 m above the sea level. It is understood that the height of the center charge of the cloud varies with the latitude. Therefore, the effectiveness of lightning warning systems that use electric field data varies as a function of this parameter (Florida – 24º N; Terrassa – 41º N; São José dos Campos – 23º S). Thus, the system based on only one sensor, assuming an area with a radius of 10 km around it and an electric field threshold of 0.9 kV/m to trigger the warning, showed a very interesting performance (POD=60% and FAR=41%) compared to the studies found in literature. The average time interval before the first lightning occurrence (LT) found in this study (13 minutes) is higher than that found in the mentioned studies. Other methodologies and additional criteria can be used to increase POD and to decrease the FAR. The use of an electric field-mill network is a choice and another one is based on the use of simultaneous

Aranguren, D., et al., 2009, “On the lightning hazard warning using electrostatic field: Analysis of summer thunderstorms in Spain”. Atmospheric Research. doi:10,1016/j.elstat.2009.01.023. Beasley, W.H., Williams, D.E., Hyland, P.T., 2008, “Analysis of surface electric-field contours in relation to cloud-to-ground lightning flashes in air-mass thunderstorms at the Kennedy Space Center”. 20th International Lightning Detection Conference (ILDC). Proceedings. Tucson. Holle, R.L., Murphy, M.J., Lopez, R.E., 2003, “Distances and times between cloud-to-ground flashes in a storm”. International Conference on Lightning and Static Electricity (ICLSE). Proceedings. Blackpool. Lengyel, M.M., 2004, “Lightning casualties and their proximity to surrounding cloud-to-ground lightning”, M.S. Thesis. University of Oklahoma. Montanya. J., et al., 2008, “Total lightning, electrostatic field and meteorological radar applied to lightning hazard warning”. 20th International Lightning Detection Conference (ILDC). Proceedings. Tucson, CD-ROM. Murphy, M.J., 1996, “The electrification of Florida thunderstorms”. PhD thesis, University of Arizona. Murphy, M.J., Cummins, K.L., 2000, “Early detection and warning of cloud-to-ground lightning at a point of interest”. 2nd Symposium on Environmental Applications. Proceedings. American Meteorological Society, Long Beach, p. 172-177. Murphy, M.J., Demetriades, N.W.S., Cummins, K.L., 2002, “Probabilistic early warning of cloud-toground lightning at an airport. 16th Conference on Probability and Statistics in the Atmospheric Sciences”. Proceedings. American Meteorological Society, Orlando, p. 126-131. Murphy, M.J., Holle, R.L., 2006, “Warnings of cloudto-ground lightning hazard based on combinations of lightning detection and radar information”. 19th International Lightning Detection Conference (ILDC). Proceedings. Tucson, CD-ROM.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 301-310, Sep. - Dec., 2011

309

Ferro, M.A.S. et al.

Murphy, M.J., Holle, R.L., 2005, “A warning method for the risk of cloud-to-ground lightning based on total lightning and radar information”. International Conference on Lightning and Static Electricity (ICLSE). Proceedings. The Boeing Co., Seattle. Murphy, M.J., Holle, R.L., Demetriades, N.W.S., 2008, “Cloud-to-ground lightning warnings using electric field mill and lightning observations”. 20th International Lightning Detection Conference (ILDC). Proceedings. Tucson.

310

Naccarato, K.P., Pinto Jr., O., Ferreira Jr., H.H., 2008, “Cloudto-ground lightning forecast based on lightning location system information and electric field-mill data”. International Conference on Grounding and Earthing (GROUND 2008) & 3th International Conference on Lightning Physics and Effects (LPE). Proceedings. Florianópolis, Brazil. Naccarato, K.P., Pinto Jr., O., 2008, “Improvements in the detection efficiency model for the Brazilian lightning detection network (BrasilDAT)”. Atmospheric Research. In press. doi:10,1016/j.atmosres.2008.06.019.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 301-310, Sep. - Dec., 2011

doi: 10.5028/jatm.2011. 03033111

João Batista P. Falcão Filho*

Institute of Aeronautics and Space São José dos Campos/SP – Brazil. falcaojbpff@iae.cta.br

Maria Luísa Collucci C. Reis

Institute of Aeronautics and Space São José dos Campos/SP – Brazil. marialuisamlccr@iae.cta.br

Algacyr Morgenstern Jr.

Institute of Aeronautics and Space São José dos Campos/SP – Brazil. algacyramj@iae.cta.br * author for correspondence

Experimental results from the sounding vehicle Sonda III test campaign in the Pilot Transonic Wind Tunnel Abstract: The Pilot Transonic Wind Tunnel of the Institute of Aeronautics and Space has conducted the first test campaign of a sounding vehicle, Sonda III. The campaign is part of a project whose activities and final results are presented in this paper. During the test campaign, many activities were performed to increase the productivity and accuracy of the tunnel. These activities included calibration procedures, corrective and preventive trials, development of auxiliary devices, and theoretical and experimental analysis. Two tasks are described in details: the development and tests performed with the static pressure probe and the automatic re-entry flap actuation system. Several tests were carried out with the Sonda III at Mach numbers ranging from 0.3 to 1.0, at stagnation pressures of 70, 94, and 110 kPa. Experimental results include global aerodynamic coefficients (using internal balance) and pressure distribution over essential regions of the test article (using pressure sensitive paint technique). Keywords: Aerodynamics, Experimental results, Sonda III, Transonic Wind Tunnel.

LIST OF SYMBOLS AND ACRONYMS A: AEB: ALA: AEDC: CD: CD0: CFD: CL: Cm: Cl: Cn: CNPq:

Cross sectional area of fuselage Brazilian Space Agency Aerodynamics Division of IAE Arnold Engineering Development Center Drag coefficient Drag coefficient at zero angle of attack Computational fluid dynamics Lift coefficient Pitching moment coefficient Roll moment coefficient Yawing moment coefficient National Council for Scientific and Technological Development CY: Side force coefficient DCTA: Department of Aerospace Science and Technology EEI: Industrial Engineering College FINEP: Brazilian National Agency for the Financing of Project and Studies IAE: Institute of Aeronautics and Space ITA: Technological Institute of Aeronautics C : Reference length M: Mach number PIC: Programmable microcontroller Received: 10/07/2011 Accepted: 04/10/2011

PSP: Pressure sensitive paint q: Dynamic pressure, defined by 0.5ρ∞ V∞2 TTP: Pilot Transonic Wind Tunnel UNITAU: University of Taubaté UNIVAP: University of Vale do Paraíba USP: University of São Paulo VLS: Satellite Launch Vehicle V∞: Velocity at free stream condition ρ∞: Static density at free stream condition σM: Standard deviation of Mach number INTRODUCTION The TTP of the IAE is a modern installation, built in 1997 and made operational in 2002. The tunnel has a conventional closed circuit and is continuously driven by an 830 kW main axial compressor and an intermittent injection system, which operates in a combined mode, for at least 30 seconds. Its test section is 300 mm wide and 250 mm high, with slotted walls, and it has automatic stagnation pressure controls (from 50 to 125 kPa), Mach number (from 0.2 to 1.3), stagnation temperature and humidity to properly simulate Mach and Reynolds numbers (Falcão Filho and Mello, 2002). Figure 1 shows the operational envelope of the TTP, in which the test capacity of the tunnel in terms of Reynolds numbers related to a typical reference chord of 27.4 mm can be seen. The TTP is a 1/8th scale of an industrial transonic project. It was initially designed to study the innovative features of the industrial facility,

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 311-324, Sep. - Dec., 2011

311

Falcão Filho J.B.P. , Reis M.L.C.C., Morgenstern Jr. A.

mainly concerning with the injection system operation in combination with the conventional main compressor operation. It was also designed to train the technical team to develop basic and academic research in high-speed regimes. Other activities include tests of vehicles with simple geometrical shapes, tests for the development of new aerodynamic transonic profiles, qualitative tests of airplanes with basic configurations, anemometric tests, amongst others. To accomplish this, the tunnel has three sets of six multi-component internal balances manufactured by MicroCraft™ for measuring forces and moments, four modules of 16 pressure channels scanners from PSI™ (2000) for pressure distribution tests, a Schlieren visualization system, hot-wire equipment, and PSP technique to determine pressure distribution over the model surface. In addition, the tunnel possesses a twodimensional probe positioning system, angle of attack remotely controlled system and re-entry flap capability (Falcão Filho et al., 2009). Figure 2 shows a partial view of the aerodynamic circuit of the TTP along with some valves and tubing from the auxiliary systems, and the structural frame in grey, which main compressor limit

Reynolds number (millions)

0.5

T0 = 313 K typical chord = 0.0274 m

0.4

INJECTION structural limit

0.3

blow-off limit

0.2

main compressor turbining limit plenum evacuation limit

0.1 0

sonic throat operation limit

0.2

0.4

0.6 0.8 Mach Number

1.2

1.4

Figure 1. The operational envelope of the TTP.

is used to open the plenum chamber door. Technical details regarding the TTP can be found in Falcão Filho and Mello’s (2002) paper. As a transonic wind tunnel, the TTP is a suitable tool to investigate important effects in the transition range (Goethert, 1961). Particularly for the IAE sounding vehicles, some physical phenomena can be more precisely assessed and used for CFD code comparisons. This resulted in a proposal for a project, which was approved to run from 2007 to 2010. The AEB, by means of the VLS Associated Technology Projects, sponsored the activities of a complete test campaign with the sounding vehicle Sonda III, named “Realização de Ensaios do VS-30 no Túnel Transônico Piloto do IAE” (Tests Development with VS-30 in the IAE Pilot Wind Tunnel). An amount of BRL 271,981.57 was provided for the project, 312

Figure 2. Partial view of the TTP aerodynamic circuit.

whose objective was to achieve know-how about sounding vehicles wind tunnel tests. Out of this amount, 48% were used in the product development, 47% in maintenance and 5% in consumables. The campaign lasted 46 months in contrast to the 41 originally planned, and it used the tunnel for an estimated 1,200 working hours. Thus, the TTP technical team conducted a test campaign to assess the aerodynamic parameters of Sonda III, using an internal balance, and to investigate the pressure distribution using PSP (PSP technique) over important regions of the model, such as the ogive, the inter-stage area, and fins. The Sonda III vehicle is a sounding rocket developed by the IAE. This is a double stage vehicle with a 300 mm diameter second stage, capable of carrying a payload of approximately 100 kg up to an altitude of 600 km. It is one of the sounding rocket families named Sonda, which started with Sonda I, first launched in 1965. Sonda III was launched 27 times between 1976 and 2002. Two models were constructed to allow the installation of an internal balance inside the fuselage: a second stage model (scale 1:11) and a two-stage complete configuration model (scale 1:20). The internal balance was fixed to a sting structure to measure the model’s aerodynamic global parameters: forces (lift, CL, drag, CD and side, CY) and moments (pitching, Cm, roll, Cl and yawing, Cn). General force and moment coefficients were calculated based on literature (Anderson, 2001), by Eq. 1 and 2: C Force =

Force qA

C Moment =

Moment qA C

(1) (2)

where:

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 311-324, Sep. - Dec., 2011

Experimental results from the sounding vehicle Sonda III test campaign in the Pilot Transonic Wind Tunnel

A is a reference area, herein adopted as the maximum cross sectional area of the fuselage, C is a reference length, considered as the maximum fuselage diameter, and q is a reference dynamic pressure, calculated from the free stream parameters (density and velocity), and given by Eq. 3: 1 q = ȡ V 2 2

(3)

Figure 3 shows the scaled models utilized in the experiments, which have configurations of angle of fins equal to 0º, 2.5º and 5.0º.

During the test campaign, many activities were performed to increase the tunnel’s productivity and accuracy of results. The present paper describes the main steps followed in implementing the project and some results obtained. PRELIMINARY TASKS Several procedures were undertaken during the campaign to better adapt the TTP for the proposed tests, which yielded important results that were then incorporated into the installation. Four different categories can be highlighted: (1) calibration procedures for the following: forced extraction mass flow from the plenum chamber, injection system operation with open and closed circuits, six-component internal balances with uncertainty determination, flow quality assessment in the test section, model’s angle of attack setting of the model, and positioning of the re-entry flaps; (2) corrective and preventive maintenance procedures for several components and subsystems to guarantee the perfect operation of the tunnel throughout the campaign, such as the main frequency inverter, auxiliary compressors of injection system, circuit pressure and humidity controls, the cooling tower, the main motor gear lubrication, the auxiliary service air system, and cleaning of the aerodynamic circuit;

(a)

(3) design development of pressure probe with 34 measuring channels to test section uniformity verification, calibration rig for multi-components internal balances loads application, two-dimensional platform for probe positioning, profile rotation device installed on the test section walls, blockage valve installed in the cooling water circuit, items for the pressure sensor scanner calibration, and plenum chamber opening mechanism; (4) carrying out of theoretical and experimental studies to improve the TTP test performances, such as determining the main parameters of the injection system and the polytropic coefficient of the reservoirs, designing a first-throat for Mach number 1.3, determining the operational parameters of the dryer system and the main parameters of the supersonic mixing chamber, as well as numerical simulations of NACA 0012 profile and Sonda III installed in the test section.

(b) Figure 3. Sonda III: (a) second stage in scale 1:11 and (b) complete vehicle in scale 1:20 models.

During the campaign, many scientific studies were presented at conferences, and the TTP team had the technical collaboration of undergraduate students from many universities (ITA, ETEP, UNIVAP, UNITAU, USP-São Carlos), under Scientific Initiation scholarships

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 311-324, Sep. - Dec., 2011

313

Falcão Filho J.B.P. , Reis M.L.C.C., Morgenstern Jr. A.

sponsored by CNPq: Goffert et al. (2008a, 2008b), Zanin et al. (2008a, 2008b), Tagawa et al. (2008), Silva et al. (2009), Goffert and Falcão Filho (2008, 2009), Souza et al. (2009), Reis et al. (2009), Goffert et al. (2010), Silva et al. (2010), Souza et al. (2010), Vargas and Falcão Filho (2010), Costa et al. (2010), Schiavo et al. (2011), Silva et al. (2011), Falcão Filho and Ubertini (2011), Falcão Filho et al. (2011), Goffert et al. (2011). Out of all the preliminary tasks conducted during the campaign, there are three very important ones worthy of mention: the internal balance calibration procedures, which are well documented in Tagawa et al. (2008) and Reis et al. (2008, 2009, 2010); the static pressure probe development; and the automatic re-entry flap actuation system. The last two will be further described.

(a)

According to Davis et al. (1986), the basic criterion to ensure good flow quality in modern transonic wind tunnels is given by the standard deviation of the Mach number over the nominal test section where the model is installed. The accuracy in this area must be periodically verified to guarantee the robustness and repeatability of the tests performed in the installation. The criterion for the standard deviation of the Mach number is described by Eq. 4: (4)

A static pressure probe was developed to assess the Mach number distribution in the TTP. The probe is composed of an ogive tip with an outer diameter of 17.2 mm and a cylinder 1,240 mm in length. The blockage area ratio of the probe is 0.39%. On the surface of the probe, there are 33 static pressure tap stations. Each station is composed of four orifices, which are circumferentially distributed in a cross section area of the cylinder. The four orifices are interconnected, which provide a single static pressure value of the flow at each station location. One tap at the tip of the probe provides the total pressure. Pressure sensors located outside of the plenum chamber supply the pressure readings. Pressure taps holes of 0.5 mm diameter guarantee good accuracy of the pressure readings. The probe wall thickness is 2.1 mm to ensure sufficient rigidity to prevent the probe from bending. Figure 4 shows the probe installed in the test section. Thanks to the rigidity imposed on the design, total bending of only 0.5 mm at the tip was observed, which was practically unnoticeable in the nominal test section region. Figure 5 shows the Mach number distribution along the central line of the test section measured by the static pressure probe. The test section re-entry flaps are closed and there is no forced mass extraction from the plenum chamber.

314

(b) Figure 4. Static pressure probe installed in the test section showing: (a) the attachment to the angle of attack support, and (b) frontal part.

Mach number

|2σM| ≤ 0.001

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

100

200

300

400

500

600

700

800

900

longitudinal distance from test section entrance (mm)

Figure 5. Distribution of the Mach number centerline (shaded area marks the nominal test section region).

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 311-324, Sep. - Dec., 2011

Experimental results from the sounding vehicle Sonda III test campaign in the Pilot Transonic Wind Tunnel

The shaded area marks the nominal test section region and the dashed vertical line marks the test section end. Two measuring stations are located further along this line, in the flap region. Mach number values in this figure were obtained by taking an average value from the test runs data readings over a period of 60 seconds. Mach number deviation observed in the nominal test section region was about 0.002 for most of the operational conditions in the tunnel. It shows that there is good flow quality in the TTP, though still behind the modern state-of-art requirement established by Eq. 4. The test section wall convergent angle and the re-entry flap positioning angle can still be adjusted for better results, which will be carried out in a near future study.

(a)

A remote control system based on stepper motors for the two flap panels (located at each lateral wall just after the test section) was developed to improve the tunnel performance and productivity (Fig. 6a). In the past, the opening of the flap angles was manually adjusted by means of wheels that commanded a worm screw, as seen in Fig. 6b, and it was necessary to stop the test to manually operate the flaps command. As the tunnel circuit is pressurized, this procedure was very tedious and time consuming. Besides, in transonic tunnels, it is important to operate the re-entry flaps remotely to investigate optimum test conditions for each test configuration and model attitude. For example, at high angle of attack during polar movements, adjustment of the flaps may be necessary. The flow chart in Fig. 7 shows the electronic components involved in the project to operate the stepper motors. The central microcomputer (1) located in the TTP control room, which is used to manage and monitor all subsystems actions in the tunnel, commands the flap to move to a determined angle. The signal is sent to the stepper motors (5) through auxiliary circuits (2), programmable microcontrolers named PIC (3) and power drivers (4). The PICs and the power drivers are supplied by a stable voltage source (6). Two hybrid stepper motors (5) were selected to operate in bi-polar and in series mode, since it is more appropriate for the driver utilized, with only two poles and with a lower raged current per phase. The power drivers chosen (4) with high performance receive signals from PICs (3). The signals are: clock, direction of movement and enabling, which command the stepper motors operation. A PIC was chosen because it can be installed very close to the drivers and motors, thereby avoiding electrical noise. PICs are very popular due to their low cost, wide availability, large user base and programming, and reprogramming capability. The microcontroller works as a remote logical unit, which receives data from the

(b) Figure 6. Re-entry flap actuation: (a) modified to remote control actuation, (b) original manual control.

6 Figure 7. Electronic logical data transference for stepper motor control using PIC.

J. Aerosp.Technol. Manag., SĂŁo JosĂŠ dos Campos, Vol.3, No.3, pp. 311-324, Sep. - Dec., 2011

315

Falcão Filho J.B.P. , Reis M.L.C.C., Morgenstern Jr. A.

central microcomputer. Then, the central microcomputer is free to perform other tasks. The PIC was programmed by a code developed in Assembler language, recorded by a Microchip™ unit, in order to send instructions to the drivers. The code contains information to control the stepper motors: the amount of pulses to allow movement, which directions to take and to emit an emergency signal. Since this logic unit is close to the motors, the communication frequency may be higher without electronic noise, thus allowing a faster motor movement. A switched mode power source supplies the PICs and the power drivers. An auxiliary electronic circuit between the microcomputer and the PIC was introduced to control voltage level and optical isolation, for operational safety reasons. The PIC receives instructions from the central microcomputer by means of a program developed in LabView®. This program reports the current positioning of the angle of flap. When the user wants to change this value, a warning signal in yellow displays the value of the flap positioning while the operation is in progress. The original white color is restored when the operation is completed. The conversion between the desired angle and the amount of pulses and motor rotation direction is embedded in this operation as well as a protocol for communication with the PIC. It was necessary to establish a relation between the angle position and the positioning ruler in the shaft. The opening flap angle can be monitored and changed during a tunnel test procedure. Its average speed is 0.7 degrees per second, which is adequate for all tests devised in the tunnel. RESULTS FROM FORCE AND MOMENT The second stage of the Sonda III model was tested for the entire range of Mach number (Reis et al., 2010), even though only hypersonic regimes data are of real interest. However, the results for the complete vehicle at transonic range are very useful due to its flight at low altitude. Eighty aerodynamic polars were planned for the complete model varying the angle of attack from -5º to +5º, for configuration of fins angle at 0º, 2.5º and 5.0º, related to the air direction. The Mach number range was from 0.2 to 1.0 for three stagnation pressure conditions: 70, 94 and 110 kPa. For vehicles like Sonda III, which have jet exits at their rear, it is common to determine the aerodynamic coefficients data reduced to coefficients based on the assumption of zero base drag, i.e., the axial force measured by the balance is subtracted from the model base area multiplied by the pressure difference between the measured value at the model base and at the free stream condition (Pope and Goin, 1978). Figure 8 316

shows the complete model of Sonda III in scale 1:20 installed in the test section of TTP, with one end of the internal balance connected to the model and the other one to the sting, which is fixed to a semi-circular sector of angle of attack positioning. The diagram in the figure clarifies the balance installation. Two pressure taps were installed at the rear part of the model to measure the base pressure, by taking an average of both taps. For all data herein reported, the assumption of zero base drag was applied.

vehicle

Internal balance

Sting

Figure 8. Sonda III complete model in scale 1:20 installed in the test section of the TTP and balance installation scheme.

The forces read by the internal balance had to be reduced to tunnel wind direction to obtain the lift and drag coefficients. Considering that the model is well aligned and the yaw angle is negligible, one can write (Pope and Goin, 1978): FL = FN cosĮ - FA senĮ

(5)

FD = FN senĮ + FA cosĮ

(6)

where: FN and FA are the forces read by the internal balance, in its normal and axial direction, respectively, FL and FD the forces in the normal and axial directions relative to the wind direction, i.e., lift and drag, as they are defined, and α is the model angle of attack.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 311-324, Sep. - Dec., 2011

Experimental results from the sounding vehicle Sonda III test campaign in the Pilot Transonic Wind Tunnel

It is important to observe that the internal balance has small deflections under stress and, during the tests, this effect on the angle of attack measurement was not considered.

All the tests were performed using an internal balance with the following load limits of force: 160 N normal, 16 N axial and side; and moments: 24 Nm pitch, 16 Nm roll and yaw. The balance was calibrated in a rig with dead loads applied to obtain the calibration matrix to be used for data reduction (Reis et al., 2009). The Mach number range was from 0.3 to 1.0, the stagnation pressure was 70, 94 and 110 kPa, and the model was installed in the test section in two different configurations: one with the fins oriented in “+” and the other in “x”, turned 45º in relation to the former. Figure 9 shows the CD variation at zero angle of attack with the model in “+” configuration. Results were as expected, characterized by a constant value for low Mach numbers and growing as it approaches sonic Mach number (for M≥0.8). No appreciable differences can be observed with the variation of stagnation pressure, i.e., with the variation of Reynolds number. Figure 10 shows drag curves as a function of angle of attack for Mach number 0.80 for the model with fins at zero degree and in “+” configuration, with stagnation pressures of 70, 94 and 110 kPa. The general appearance of the curves is quite close to the theoretical prediction, with a curve that has a parabolic tendency. It is observed that the Reynolds number variation (stagnation pressure) has not caused a significant impact on the curves. It can also be seen that the curve for 70 kPa presented some oscillation, because during the tests this low pressure value could not be kept very stable.

CD0

0.6 0.5 0.4

p0=94 kPa p0=110 kPa

0.3 0.2

0.4

0.8

0.6

Mach number

Figure 9. CD at zero angle of attack for “+” configuration with fins at zero degree for stagnation pressures at 94 and 110 kPa. M = 0.80 configuration in “+” 0.60 0.55

Drag coefficient (CD)

Finally, in order to obtain the correct value of aerodynamic forces and moments, many influences should be considered, such as: model support influence, upwash, downwash, buoyancy effects, and model blockage area ratio influence. In a transonic wind tunnel, many of these effects may be diminished by adjusting the flow conditions in the test section, actuating in the wall deflection and in the opening angle of re-entry flaps. In most cases, it is possible to obtain good results in transonic wind tunnels simply by ensuring that the blockage area ratio is kept below 1% (Pope and Goin, 1978). The data presented herein represent the first approach for the aerodynamic loads determination for Sonda III, as no other correction was employed. However, since the blockage area ratio of the model is 0.81% and the tunnel was adjusted to typical operational conditions of flaps opening (10o), reliable results are expected for low Mach numbers (M∞<0.7) and reasonable results for higher speeds.

Configuration in “+”

0.7

0.50

0.45

-6

-4

-2

0.40

p0=94kPa p0=70kPa p0=110kPa 0

Angle of attack

Figure 10. CD for Mach number 0.80, for configuration in “+”, and with fins at zero degree for stagnation pressures of 70, 94, and 110 kPa.

Figure 11 shows a typical result of the variation of the CL with respect to the angle of attack for Mach number 0.5, with stagnation pressure of 94 kPa, fins at 0 degree of deflection and configuration in “+”. The sequence followed in positioning the angle of attack was 0º, 1º, 2º, 3º, 4º, 5º, 4º, 3º, 2º, 1º, 0º, -1º, -2º, -3º, -4º, -5º, -4º, -3º, -2º, -1º and 0º. It is interesting to repeat some angle positions in order to check the repeatability of the tests. The curve shows a likeness to the theoretical prediction (polynomial of third degree) with a hysteresis effect almost nonexistent. In general, the curves for other test conditions showed similar trend. Figure 12 shows a comparison of typical curves of CL, for configuration in “+”, fins at 0 degree and stagnation pressure of 94 kPa. Here again the thirddegree polynomial behavior of the curves is very clear. It is also observed that the curves do not pass

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 311-324, Sep. - Dec., 2011

317

Falcão Filho J.B.P. , Reis M.L.C.C., Morgenstern Jr. A.

CL 1.5

configuration in “+” fins at 0 degree Mach number 0.5

1.0 0.5 0.0

-6

-4

-2

2 4 6 angle of attack (degree)

-0.5

an irregular plane. But, considering the value of 0.06, it is possible to associate the same effect to the normal direction. In this way, a corresponding angle of attack of 0.1 degree causes an increase of about 0.06 in CL. A misalignment of 0.1 degrees represents a total of 0.7 mm linear misalignment between the extremities of the model, which could not be observed by the available alignment device. In Fig. 14 the hysteresis effect is practically undetectable. The same behavior was observed for all other flow conditions.

-1.0 2.0

-1.5

Figure 11. CL for configuration in “+”, fins at 0 degree of deflection, Mach number 0.5, and stagnation pressure of 94 kPa.

M = 0.3 M = 0.4 M = 0.5 M = 0.6 M = 0.7 M = 0.8 M = 0.9 M = 1.0

-5

-3

1.5 1.0 0.5 0.0 0.2 -0.5

CL 1.70 1.20

-1.0

0.70

-1.5

0.20 1-0.30

angle of attack (degree) 1

-0.80

-2.0

CD 0.4

0.3

0.5

0.6

0.7

0.8

M = 0.3 M = 0.4 M = 0.5 M = 0.6 M = 0.7 M = 0.8 M = 0.9 M = 1.0

Figure 13. Polar ‘lift-drag’ for configuration in “x”, fins at 0 degree and stagnation pressure of 94 kPa.

-1.30 -1.80

CY 0.15

Figure 12. CL for configuration in “+”, fins at 0 degree, and stagnation pressure of 94 kPa.

perfectly through the origin, indicating a possible small misalignment of the model relative to the airflow, since the model is axisymmetric. It is common to represent lift and drag in the same graphic, with a variation of the angle of attack. This graphic is called drag polar and, in Fig. 13, polar curves for all Mach numbers, for configuration in “x”, fins angle at 0 degree and stagnation pressure of 94 kPa are represented. All curves related to low Mach number are practically coincident. For Mach number equals and greater than 0.8, there is a noticeable increase of CD. Figure 14 shows the CY, for the “+” configuration, fins at zero degree of deflection, stagnation pressure of 94 kPa, and for angle of attack sweep of: 0º, 1º, 2º, 3º, 4º, 5º, 4º, 3º, 2º, 1º, 0º, -1º, -2º, -3º, -4º, -5º, -4º, -3º, -2º, -1º and 0º. A zero value throughout the range was expected, however a CY of about 0.02 at -5º of angle of attack up to 0.06 at +5º of attack can be observed. This indicates that the angle of attack mechanism travels in 318

0.10 0.05

-6

-5

-4

-3

-2

0.00 -1 0

angle of attack (degree) 1

-0.05 -0.10 -0.15

Figure 14. CY in the “+” configuration, fins at zero degree of deflection, and stagnation pressure of 94 kPa.

Figure 15 shows the average CYs and their standard deviation for all 80 test runs. From run 1 to 24, the model was in configuration in “+” and from 25 to 48 in “x”. No significant variation of CY was observed, indicating that the model’s axisymmetric feature was observed. However, as already observed, the value reflects a misalignment of about 0.7 mm along the model’s body. From runs 49 to 80, the model underwent

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 311-324, Sep. - Dec., 2011

Experimental results from the sounding vehicle Sonda III test campaign in the Pilot Transonic Wind Tunnel

Cn 0.5 0.4

Side force coefficient (CY) Standard deviation of CY

0.12 0.10

0.3 0.2

0.08

0.1

0.06 -6

0.04

-5

-4

-3

-2

0.0 -1 0 -0.1

angle of attack (degree) 1

-0.2

0.02 0.00

Mach number 0.8 Stagnation pressure 94 kPa

40 runs

configuration in “+” configuration in “x”

-0.3 -0.4 -0.5

Figure 15. Average CY and corresponding standard deviation for all 80 runs.

Figure 17. Cl for the “+” and “x” configurations.

several modifications from “x” to “+” and it also had parts with different fin angles changed (Fig. 3b). For these runs, one can observe a little decrease in the average CY but an increase in the standard deviation. It is possible to conclude that the tunnel support had, in fact, a misalignment of about 0.1 degree, which means that repair actions must be carried out to achieve better results.

“+” configuration, the curve presented a typical stable third-degree polynomial curve behavior with angle of attack variation, and crossing the origin of the graph. For the “x” configuration the same third-degree polynomial curve behavior was present (rotated counterclockwise from “+” configuration) and the curve does not pass through the origin. The influence of the fins in the “x” configuration is clear.

A similar investigation was performed for the Cl, as shown in Fig. 16. From runs 1 to 48 the model’s fins were at 0 degrees of deflection, from 49 to 64 at 2.5 degrees of deflection and from 65 to 80 at 5.0 degrees of deflection, as can be clearly seen in the figure. The observed standard deviations were very small for all runs. However, for fins at 0 degrees the average value for some runs was very different from the expected zero value, related to low Mach number (0.3, 0.4 and 0.5), indicating that a more precise internal balance should be used for these low flow conditions.

It is possible to determine an equation to be added to the values from configuration “+” to obtain the values from configuration “x”, given by Eq. 7.

0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0 -0.1 -0.2 -0.3

Cn = -0.05 + 0.03Į

(7)

where: α is the angle of attack. RESULTS FROM PRESSURE DISTRIBUTION

Roll moment coefficient (CI) Standard deviation of CI

Tests at some Mach number conditions were performed to determine pressure distribution over the second stage model, using the PSP technique.

runs

Figure 16. Average Cl and its standard deviation, for all 80 runs.

Figure 17 shows a comparison of Cm for the “+” and “x” configurations for Mach number 0.8 and stagnation pressure of 94 kPa. It is interesting to note that with the

Figure 18 shows an image obtained with PSP technique, for the front region of the model at angle of attack 0 degree and Mach number 1.0. The free stream static pressure is 49.6 kPa and in Figure 18 a graphic of the pressure distribution along the line traced in the PSP image can be seen. In the graphic, it is possible to observe an abrupt pressure decrease just after the cone end. From the local lowest pressure value of 28.0 kPa it is possible to approximately determine the maximum Mach number, which is equal to 1.44 at the expansion. The pressure rises up to the final level of 48.6 kPa at the recovery region, which corresponds

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 311-324, Sep. - Dec., 2011

319

Falcão Filho J.B.P. , Reis M.L.C.C., Morgenstern Jr. A.

1200

Field Pressure

60000

88000

55000

1100

50000

1000

45000

900

35000

800

25000

700

20000

80000

40000

72000

30000

64000 0

500

1000

600

1500

56000 48000

500

40000

400

32000

300 200

24000

100

16000 0

200

400

600

800

1000

1200

1400

1600

Figure 18. Image obtained with PSP technique showing pressure distribution over the model front region surface, at Mach number 1.0 and angle of attack of 0 degree.

to Mach number 1.0. One conventional pressure tap was drilled at the model fuselage, at 148 mm from the nose tip of the fuselage, which was connected to a pressure sensor. The pressure measured by the sensor was 0.230 kPa higher than the value obtained by the PSP technique, representing a very low relative error of 0.5%. Figure 19 shows the same region for all Mach numbers in the range from 0.3 to 1.0. It is interesting to observe that the expansion region grows when the Mach number increases. For free stream Mach number of 0.8, 0.9 and 1.0, the maximum local Mach number values found were 1.17, 1.45 and 1.55, respectively. Figure 20 shows an image of inter-stages regions of the complete model at Mach number 0.7 and angle of attack of zero degree. Two plateaus of pressure values were observed on the fuselage. One at the second stage (pressure about 73.0 kPa), and the other at the first stage (pressure about 71.0 kPa). The plateau in the second stage has a value that is a little higher and grows with the proximity of the inter-stage cone. This can be explained by the cone influence at subsonic Mach number and by the presence of fins. The pressure at the inter-stage cone was greater, as expected, approximately 19%, and the Mach number decreased to 0.46. In the expansion region, the sudden pressure decrease indicates a local Mach number of 320

about 0.93 (indicated by the narrow dark blue color vertical strip in Fig. 20). CONCLUSIONS The main activities related to the first test campaign of the sounding vehicle Sonda III from IAE held in TTP were described. The main tunnel characteristics and model configurations were described, as well as the internal balance used. The preliminary tasks undertaken to the tunnel installation were presented, and important developments were described in more details. They are as follows: a) The pressure probe to assess the flow quality in the test section, showing the longitudinal Mach number distribution for the whole Mach number range; and b) The implementation of automatic control for the re-entry flaps that optimizes the tunnel testing procedures, along with a description of the main features of the design in which a programmable microcircuit is used. Important results of global aerodynamic loads and pressure distribution over the model surface were presented. Since completely accurate procedures of testing corrections for the tunnel were not yet developed, optimum expected conditions were applied to the tunnel controls and good

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 311-324, Sep. - Dec., 2011

Experimental results from the sounding vehicle Sonda III test campaign in the Pilot Transonic Wind Tunnel

M=0.3

M=0.4

M=0.5

M=0.6

M=0.7

M=0.8

M=0.9

M=1.0

Figure 19. Images obtained with PSP technique showing pressure distribution over the model front region surface, at Mach number from 0.3 to 1.0 and angle of attack at 0 degree.

results are envisaged using a model in scale 1/20, with a 0.8% of blockage area ratio. Global load coefficients were determined for Mach number range from 0.3 to 1.0 and stagnation pressure of 70, 94 and 110 kPa, for a complete version of the model, with

configurations of fins at 0º, 2.5º and 5.0º of deflection, and varying the angle of attack from -5º to +5º. The curve of CD0 indicated a CD of about 0.39 from Mach number of 0.3 to 0.7, reaching 0.59 at Mach number 1. The presented curves of CD0, CD × α, CL, and Cm were in accordance with the theoretical predictions. Curves of CY were used

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 311-324, Sep. - Dec., 2011

321

Falcão Filho J.B.P. , Reis M.L.C.C., Morgenstern Jr. A.

85000 80000

1.05e + 00

75000 70000 65000 60000 55000 50000 45000

98000 91000 84000 0

500

1000

1500

77000

600 500

70000

400

63000

300

56000

200

49000

100

42000

0 0

200

400

600

800

1000

1200

1400

1600

Figure 20. Image obtained with PSP technique showing pressure distribution over inter-stage region of the model at Mach number 0.7 and stagnation pressure of 94 kPa, and angle of attack zero degree.

to verify the alignment conditions in the test section. A misalignment of about 0.1 degree was found, which did not compromise the accuracy for the whole campaign, but this fact will be investigated in more details in the future. No appreciable impact was observed with stagnation pressure (Reynolds number) variation. A noticeable change could be seen in the curve of Cm when the model was changed from the “+” to the “x” configurations, due to the aerodynamic influence of the fins. PSP was used to determine the pressure distribution over essential regions of the model, such as the ogive, interstage sector, and fins. The Mach number for expansion regions that is an important feature in wind tunnel testing was determined. A very important step was taken with the conclusion of the present test campaign, enabling the technical team of TTP to proceed with aerodynamic testing in transonic range for models of real interest for the Brazilian aerospace industry. ACKNOWLEDGMENTS The authors are thankful to the AEB for the financial support of project 44 0000 “Realização de Ensaios do VS30 no Túnel Transônico Piloto do IAE”, and to CNPq under grants 103520/2007-4, 101945/2007-8, 103518/2007-0, 101945/2007-8, 100656/2009-9, 103551/2008-5, 113853/2007-6, 104775/2008-4, 119235/2009-9, 119242/2009-5, 102506/2010-8, 106254/2008-1.

322

REFERENCES Anderson, J. D. Jr., 2001, “Fundamentals ofAerodynamics”, McGraw-Hill Higher Education, 3rd Edition, New York. Costa, R. S., Avelar, A. C., Falcão Filho, J. B. P., 2010, “Pressure Measurements Over a NACA 0012 Profile in a Transonic Wind Tunnel Using Pressure Sensitive Paint (PSP)”, Proceedings from 13rd Brazilian Congress of Thermal Sciences and Engineering, Uberlândia, Minas Gerais, Brasil. Davis, M. W., Gunn, J. A., Herron, R. D., Kraft, E. M., 1986, “Optimum transonic wind tunnel,” AIAA 14th Aerodynamic Testing Conference, 14, West Palm Beach, AIAA-86-0756-CP. Falcão Filho, J. B. P., Mello, O. A. F., 2002, “Descrição Técnica do Túnel Transônico Piloto do Centro Técnico Aeroespacial,” Anais ... IX Congresso Brasileiro de Ciências Térmicas e Engenharia, ENCIT-2002, Caxambu-MG, artigo CIT02-0251. Falcão Filho, J. B. P., Avelar, A. C., Reis, M. L. C. C., 2009, “Historical Review and Future Perspectives for the PTT – IAE Pilot Transonic Wind Tunnel,” Journal of Aerospace and Technology and Management, ISSN 1984-9648,Vol. 1, No. 1. Falcão Filho, J. B. P., Ubertini, G. P. A., 2011, “Pressure Probe Development and Tests in a Transonic Wind

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 311-324, Sep. - Dec., 2011

Experimental results from the sounding vehicle Sonda III test campaign in the Pilot Transonic Wind Tunnel

Tunnel Calibration,” proceedings from 21st Brazilian Congress of Mechanical Engineering, Natal, RN, Brazil (accepted for publication). Falcão Filho, J. B. P., Souza, F. M., Oliveira Neto, P. J., Rocha, A., Lima, D. S. A., 2011, “Automatic Control of Flaps in a Transonic Wind Tunnel Installation,” proceedings from 21st Brazilian Congress of Mechanical Engineering, Natal, RN, Brazil (accept for publication). Goethert, B. H., 1961, “Transonic Wind Tunnel Testing”, Pergamon Press, New York. Goffert, B., Tagawa, G. B. S., Zanin, R. B., Reis, M. L. C. C., Falcão Filho, J. B. P., 2008a, “Ensaio de Calibração de Turbina de Inserção do Sistema de Extração Forçada de Massa do Túnel Transônico Piloto do IAE”, Anais ... V Congresso Nacional de Engenharia Mecânica, CONEM-2008, Salvador-Bahia, artigo 0941. Goffert, B., Truyts, C. F., Lima, D. S. A, Falcão Filho, J. B. P., 2008b, “Control of Injection System for the Pilot Transonic Wind Tunnel of IAE in Closed Circuit”, Proceedings ... XII Brazilian Congress of Thermal Engineering and Sciences, ENCIT-2008, Belo Horizonte-MG, article 1-5054. Goffert, B., Falcão Filho, J. B. P., 2008, “Determinação do Coeficiente Politrópico Associado aos Reservatórios de Ar Comprimido do Túnel Transônico Piloto do IAE”, Anais ... V Congresso Nacional de Engenharia Mecânica, CONEM-2008, Salvador-Bahia, artigo 1029. Goffert, B., Falcão Filho, J. B. P., 2009, “Euler Equations Applied to Flow Over NACA 0012”. In: International Congress of Mechanical Engineering, 2009, Porto Alegre - RS. Anais do COBEM 2009 COB09-1136, 2009. Goffert, B., Vargas, M. M., Falcão Filho, J. B. P., 2010, “Verification and Validation of Laminar NavierStokes Applications,” Proceedings from 13rd Brazilian Congress of Thermal Sciences and Engineering, Dec. 05-10, Uberlândia, Minas Gerais, Brasil. Goffert, B., Ubertini, G. P. A., Falcão Filho, J. B. P., 2011, “Design of a Supersonic First-Throat for a Transonic Wind Tunnel and Numerical Evaluation,” proceedings from 21st Brazilian Congress of Mechanical Engineering, Natal, RN, Brazil (accept for publication). Pope, A., Goin, K. L., 1978, “High-Speed Wind Tunnel Testing”, John Wiley & Sons, New York.

PSI, 2000, “ESP-16BP Pressure Scanner User’s Manual,” Catálogo de produto da firma Esterline Pressure Systems, 3rd Edition, Retrieved in June 10th, 2011, from www.pressuresystems.com. Reis, M. L., Castro, R. M., Falcão Filho, J. B. P., Mello, O. A. F., 2008, “Calibration Uncertainty Estimation for Internal Aerodynamic Balance”, Proceedings ... 12th IMEKO TC1-TC7 joint Symposium on Man, Science & Measurement, Annecy, France. Reis, M. L. C. C., Falcão Filho, J. B. P., Paulino, G., Truyts, C., 2009 “Aerodynamic Loads Measurement of a Souding Rocket Vehicle Tested in Wind Tunnel,” XIX IMEKO World Congress, Fundamental and Applied Metrology, September 6/11, 2009, Lisbon, Portugal. Reis, M. L. C. C., Falcão Filho, J. B. P., Mello, O. A., 2010, “Wind Tunnel Tests of the Sonda III Aerospace Vehicle,” Anais do VI Congresso Nacional Engenharia Mecânica, 18 a 21 de agosto, Campina Grande, Paraíba, Brasil. Schiavo, L. A. C. A., Reis, M. L. C. C., Falcão Filho, J. B. P., Truyts, C. F., 2011, “Aerodynamic Tests of the AGARD Model B in a Transonic Wind Tunnel,” proceedings from 21st Brazilian Congress of Mechanical Engineering, Natal, RN, Brazil (submitted). Silva, A. F. C., Braz, R. O., Avelar, A. C. B. J., Falcão Filho, J. B. P., 2009, “Study of the Mach Number Uniformity Over a Horizontal Plane Inside the Test Section of a Wind Tunnel,” In: International Congress of Mechanical Engineering, 2009, Porto Alegre - RS. Anais do COBEM 2009 - COB09-1053. Silva, A. F. C., Ortega, M. A., Falcão Filho, J. B. P., 2010, “Diffuser Design for a Supersonic/Subsonic Mixing Chamber,” Proceedings from 13rd Brazilian Congress of Thermal Sciences and Engineering, Dec. 05-10, Uberlândia, Minas Gerais, Brasil. Silva, A. F. C., Godinho, M. B. C., Ortega, M. A., Falcão Filho, J. B. P., 2011, “Supersonic/Subsonic Mixing Chamber Experimental Analysis,” proceedings from 21st Brazilian Congress of Mechanical Engineering, Natal, RN, Brazil (accepted for publication). Souza, F. M., Falcão Filho, J. B. P., De Oliveira Neto, P. J., 2009, “First Throat Design of a Transonic Wind Tunnel,” In: International Congress of Mechanical Engineering, 2009, Porto Alegre - RS. Anais do COBEM 2009 - COB09-1142. Souza, F. M., Neto, P. J. O, Silva, A. R., Lima, D. S. A., Reis, M. L. C. C., Falcão Filho, J. B. P., 2010,

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 311-324, Sep. - Dec., 2011

323

Falcão Filho J.B.P. , Reis M.L.C.C., Morgenstern Jr. A.

“Metodologia de Calibração de Sistema de Medida de Pressão em Túneis de Vento,” Anais do VI Congresso Nacional de Engenharia Mecânica, 18 a 21 de agosto, Campina Grande, Paraíba, Brasil. Tagawa, G. B. S., Reis, M. L. C. C., Falcão Filho, J. B. P., 2008, “Ajuste de Curva de Calibração de uma Balança Interna Multi-Componente”, Anais ... V Congresso Nacional de Engenharia Mecânica, CONEM-2008, Salvador-Bahia, artigo 1124. Vargas, M. M., Falcão Filho, J. B. P., 2010, “Análise Numérica Bidimensional do Escoamento em Modelo de Veículo de Sondagem,” Anais do VI Congresso Nacional

324

de Engenharia Mecânica, 18 a 21 de agosto, Campina Grande, Paraíba, Brasil. Zanin, R. B., Reis, M. L. C. C., Falcão Filho, J. B. P., 2008a, “Análise da Uniformidade Longitudinal do Número de Mach na Seção de Testes do Túnel Transônico Piloto do IAE em Circuito Aberto”, Anais ... V Congresso Nacional de Engenharia Mecânica, CONEM-2008, Salvador-Bahia, artigo 1031. Zanin, R. B., Braz, R., Avelar, A. C. B. J., Falcão Filho, J. B. P., 2008b, “Analysis of the Drier System of the Pilot Transonic Wind Tunnel of IAE”, Proceedings ... XII Brazilian Congress of Thermal Engineering and Sciences, ENCIT-2008, Belo Horizonte-MG, article 1-5117.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 311-324, Sep. - Dec., 2011

doi: 10.5028/jatm.2011. 03032211

Alexandre Garcia*

Institute of Aeronautics and Space São José dos Campos/SP – Brazil alexandregarciaag@iae.cta.br

Sidney Servulo Cunha Yamanaka

Institute of Aeronautics and Space São José dos Campos/SP – Brazil sidneysscy@iae.cta.br

Alexandre Nogueira Barbosa

Institute of Aeronautics and Space São José dos Campos/SP – Brazil nogueiraanb@iae.cta.br

Francisco Carlos Parquet Bizarria

Institute of Aeronautics and Space São José dos Campos/SP – Brazil bizarriafcpb@iae.cta.br

VSB-30 sounding rocket: history of flight performance Abstract: The VSB-30 vehicle is a two-stage, unguided, rail launched sounding rocket, consisting of two solid propellant motors, payload, with recovery and service system. By the end of 2010, ten vehicles had already been launched, three from Brazil (Alcântara) and seven from Sweden (Esrange). The objective of this paper is to give an overview of the main characteristics of the first ten flights of the VSB-30, with emphasis on performance and trajectory data. The circular 3σ dispersion area for payload impact point has around 50 km of radius. In most launchings of such vehicle, the impact of the payload fell within 2 sigma. This provides the possibility for further studies to decrease the area of dispersion from the impact point. Keywords: VSB-30, Sounding rocket, Trajectory, Performance.

Wolfgang Jung

German Aerospace Center Oberpfaffenhofen – Germany wolfgang.jung@dlr.de

Frank Scheuerpflug

German Aerospace Center Oberpfaffenhofen – Germany frank.scheuerpflug@dlr.de * author for correspondence

INTRODUCTION The VSB-30 vehicle is a two-stage, unguided, rail launched sounding rocket, consisting of a solid propellant S31 rocket booster, a boost adapter, the second stage S30, payload, a recovery and a service systems. Motor and payload are connected by an adapter section and they are separated by pneumatic pistons. The vehicle is designed to fly in a spin stabilized unguided mode. The spin stabilization is achieved by using canted fins. To reduce impact dispersion, the vehicle is equipped with three spinup motors, installed in the booster adapter. The fins are arranged in the standard three-fin configuration and they are nominally set to 18’ (S31) and 21’ (S30), respectively, causing the vehicle to spin from lift-off through burnout. The roll rate at burnout is approximately 3.3 Hz. Total action time for the S31 is 16.0 seconds and 32.0 seconds for the S30. Both motors have a 22-inch diameter (557 mm) (Jung and Gomes, 2006). In 1976, Germany’s national Texus sounding-rocket programme was started and it continues to the present day. Texus (Technologische Experimente unter Schwerelosigkeit) Received: 30/05/11 Accepted: 23/09/11

used British Skylark seven rockets, which provided about six minutes of microgravity. Only recently, in 2005, the Skylark ceased to be commercially available, so that from Texus 42 onwards (November 2005) the Skylark Seven has been replaced by the Brazilian VSB-30 rocket, (Seibert, 2006). In 1996, the German Aerospace Center (DLR) proposed to Comando Geral de Tecnologia Aeroespacial (DCTA) the adaptation of its Mini-TEXUS payload to the first stage of SONDA III, which had its first flight in 1976. This new single stage vehicle was known as VS-30. In 2001, the Unified Microgravity Program for Sounding Rockets proposed to DCTA the development of a boosted version of the VS-30. The challenge was accepted, and the development of the S31 booster motor started. The qualification of the motor consisted of three static firings. The maiden flight was from the Alcântara Launcher Center (CLA) in 2004, (Palmério et al., 2003). The first operational flight from Esrange was in 2005. TEXUS is the European/German sounding rocket programme, which serves the microgravity programmes of the German Aerospace Center (DLR) and of the European Space Agency (ESA). The launches are conducted from Esrange, in Sweden. Figure 1 shows a sequence of pictures from the time of the launch of VSB-30 V07 in 2010 from the CLA.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 325-330, Sep. - Dec., 2011

325

Garcia, A. et al.

Figure 1. VSB-30 being launched from Alcantara Launcher Center.

OBJECTIVES

FLIGHT EVENTS

The objective of this paper is to give an overview of the main characteristics of the first ten flights of the VSB-30, with emphasis on performance and trajectory data. After briefly delving into the history of the VSB-30 sounding rocket in the Introduction, the idea is to give a brief overview of the architecture, the sequence off light events in a typical mission, the list of flown missions, and, finally, the trajectory and flight information for each mission. The final sections of this article deal with the uncertainty in the impact point of the vehicle and the factors affecting it.

The VSB-30 flight events sequence is presented in Table 1 and Fig. 3 (Garcia, 2010).

ARCHITECTURE The basic architecture of VSB-30 is presented in Fig. 2 (Jung, 2006). Payload Recovery system including GPS service module, TV links, experiments, etc.

Table 1. Main events of flight.

Time Altitude Range (s) (km) (km) First stage ignition 0.00 0.051 0.00 Lift-off 0.45 0.056 0.00 Burn-out of 1st stage / separation 13.5 3.617 0.450 Ignition of 2nd stage 15.0 4.323 0.552 Apogee of 1st stage 31.0 6.511 0.970 nd Burn-out of 2 stage 44.0 43.1 7.32 Nose cone ejection 55.0 64.3 11.3 Yo-Yo release 56.0 66.1 11.6 Second stage /payload separation 59.0 71.7 12.7 Impact of 1st stage 106.0 0.00 1.1 Payload apogee 259.0 252.7 81.1 Flight time above 100 km 368 s Drogue chute deployment 492.6 6.1 154.7 Payload impact (without parachute) 497.0 0.00 155.6 Events

LIST OF CAMPAIGNS Second stage Payload adapter, separation ring, conical modules, S30 solid-propellant motor, second stage fin assembly, consisting of tail can and three fins.

First stage Boost adapter, S31 solid-propellant motor, first stage fin assembly, and tail can be equipped with three fins, spin-up motors, etc.

Figure 2. Basic architecture from VSB-30.

326

Table 2 provides the selected information about the first 10 VSB-30 campaigns. IMPACT POINT DISPLACEMENT The impact point showed in Fig. 4 for each flight of the VSB-30 is a statistical displacement of the actual impact point from the predicted one. It is used to calculate the probability of impact within a given distance from the nominal impact point. The distance is referenced to a sigma value. For VSB-30 the radius of 3 sigma is around 50 km. The following procedure was used to determine the actual impact displacement or missdistance for each flight (Bristol Aerospace, 1968).

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 325-330, Sep. - Dec., 2011

VSB-30 sounding rocket: history of flight performance

Figure 3. VSB-30 flight events.

Table 2. List of VSB-30 campaign.

Logo mark

Flight number / payload name and mass

Launch day

Country/ Launch center

Logo mark

Flight number / payload name and mass

Launch day

Country/ Launch center

VSB-30 V05 Texus 44 372.0 (kg)

February 07, 2008

Sweden, Esrange

VSB-30 V06 Texus 45 (367.0 kg)

February 21, 2008

Sweden, Esrange

VSB-30 V01 Cajuana (392.0 kg)

October 23, 2004

Brazil, Alcântara

VSB-30 V02 Texus 42 (372.2 kg)

December 01, 2005

Sweden, Esrange

VSB-30 V07 Maracati II (391.25 kg)

December 12, 2010

Brazil, Alcântara

VSB-30 V03 Texus 43 (407.4 kg)

May 11, 2006

Sweden, Esrange

VSB-30 V08 Maser 11 (383.0 kg)

May 15, 2008

Sweden, Esrange

VSB-30 V09 Texus 46 (393.7 kg)

November 22, 2009

Sweden, Esrange

VSB-30 V010 Texus 47 (362.1 kg)

November 29, 2009

Sweden, Esrange

VSB-30 V04 Cumã II (350.9 kg)

July 19, 2007

Brazil, Alcântara

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 325-330, Sep. - Dec., 2011

327

Garcia, A. et al. 50 40

20 10 0 -10 -20 -30 -40

Down Range Dispersion (km)

30 Legend VSB-30 V01 - Cajuana VSB-30 V02 - Texus 42 VSB-30 V03 - Texus 43 VSB-30 V04 - Cumã II VSB-30 V05 - Texus 44 VSB-30 V06 - Texus 45 VSB-30 V07 - Maracati II VSB-30 V08 - Maser 11 VSB-30 V09 - Texus 46 VSB-30 V10 - Texus 47

-50 -50

-40

-30

-20

-10

Cross Range Dispersion (km)

Figure 4. VSB-30 actual dispersion operational flights.

Step 1 Impact displacement inrange (∆InR) Step 2 Impact displacement in cross range (∆Cr) Step 3 Total impact displacement (∆Total)

InR " I p I a cos( Cr " I a sen(

Total "

FLIGHT PERFORMANCE

)

Figure 5 gives the actual vehicle altitude versus the range for each of the first ten VSB-30 campaigns.

Table 3 gives a comparison between nominal and actual data for Apogee and Ground Range. It also gives the flight time above 100 km and launcher settings values, for all campaigns. The setting and performance differences between each flight are associated with the rocket characteristic and the campaign objective.

InR 2 Cr 2

where: Ip: predicted impact range; Ia: actual impact range; αp: predicted impact azimuth; αa: actual impact azimuth.

300

250 Legend VSB-30 V01 - Cajuana

Altitude (km)

200

VSB-30 V02 - Texus 42 VSB-30 V03 - Texus 43 150

VSB-30 V04 - Cumã II VSB-30 V05 - Texus 44 VSB-30 V06 - Texus 45

100

VSB-30 V07 - Maracati II VSB-30 V08 - Maser 11 VSB-30 V09 - Texus 46

VSB-30 V10 - Texus 47

0 0

100

150

200

250

300

Ground Range (km)

Figure 5. VSB-30 flight performance. 328

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 325-330, Sep. - Dec., 2011

VSB-30 sounding rocket: history of flight performance

Table 3. Flight numerical information.

Apogee (km) Flight number Nominal Actual 01 254.0 236.0 02 257.0 264.4 03 248.5 237.0 04 272.0 250.0 05 270.0 264.4 06 270.0 273.0 07 252.7 242.0 08 258.0 251.8 09 256.1 253.3 10 264.0 264.2

Ground range (km) Launcher setting Nominal Actual Elevation (degree) Azimuth (degree) 188.0 189.0 84.0 45.0 75.0 106.0 87.7 350.0 72.3 74.0 87.7 350.0 208.5 224.7 83.5 65.0 74.0 107.0 87.9 350.0 74.0 78.6 87.9 350.0 155.6 141.0 85.0 60.0 72.0 56.0 87.9 352.5 75.9 104.0 87.8 352.3 75.0 58.3 87.8 352.3

DISPERSION AREA The rocket motions during flight are governed by a number of forces and factors: motor thrust, aerodynamic forces, gravity, wind, wind shear, atmospheric friction, decrease of the rocket mass resulting from the consumption of propellants during flight, the movement of the center of gravity during that process, and forces generated by air rudders and jet vanes (Ordway et al., 2007). The main contributors to the dispersion of impact parts of this type of rockets are manufacturing or assembly misalignments and the wind (James, 1961). Wind dispersion could be made less significant by the use of wind-compensation techniques and automations procedures to set the position of the launcher. In Brazil, the calculation of the adjustment of the azimuth and elevation angles is made for a dedicated software called “Guará” (Yamanaka and Gomes, 2001). This software needs the real-time information about direction and velocity of the wind. The dispersion area is calculated by using rocket parameter error of mass, thrust, aerodynamic, winds, drag of parachute, launcher setting, and so on. The circular 3σ dispersion area for the VSB-30 payload has around a radius of 50 km, using the parameters error listed in Table 4. For that analysis, all perturbation factors are considered. Statistically it implied that the dispersion factors will, in more than 99% of all cases, be less than or equal to the values assumed. Impact displacements, caused by the individual factors, are then combined by the square root of the sum of the squares (of the displacements) to give a 3σ tolerance, (Scheuerpflug, 2009). We can see in Fig. 5 that all impact points for the ten VSB-30 are into the dispersion area. The mathematical method used to calculate the dispersion is the residual sum of square (RSS) by a software called TDA (Garcia and Louis, 2005). COMMENTS AND CONCLUSIONS

Flight time above 100 km (s) 347 380 346 365 380 390 354 365 366 380

Table 4. Error parameters used for VSB-30.

Flight events First stage (S31 burning)

Parameters Error Thrust variation 3% Thrust misalignment in pitch 0.1º Thrust misalignment in yaw 0.1º Aerodynamic drag 20% Weight variation 1% Fin misalignment 0.01º Launcher elevation error 0.5º Launcher azimuth error 3.0º Head wind 2 m/s Cross wind 2 m/s Second stage (S30 burning) Parameters Error Thrust variation 3% Thrust misalignment in pitch 0.1º Thrust misalignment in yaw 0.1º Aerodynamic drag 20% Weight variation 1% Ignition time variation 2s The circular 3σ dispersion area from payload impact point for the VSB-30 has around a radius of 50 km. In most launchings of this vehicle, the impact of the payload fell within 2 sigma. This provides the possibility for further studies to decrease the area of dispersion from the impact point. There are other options to decrease the dispersion area, one of that is an automatic system to set the launcher position in real time, the nearest possible moment of the launching, after any wind perturbation on launcher in the time of the rocket countdown, considering all the terms of the flight security in an unguided sounding rocket campaign (Garcia, 2007).

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 325-330, Sep. - Dec., 2011

329

Garcia, A. et al.

Contributions concerning the possibility of decreasing the impact point dispersion area could minimize many operational problems, such as coats, payload rescue, maritime interdiction area, etc. Other specific objectives of the VSB-30 are: to support the Project of AEB Microgravity allowing organizations, to teach, to research and to develop by performing scientific experiments and technology through suborbital flights; to enhance the partnership with the German Aerospace Center (DLR) in the space-related suborbital launch vehicle and to perform experiments in microgravity.

James, R. L. A., 1961, “Three-Dimensional Trajectory Simulation Using Six Degrees of Freedom with Arbitary Wind. Technical Note D-641”, National Aeronautics and Space Administration. Langley Research Center. Washington – USA. 28p. Jung, W. G., R.M., 2006, “TEXUS 43 – Pre-flight Report. Document number TX43_PFR_1.0”, German Aerospace Center – DLR. Ordway, F. I. III, Dahm, W. K., Konrad, D., Haeussermann, W., Reisig, G., Stuhlinger, E., et al., 2007, “A memoir: From peenemünde to USA: A classic case of technology transfer”, Acta Astronautica, Vol. 60, No. 1, p. 24-47.

Bristol Aeroespace, 1968, “Indoctrination Training Program”, Black Brant IV – SAAP Project. Lecture Course. Part I – Vehicle Course.

Palmério, A.F., da Silva, J. P. C. P., Turner, P., Jung, W., 2003, “The Development of the VSB-30 Sounding Rocket Vehicle”. Proceeding of the 16th ESA Symposium on European Rocket and Ballon Programmes and Related Research, St. Gallen, Switzerland.

Garcia, A., 2010, “Pre-Flight Report VSB30 V07 – MARACATI II. Document Number 528-000000/F4031”, Institute of Aeronautics and Space – Brazil.

Scheuerpflug, F., 2009, “TEXUS 46 – Pre-flight Report. Document number TX46_PFR_1.1”, German Aerospace Center – DLR.

Garcia, A., 2007, “Automation Applied to Sounding Rocket Launcher for Compensation of Wind Influence”, 145 pages. Doctorate Thesis. UNESP – Paulista State University – Campus of Guaratinguetá.

Seibert, G., 2006, “The History of Sounding Rockets and Their Contribution to European Space Research”. ESA Publications Division. Editor: Bruce Battrick. ISSN: 1683-4704. ISBN: 92-9092-550-7.

Garcia, A., Louis, J. E., 2005, “Software Trajectory Dispersion Area – TDA”. Document Number ASE-RT002-2004. Institute of Aeronautics and Space – Brazil.

Yamanaka, S.S.C., Gomes, R.M., 2001, “Launch Pad Setting Calculation – GUARÁ. Technical Report 024/ASEV/01”, Institute of Aeronautics and Space – Brazil.

REFERENCES

330

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 325-330, Sep. - Dec., 2011

doi: 10.5028/jatm.2011. 03033611 Antonio Gil Vicente de Brum*

Universidade Federal do ABC, Santo André/SP – Brazil antonio.brum@ufabc.edu.br

Annibal Hetem Jr.

Universidade Federal do ABC, Santo André/SP – Brazil annibal.hetem@ufabc.edu.br

Israel da Silveira Rêgo

Universidade Federal do ABC, Santo André/SP – Brazil israel.rego@ufabc.edu.br

Cayo Prado Fernandes Francisco

Universidade Federal do ABC, Santo André/SP – Brazil cayo.francisco@ufabc.edu.br

Andre Fenili

Universidade Federal do ABC, Santo André/SP – Brazil andre.fenili@ufabc.edu.br

Fernando Madeira

Universidade Federal do ABC, Santo André/SP – Brazil fernando.madeira@ufabc.edu.br

Preliminary development plan of the ALR, the laser rangefinder for the ASTER deep space mission to the 2001 SN263 asteroid Abstract: The Brazilian deep space mission ASTER, as temporarily named, plans to send a small spacecraft to encounter and investigate the triple asteroid 2001-SN263. The launch is scheduled (initially) to occur in 2015, arriving in 2018. The main motivation of the mission is the development of technology and expertise to leverage the national space sector. Within the scientific goals, the investigation of the still unknown asteroid 2001-SN263. The main project guideline is to aggregate the widest possible Brazilian involvement in the platform, the development and operation of subsystems, integration, payload, as well as in the tracking, navigation, guidance and control of the probe. To meet this guideline, among others, the decision for the development of a laser altimeter in Brazil to fly in the mission was taken. This effort is currently coordinated by a group of researchers from the aerospace engineering personnel of UFABC. This article presents the preliminary development plan for the design of this instrument, which was called ALR (ASTER Laser Rangefinder). Keywords: Deep space mission, Laser altimeter, Laser rangefinder, ASTER, ALR.

Flavio Caldas da Cruz

Universidade Estadual de Campinas, Campinas/ SP – Brazil flavio@ifi.unicamp.br

Marcelo Assafin

Universidade Federal do Rio de Janeiro, Rio de Janeiro/RJ – Brazil massaf@astro.ufrj.br *author for correspondence

THE FIRST BRAZILIAN MISSION TO DEEP SPACE: ASTER

dimensions, masses and consumption in the project of scientific instruments.

The Brazilian space mission ASTER, as it was temporarily named, plans to send a small spacecraft to find, orbit and investigate the still unknown triple asteroid system 2001SN263 for approximately six months. Launch is initially scheduled for 2015 (first launch window), arriving in 2018.

The main motivation of this mission is the development of technology and expertise in order to leverage the national space sector. Thus, the main guideline is to aggregate the widest Brazilian involvement in the platform, the development and operation of subsystems, integration, payload, as well as in tracking, navigation, guidance and control of the probe.

The mission will be conducted from the low-cost FinnishRussian platform MetNet, with initial liquid mass of 152 kg and total mass directed to a 30 kg payload and nominal power of 2.0-2.1 kW. Such restrictions demand reduced Received: 04/0811 Accepted: 22/09/11

In terms of scientific goals, the investigation should bring some indications on the formation of this triple system and each of its main three asteroids will be separately investigated as to dynamic and orbital properties, shape, size, volume, mass distribution, mineral composition, topography and surface texture, gravitational field and rotation speed.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 331-338, Sep. - Dec., 2011

331

Brum, A.G.V. et al.

In order to fulfill these goals, different scientific instruments are planned to compose the payload of the mission. Among them, there are a multispectral camera (wide and narrow-band), a laser altimeter, and an infrared spectrometer. The scientific data will be sent to Earth via telemetry. The references Sukhanov et al. (2010), Araujo et al. (2010) e Winter et al. (2010) provide more details on the ASTER mission and the target system. To meet the project guidelines, the laser altimeter was developed in Brazil, and the engineering coordination of this initiative was a responsibility of the research group of Navigation, Instrumentation and Aerospace Systems (NISA, in Portuguese), comprised of professors of the Aerospace Engineering course at Universidade Federal do ABC. This article presents the preliminary plan to develop this equipment, which is called ALR (ASTER Laser Rangefinder). ALR – THE LASER ALTIMETER FOR THE ASTER MISSION Concept

Due to the high number of functions and the importance of each of them, the laser altimeter is essential for the mission. Scientific Objectives of ALR •

A model of the shape of the main asteroids in the system should be obtained with ≤10 m precision, in terms of spatial resolution (surface) and height (topography), in relation with the CG of the asteroid, in illuminated regions or those with no illumination.

•

The mass of the asteroids should be defined (level of precision to be determined).

•

Gravitational terms J1 and J2, which describe how the mass of the asteroid is distributed, should be determined (level of precision to be determined).

Using the data obtained by ALR •

Obtaining the topographic profiles of the asteroids;

•

Obtaining a model of the global shape of the asteroids;

•

Supporting the study of geodesic parameters of the asteroids (coordinate system, rotation, etc.);

•

Supporting the definition of the orbit and the modelling of gravitational data;

•

Supporting spacecraft maneuvers;

•

Measuring surface roughness and albedo (at laser frequency);

Functions

•

Supporting autonomous navigation (as a sensor) at the approximation stage.

The ALR will contribute with the geodesic and geophysical characterization of the asteroids that are part of the triple system. The data in the equipment will be combined with data from the imaging cameras to gather up topographic features of the targets.

The importance to develop ALR in Brazil

The instrument will measure the length of time for a laser pulse to leave the instrument, reflect on the surface of the target asteroid and return to the instrument. Thus, the distance and relative velocity between the spacecraft and the asteroid system can be precisely measured. A topographic profile of the target corresponding to the course of the beam over the surface will be produced. By interpolation, a model of the global shape of the asteroid will be obtained. From measurements of amplitude and shape of the returning pulse, the reflectivity of the surface, its inclinations and roughness (within the area illuminated by the laser pulse = footprint) will be modeled. The characteristics of the equipment are defined and presented this study, in order to enable the described investigation.

The instrument will also be useful for the navigation in the approximation stages – distance of about 30 km from the center of gravity (CG) of the system – and closest to each of the asteroids (decision of reaching distances <30 km), thus providing precise information concerning the distance of the spacecraft to the target, and the relative velocity between them. Also, ALR will be used to calibrate the infrared spectrometer. 332

The development plan that better meets the technological objectives of the mission and of Brazil, according to the National Plan of Space Activities of the Brazilian Space Agency – PNAE/AEB (PNAE for the decade 2005-2014; www.aeb.org.br) , should prioritize the construction and qualification for the space of scientific equipment and instruments, together with the academy, research institutes and national companies, thus strengthening the national space sector as to the development of new essential technologies and human qualification.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 331-338, Sep. - Dec., 2011

Preliminary development plan of the ALR, the laser rangefinder for the ASTER deep space mission to the 2001 SN263 asteroid

ABOUT THE DEVELOPMENT OF ALR

the equipment can be useful to verify the spacecraft pointing (coarse attitude) and the distance to the asteroids, that should be measured with ±10 m precision.

Team Team of Science Marcelo Assafin – Observatório do Valongo/UFRJ (Coordinator of Science/ALR) Roberto Vieira Martins – Observatório Nacional, RJ Julio Inacio B. Camargo – Observatório Nacional, RJ Team of Engineering (UFABC) Antonio Gil Vicente de Brum (ALR coordinator of Engineering) Annibal Hetem Jr. Cayo Prado F. Francisco Fernando Madeira Israel da Silveira Rêgo André Fenili Laser/detector development Flávio C. Cruz – IFGW/UNICAMP e company BR Labs (based in Campinas) Other partners Álvaro Alberto Cuccolo – OMNISYS Engineering/ UFABC Requirements related to the operation From initialization, the equipment should operate continuously throughout the whole period of the mission investigation (T≤1 year). The equipment should initialize its operation less than 50 km from the target. In this first moment, it may be useful for the spacecraft navigation providing precise information concerning distance and relative velocity in the maneuvers to establish the encounter with the triple system. After the encounter is established (rendezvous), an initial investigation of the system will take place. In this phase,

According to the results of the preliminary investigation, the signal to start the maneuvers and obtain a position that is closer to the system will be given. In this case, the new distance between the spacecraft and the CG of the system will be defined by the mission management after more data on the formation and dynamics of the system are obtained. For now, there are few details about the orbits in this system. Table 1 gathers the available information. According to Winter et al. (2010), the mass of asteroid 1 is much greater than the mass of the other asteroids (M1=1.149x1013 kg; M2=8.0%M1; M3=.6%M1), the mean orbital radius of asteroid 2 around asteroid 1 is ±17 km, which shows that the encounter to a distance much shorter than 30 km cannot occur, unless the satellite asteroids have coplanar orbits. The maximum approximation of the spacecraft in this safer perspective should remain, however, from 20 to 30 km. In this position, ALR can be used to investigate the dynamic characteristics of the system, as well as those related to shape, dimension and topography of the asteroids, with distance measurements with <10 m error. After this investigation, a new approximation can be conducted. The type of approximation will depend on the decision of the mission management. Both the establishment of a closer position, within the rendezvous with the system, and the one obtained from the establishment of an orbit around one of the asteroids are considered at the moment. The first asteroid to be investigated at a higher resolution will possibly be the biggest one (asteroid 1). Studies conducted by Araujo et al. (2010) point to the possibility to establish an orbit around this asteroid with radius between 10 and 13 km. Orbits around asteroid 2 (intermediate mass) with radius of approximately 5 km are also under analysis. In these cases, ALR will be used to map the topographic characteristics of these asteroids with 1 m resolution. Also, at the end of the

Table 1. Orbital and Physical Data.

Asteroid A1 A2 A3

Orbit Sun A1 A1

a1 1.99 AU 17 km 4 km

e1 0.478 * *

I1 6.69o * *

Period1 ~2.8 years ~147 hours ~46 hours

Radius1 1.4 km 0.5 km 0.2 km

Mass2 M1 = 1.15x1013 kg M2 ≅ 7.9x10-2 M1 M3 ≅ 5.7x10-3 M1

*Still not determined; 1Nolan et al. (2008); 2calculations for density = 1.0 g/cm3, estimated density: 1.3 ± 0.6 g/cm3 (Becker et al., 2008); Orbital data a, e, I refer to values of semi-major axis, eccentricity and inclination of the elliptical orbit, respectively. 1 AU (astronomical unit) = mean distance Earth-Sun ≅ 150 x 106 km. SOURCE: Winter et al. (2010).

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 331-338, Sep. - Dec., 2011

333

Brum, A.G.V. et al.

mission, a closer approximation and possible contact with one of the asteroids may occur. Since the mission will not have other devices available to measure the distance to the surface of the target, it is recommended that the operational range of the ALR be expanded to the maximum possible proximity (<50 m). In order to meet these objectives, the equipment should be able to operate at a range that varies from 50 km to less than 50 m, with 10 m and 1 m precision, respectively. In terms of footprint (area that is illuminated by the laser beam over the surface of the target, and depends on the beam divergence and the distance to the target), at initialization and first trigger of ALR (from 40 to 50 km away from the target), beam divergence should provide a return signal of the pulse that was sent, in order to confirm the spacecraft pointing and the proper functioning of the equipment. Within the limits of operation of the equipment (approximately 40 km), the first values concerning the distance to the asteroids can be determined, thus confirming the relative positioning of the spacecraft. The first scientific use of ALR should be about 20 and 30 km away from the main target, asteroid 1. In this case, the beam divergence, α (halfcone angle), should be small, and the power should be enough to obtain a return signal that is related to a 10 to 20 m footprint (radius of the illuminated area). This is sufficient to conduct the scientific analyses related to this approximation stage. For the stage of closest approximation to asteroid 1, about 10 km far, the ray of the footprint should be from 5 to 10 m, which implies a beam divergence angle between 0.028° and 0.057°, in which α=atan (d/D), with d=footprint radius and D=distance to the target; 0.028°=500 µrad. It is also recommended to observe that subsequent studies on the dynamics of the bodies in the triple system should confirm the values of the cited characteristics, related to the transmitter optics. The footprint should also be planned in order to enable the best possible mapping of the target surface. The distance between successive measurements, as a function of the shootings frequency and of the spacecraft-target relative motion (including the target rotation can not exceed a maximum value that would make it impracticable to obtain an improved surface model in terms of horizontal resolution. The distance considered as ideal is the one which results in approximate continuous measurements of the terrain topography.

Thus, the development of modules in Brazil (and their parts) may be conducted as a result of a partnership between the academy and institutions of the national aerospace sector. The initiative is coordinated by the aerospace engineering team of UFABC. Modular division There are three modules to be developed: transmitter, receptor and electronics. The design regarding the parts and the whole system aims to simplify all the items related to the instrument operation. Thus, the design of the LIDAR, which flew with the Hayabusa mission, was the basis for the division proposed here and for the ALR design in general (Tsuno et al., 2006; Hashimoto, Kubota e Mizuno, 2003). Concept of the joint operation The laser electronics controls the laser operation and directs pumping diodes inside the laser head, in which the laser beam is generated. The laser beam is collimated and expanded by the beam expander; then, it leaves the instrument through the exit optics of the transmitter, and goes towards the surface of the target asteroid. The signal reflected by the target is received by the reception telescope (receptor) and focused on the detector plan. The rangefinder has algorithms implemented in the receptor electronics to analyze the return signal. DESIGN GENERAL CHARACTERISTICS •

Integrated design aiming to minimize the size and weight of the instrument.

•

Minimizing telemetry, commands and functions.

•

Using simple and independent optics: transmitter and receptor.

The classic Cassegrain telescope, with optical deposition and a light material mirror, but also hard and resistant (silicon carbide or aluminum).

Obs.: The possibility of integrated optics is investigated (receptor and transmitter).

Design guidelines 1. Reduced configuration (mass, dimensions, power, resource consumption). 334

2. Modular approach to develop the system.

•

Electronics of (development).

low-power

signal

processing

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 331-338, Sep. - Dec., 2011

Preliminary development plan of the ALR, the laser rangefinder for the ASTER deep space mission to the 2001 SN263 asteroid

Telemetry and simple commands: on, off, operation start, end of operation, gain adjustments, etc. The desired characteristics for the instrument translated in terms of design parameters are presented in Table 2. Table 2. Characteristics wanted for ASTER Laser Rangefinder.

Mass Dimensions Power Operational Range Precision Wavelength Pulse-width Repetition rate (frequency) Pulse energy Transmitter’s divergence Life period Footprint: Temperature: Self-calibration during flight

Reduced (<5 kg) Reduced (Max: 37.5 x 23.0 x 35.0) cm3 ≤20 W 50 m≤D≤50 km Vertical: 10 m (D<50 km) 1 m away (D≤10 km) Horizontal: <10 m 1.064 µm To stablish (ns) 1 Hz = >fixated. Should be sufficient for topographic analysis (can be variable) 5 to 16 mJ (depending on the distance, reflectivity, divergence and wavelength) <500 µrad >1 year <10 m (radius; 10 km away) ±2°C (depending on the detector) Yes

Brief description of the parts A preliminary version of ALR and its most important elements is demonstrated in Fig. 1. The most important parts are the transmitter, the receptor and the control unit. The latter is part of the electronics set of the equipment (which includes power sources, not shown here). In Fig. 1, the electric connections are represented by full lines, and optical paths are represented by dotted lines. Generally, ALR will use the time gap between transmission and reception of a laser pulse to calculate the distance between the spacecraft and the asteroid. A new aspect related to the existing equipment is the possibility to use integrated optical transmission and reception. In this case, the function of the optics interchange, from transmitter to receptor, by means of an electro-optical key. However, the choice for independent optics is still more likely. Transmitter and receptor The transmitter will be operated by shooting commands programmed in the control unit of the equipment. Operation can also be externally commanded, via telemetry. Triggering the transmitter or not will depend on the signal sent by the Attitude and Orbit Control Subsystem (AOCS), which can confirm the correct spacecraft pointing. In operation, ALR will send the collected information to a data processing and storage unit. Such data should be used on board (by AOCS, as information concerning Info: Time gap

Receptor

Stop

Interval timer

External commands Gap

Control Unit

Start

Secondary

electrooptical key

Transmitter

Shoot

Primary

Figure 1. Preliminary block diagram of the ASTER Laser Rangefinder program. Here is the version of the study with unified optics. Independent optics is still the first option. J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 331-338, Sep. - Dec., 2011

335

Brum, A.G.V. et al.

navigation and support to the operation of other scientific instruments) and/or sent to telemetry. The laser transmitter consists of a Nd: YAG laser, diodepumped, with a Q-switch, operating at 1.064 µm. The technology of this laser is mature and can produce high energy pulses that meet the criteria of low power consumption and reduced size. The operation in the close infrared gives this laser a wide variety of efficient optics and detectors that are commercially available, which positively influences the decision to use it. The optical diode-pumped laser assures the long-life low power consumption of typically 50,000 hours or up to 18 million 10 pps shooting. A unit of laser diode operating at 808 nm will basically consist of its current source and thermoelectric coolers to stabilize the emission wavelength. Pumping geometry will be linear, including the diode laser attached to the Nd: YAG crystal, two infrared mirrors and a LiNbO3 electro-optic modulator for the Q-switch. The laser source operating in Q-switching mode should be able to produce minimum energy pulses of 10 mJ to 1.064 µm, lasting up to 15 ns with repetition rates varying from 1 to 10 pps, thus producing laser pulses with 1 MW of optic power. The energy of the laser pulse should be high, because, according to Reddy et al. (2008), the 2001SN263 is a C-type asteroid, which means its albedo should be relatively low at 1.064 µm. Besides, the repetition rate influences the horizontal resolution along the beam path over the asteroid surface. However, the commitment between the horizontal resolution, energy consumption and dissipation of heat was used to establish a value for the repetition rate of the order of 1 pps. Beam divergence and its pattern are mainly determined by the geometry of the optical cavity and have an impact on the size of the footprint over the asteroid terrain. The operation of the receptor will be controlled by a temporal gate, which is also internally programmed in the control unit of ALR. Generally, the gate only triggers the receptor within the time gap in which the return pulse is expected. Thus, the time to gather random noise decreases, and the relation signal/noise is maximized. A start signal is sent to the counting clock when the transmitter generates the exit pulse. When the echo or return pulse is received by the receptor, a stop signal is sent to the counting clock. The returning signal is received and sent to the processing unit. The results reveal the distance to the surface, as well as additional data concerning inclination, roughness and albedo, which are stored to be possibly used by AOCS or by other instruments in the spacecraft, and/or to be sent out by telemetry. The registers in the clock are then formatted for the next reading. As to the optics of the instrument, in the version with independent optics, the transmitter optics is simplified 336

and has functions such as collimation and expansion of the exit beam according to the specification of the mission (divergence <500 µrad). A Galilean telescope with proper filters and lens can be used for this purpose. Another possibility that is considered for this moment is unified optics, which consists of coaxial and reflection arrangement with a Schmidt-Cassegrain telescope. A typical diameter for this telescope is around 10 to 12 cm. Both internal mirrors will be made of a hard and resistant material, such as silicon carbide (SiC) or aluminum, the latter being much lighter (however, other properties are not so favorable and need to be investigated, such as high thermal conductivity). The choice for aluminum requires a nonthermal design, like the one used in the laser rangefinder that flew in the NEAR-Shoemaker mission (Cole, 1998). With the coaxial configuration, transmitter and receptor share the same primary optical elements; however, an electro-optical key is necessary, with only one Pockels cell combined with a polarizer. Thus, a selection may be imposed on the state of polarization of the exit and echo pulses. Since the interval between the transmission and reception of the laser pulse can range from 0.3 ms and 66 µs during the phases of the missions, fast commutation will not be required, which makes this technically viable. Electronics According to the guidelines of the mission, it is necessary to optimize resources. In order to meet this objective, the same electronic unit should control both the transmitter and the receptor (integrated electronics). The transmitter electronics includes feeding, triggering and shooting of the laser source, besides the electronics that is dedicated to controlling the operation. The receptor electronics is in charge of its operation, which includes the photo detector element (PD) and the preamplifier, besides the control of the receptor (integrated with the transmitter). This item also detects the pulse reflected by the target (sent by the transmitter), treats the signal, obtains the shape of the received pulse wave, processes this signal, storages and sends data (to other equipment on board or to Earth, via telemetry (use of memory), among others). Internal algorithms After being implemented in electronic units, they first treat the signal, obtain the shape of the wave, analyze the pulse, detect the border, calculate the distance, etc. Algorithms are implemented in the electronic control unit and conduct the processing of the received signal/ pulse, providing the wanted information – calculation of the distance to the target, for example, and other items to

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 331-338, Sep. - Dec., 2011

Preliminary development plan of the ALR, the laser rangefinder for the ASTER deep space mission to the 2001 SN263 asteroid

be conducted in real time to use on board (other devices/ instruments), or to send to Earth. Housing Metal box and fixation of the equipment inside it. The design directions for this element are: - Reduced dimensions (reference: 37.5 x 23.0 x 35.0 cm3); - Fixation with external vibration damping; - Armoring/isolation (thermal and others); -

Interfaces of internal/external communication (input/ output of energy and data);

- Internal/external power sources (total low consumption: ≤20 W); - Heat sinks and thermal control (initial temperature: ±2º C). Integration and tests The integration and validation of the instrument should be conducted under contract by the aerospace company with proper structure and experience in this type of activity. The participation of the Engineering team of UFABC will be in the contract, since it aims to develop the expertise in the field. For this stage, the company OMNISYS Engenharia (São Bernardo do Campo/SP) was contacted. CONCLUSIONS AND FINAL COMMENTS The items to develop ALR for the ASTER mission were defined and presented. The subdivision of this development in parts (modules) and the characteristics of each item and of the whole instrument were previously described, in order to meet the needs and the technological and scientific objectives of the mission. The schedule for the modular development implies the existence of a specific timeline for each of the parts described. The parallel development of the parts should be inside this scenario, minimizing time and optimizing resources. However, the high level of integration and simplification required by the design can only be achieved with more interchange of information among the developing teams. So, the development is centralized in the Engineering group of UFABC. Each item has its own costs spreadsheet. The total estimated cost is compounded by the sum of costs of the

parts previously listed. The total estimated cost of the mission is initially of US$ 35 million. The development will be performed in stages (A, B, C, D). Stage A consists of the creation of the development plan. In the other stages, the models are created and tested. The creation of a “spider” (the setting of unqualified elements or modules on a plate that is different from the final version; it is used to test the module’s functionality), an engineering model, a qualification model, a prototype and a flight model. If possible, the intention is to use the qualification model as a flight model. In this case, there is one less model to make, causing the costs to reduce. At the moment, the “spider” is being created. The ASTER Project has been recently included in the activities of the National Institute for Space Research (INPE). At the moment, the management of the mission wants to include the Project as a strategic activity, aligned with Plano Nacional de Atividades Espaciais (PNAE), to allocate resources related to developments in the space sector. Also, other financing programs (Research and Projects Financing – FINEP and São Paulo Research Foundation etc.) are considered. The use of regular research scholarships is also included (scientific initiation, internships, masters, doctorate, technical development etc.), since the development of expertise in the sector is one of the purposes of the mission. The development model used to create ALR, based on the establishment of multiple partnerships, with personal and institutional qualification, should lead the way to projects concerning the development of the national aerospace sector. This approach to the development of aerospace instrumentation has been used successfully abroad. In Brazil, however, it is new and should contribute much to future developments in the sector. REFERENCES Araujo, R. A. N. de, Winter, O. C., Prado, A. F. B. de A., Sukhanov A., 2010, “O Sistema Triplo de Asteróides 2001SN263: Dinâmica Orbital e Regiões de Estabilidade”. Trabalho apresentado no XV Colóquio Brasileiro de Dinâmica Orbital, Hotel Alpina, Teresópolis, RJ, Brasil. Becker, T., Howell, E.S., Nolan, M.C., Magri, C., 2008, “Physical Modeling of Triple Near-Earth Asteroid 153591(2001 SN263)”. American Astronomical Society, DPS meeting #40, #28.06; Bulletin of the American Astronomical Society, Vol. 40. p.437. Cole, T. D., 1998, “NEAR Laser Rangefinder: A tool for the mapping and topologic study of Asteroid 433 Eros”. JHU/APL Tech. Dig., Vol. 19, No. 2. pp. 142-157.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 331-338, Sep. - Dec., 2011

337

Brum, A.G.V. et al.

Hashimoto, T., Kubota, T., Mizuno, T., 2003, “Light weight sensors for the autonomous asteroid landing of MUSES-C mission”. Acta Astronautica, Vo. 52, No. 2-6, pp. 381-388.

Sukhanov A.A.; Velho, H. F. de C.; Macau, E. E.; Winter, O. C., 2010, “The Aster project: Flight to a near-Earth asteroid.” Cosmic Research, Vol. 48, No. 5, pp.443-450. doi:10.1134/S0010952510050114.

Nolan, M. C., Howell, E. S., Benner, L. A. M., Ostro, S. J., Giorgini, J. D., Magri, C., Margot, J., Shepard, M., 2008, “Planetary Radar Imaging of Binary Asteroids”. Presentation in the Binary Asteroid Dynamics Workshop”, Observatory Meudon, Paris, France, from http://www.asu.cas.cz/~asteroid/paris/add-ons.htm

Tsuno, K., Okumura, E., Katsuyama, Y., Mizuno, T., Hashimoto, T., Nakayama, M., Hashimoto, H., 2006, “Lidar on board asteroid explorer Hayabusa”. Proc. 6th Internat. Conf. On Space Optics, ESTEC, Noordwijk, The Netherlands (ESA SP-621, June 2006).

Reddy, V., Gaffey, M. J., Schaal, M. Takir, D., 2008, “Physical Characterization of First Triplet Near-Earth Asteroid (153591) 2001 SN263”. Asteroids, Comets, Meteors 2008 held July 14-18, 2008. In Baltimore, Maryland. LPI Contribution No. 1405, paper id. 8244.

338

Winter, O. C., Araujo, R. A. N. de, Prado, A. F. B. de A., Sukhanov A., 2010, “A Study of the Orbital Dynamics of the Asteroid 2001 SN263”. Proceedings of the 9th Brazilian Conference on Dynamics Control and their Applications. Serra Negra, SP. ISSN 2178-3667, from http://www.sbmac. org.br/dincon/trabalhos/PDF/minisymposia/68649.pdf

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 331-338, Sep. - Dec., 2011

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

Modeling and simulation of capacitive microaccelerometer

Janderson Rocha Rodrigues Technological Institute of Aeronautics São José dos Campos/SP – Brazil jrr@ita.br Thesis submitted for Masters in Aeronautical and Mechanical Engineering at the Technological Institute of Aeronautics, in 2011. Advisors: Doctors Luiz Carlos Sandoval Góes and Carlos Fernando Rondina Mateus Keywords: MEMS, Microaccelerometer, Modeling, Simulation. Abstract: This work presents the modeling and simulation of a silicon bulk-micromachined capacitive microaccelerometer. The device operating principle is presented as well as its geometric characteristics coming from the microfabrication process. A generalized description is used to identify and subdivide the device physical elements in terms of their functionality. Through this method, the elastic sensing, the capacitive sensing, and the signal conditioning are identified. The static and dynamic analytical models of the sensing elements are presented and simulated using the software Matlab/Simulink®. The validity of each model is discussed and its results related to the device main specifications are shown. The microaccelerometer 3D geometric model is simulated numerically using the software COMSOL Multiphysics®. The numerical results are compared with the analytical and they show an excellent agreement, validating the analytical models.

Production of TiN coatings by EBPVD on titanium alloys obtained via powder metallurgy Eduardo Tavares Galvani Technological Institute of Aeronautics São José dos Campos/SP – Brazil eduardotgalvani@yahoo.com.br

Thesis submitted for Master of Science at the Technological Institute of Aeronautics, in 2011. Adviser: Doctor Vinicius André Rodrigues Henriques Keywords: Coatings, Titanium nitride, Titanium alloys, Powder metallurgy.

EB-PVD,

Abstract: The high strength-to-weight ratio, excellent corrosion resistance, and biocompatibility of the titanium and its alloys are strategic materials to aerospace, chemistry, naval, and surgical implant industries. However, the production and processing of titanium and its alloys are very expensive due to their high reactivity. An alternative is the production of parts by powder metallurgy. The Institute of Aeronautics and Space (IAE) is the main research center of titanium technology in Brazil and in the last ten years it has developed an effective research line, involving titanium alloys production via powder metallurgy techniques. Despite their special properties, titanium and its alloys have poor wear resistance in applications exposed to friction because of the mechanical instability of its passive layer. Thus, the purpose of this dissertation is the development of a physical vapor deposition process using an electron beam furnace (EB-PVD), in order to produce titanium nitride coatings on Ti-13Nb-13Zr substrates obtained by powder metallurgy, which aim to increase their wear resistance. The EB-PVD technique was selected for its ability to obtain hard, thick, and dense multilayer coatings with high adhesion properties. Substrates of Ti-13Nb-13Zr alloy were obtained from the elemental blending of hydrogenated powders, followed by a sequence of cold uniaxial and isostatic pressing and vacuum sintering. The coatings were produced on a twolayer (Ti/TiN) system, in which, initially, a thin layer of pure titanium is deposited to increase the adhesion properties to the substrate. The final layer of TiN is obtained by the combined action of the vaporization cloud from the titanium target against a nitrogen flow-directed to the substrate. The coatings were characterized by scanning electron microscopy, atomic force microscopy, X-ray diffraction, Vickers micro-hardness and tribological analyses including friction, wear, and adhesion tests. The results showed that it is possible to produce TiN coatings with high homogeneity without using special methods of assisted ionization. The coatings have significantly increased the superficial hardness of the substrates, presenting high adhesion and wear properties. It could be observed a clear tendency to increase micro-hardness values with the increasing of the substrate temperature and with the reduction of the nitrogen flow.

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 339-340, Sep. - Dec., 2011

339

Thesis abstracts

Synthesis of polyaniline on a pilot scale for the processing of microwave absorbers with silicone matrix Joseane Mercia da Rocha Pimentel Gonçalves Technological Institute of Aeronautics São José dos Campos/SP – Brazil jomerpi@hotmail.com Thesis submitted for Masters in Aeronautical and Mechanical Engineering at the Technological Institute of Aeronautics, in 2011. Advisors: Doctor Mirabel Cerqueira Rezende Keywords: Polyaniline, Conducting polymers, Radar absorbing materials, Synthesis, Materials engineering. Abstract: The intrinsic electrical conductivity and the low density are properties that make the conducting polymers attractive additives in the development of radar absorbing materials (RAM). In the class of conducting polymers, the polyaniline (PAni) is notable because its production does not require special conditions, and the average cost of reagents is low. However, the commercial availability of this polymer with defined characteristics is still insufficient to attend applications with high demands. Thus, this work aims to study the PAni production in pilot scale, with the maximum repeatability of its physicochemical characteristics, for supporting the RAM processing based on silicone matrix. Results of infrared spectroscopy with Fourier transform, Raman spectroscopy, and X-ray diffraction confirm the success of PAni synthesis in pilot scale, with different samples presenting similar characteristics. Electrical conductivity data of different PAni batches present the same order of magnitude (3.96 to 5.63 x 10-3 S.cm-1). The observed variations of this magnitude are attributed to unintentional changes of temperature of the medium reaction (-3º to 5º C). RAM samples obtained with silicone matrix and the PAni synthesized in the pilot scale showed excellent results in the incident radiation attenuation (>99.9%). Variations in the maximum attenuation with the frequency are matched with the thickness of processed RAM, as it was expected.

340

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 339-340, Sep. - Dec., 2011

AD HOC REFEREES

AD HOC REFEREES Besides the participation of Editorial Board, the Journal of Aerospace Technology and Management had a collaboration of specialists as reviewers to evaluate the manuscripts. To them, the JATM thanks for the contribution in Vol. 3 (2011) Alberto Adade Filho - Technological Institute of Aeronautics - São José dos Campos - Brazil Alberto W. S. Mello Junior - Institute of Aeronautics and Space - São José dos Campos - Brazil Alvaro Jose Damião - Institute for Advanced Studies - São José dos Campos - Brazil Anderson Ribeiro Correia - Technological Institute of Aeronautics - São José dos Campos - Brazil Antonio Carlos da Cunha Migliano - Institute for Advanced Studies - São José dos Campos - Brazil Arcanjo Lenzi - Federal University of Santa Catarina- Florianópolis- Brazil Ariovaldo Felix Palmério - Institute of Aeronautics and Space - São José dos Campos - Brazil Avandelino Santana Jr - Institute of Aeronautics and Space - São José dos Campos - Brazil Bernardo Santos Aflalo - EMBRAER - São José dos Campos - Brazil Carlos Alberto Rocha Pimentel - Universidade Federal do ABC - São Paulo - Brazil Cayo Prado Fernandes Francisco - Universidade Federal do ABC - São Paulo - Brazil Cesar J. Deschamps - Federal University of Santa Catarina - Florianópolis - Brazil Claudio Jorge Pinto Alves - Technological Institute of Aeronautics - São José dos Campos - Brazil Cristina Moniz Araújo Lopes - Institute of Aeronautics and Space - São José dos Campos - Brazil Davidson Moreira - Univerty Center of Patos de Minas - Patos de Minas - Brazil Denise Ferrari - Technological Institute of Aeronautics - São José dos Campos - Brazil Dermeval Carinhana- Institute for Advanced Studies- São José dos Campos - Brazil Elbert E. Nehrer Macau - National Institute for Space Research - São José dos Campos - Brazil Elias B. Teodoro - Federal University of Uberlândia - Uberlândia- Brazil Elizangela Camilo - Institute of Aeronautics and Space - São José dos Campos - Brazil Evandro Moimaz Anselmo - University of São Paulo - São Paulo - Brazil Fábio A. de Almeida - Institute of Aeronautics and Space - São José dos Campos - Brazil Fatima Matiello - National Institute for Space Research - São José dos Campos - Brazil Flávio Donizete Marques - University of São Paulo - São Carlos - Brazil Francis H.R. França- Federal University of Rio Grande do Sul - Porto Alegre- Brazil Gilberto Montibeller Neto - London School of Economics - London- England Humberto Araújo Machado - Institute of Aeronautics and Space - São José dos Campos - Brazil Ijar M. Fonseca - National Institute for Space Research - São José dos Campos - Brazil Ion Georgiou - Getulio Vargas Foundation - São Paulo - Brazil Isabel Cristina dos Santos - University of Taubaté - Taubaté - Brazil Jacques Waldmann - Technological Institute of Aeronautics - São José dos Campos - Brazil Joana Ribeiro - Institute of Aeronautics and Space - São José dos Campos - Brazil João Andrade de Carvalho Jr - State University of São Paulo - Guaratinguetá - Brazil João Batista P. Falcão Filho - Institute of Aeronautics and Space - São José dos Campos - Brazil Jose Eduardo Mautone Barros - Federal University of Minas Gerais - Belo Horizonte - Brazil José Nivaldo Hinckel - National Institute for Space Research - São José dos Campos - Brazil Karl Heinz Kienitz - Technological Institute of Aeronautics - São José dos Campos - Brazil Lais Maria Resende Mallaco - Institute of Aeronautics and Space - São José dos Campos - Brazil Luciene Dias Villar - Institute of Aeronautics and Space - São José dos Campos - Brazil Luis Cláudio Rezende - Institute of Aeronautics and Space - São José dos Campos - Brazil Luiz Carlos Gadelha - National Institute for Space Research - São José dos Campos - Brazil Márcio S. Luz - Department of Aerospace Science and Technology - São José dos Campos - Brazil M. Navabi - Shahid Beheshti University - Teerã - Irã Marcos Daysuke Oyama - Institute of Aeronautics and Space - São José dos Campos - Brazil Maria Luisa Gregori - Institute of Aeronautics and Space - São José dos Campos - Brazil Marisa Roberto - Technological Institute of Aeronautics - São José dos Campos - Brazil Mauricio Guimarães Silva - Institute of Aeronautics and Space - São José dos Campos - Brazil Michelle Leali Costa - State University of São Paulo - Guaratinguetá - Brazil Miriam Kasumi - Institute of Aeronautics and Space - São José dos Campos - Brazil Moarcir Lacerda - Federal University of Mato Grosso do Sul - Campo Grande - Brazil Olympio Achilles de Faria Mello - EMBRAER - São José dos Campos - Brazil J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 330-331, Sep. - Dec., 2011

341

AD HOC REFEREES

Osmar Pinto - National Institute for Space Research - São José dos Campos - Brazil Othon Cabo Winter - State University of São Paulo - Guaratinguetá - Brazil Paula Rahal - State University of São Paulo- São José do Rio Preto - Brazil Paulo Gilberto de Paula Toro - Institute for Advanced Studies - São José dos Campos - Brazil Pedro Paulo Leite de Prado - University of Taubaté -Taubaté - Brazil Pedro José de Oliveira Neto - Institute of Aeronautics and Space - São José dos Campos - Brazil Pierre Kaufmann - Mackenzie University - São Paulo - Brazil Rafael Sfair - State University of São Paulo - Guaratinguetá - Brazil Roberto Gil Annes da Silva - Institute of Aeronautics and Space - São José dos Campos - Brazil Rodolpho Vilhena de Moraes - State University of São Paulo - Guaratinguetá - Brazil Rogeria Eller - Technological Institute of Aeronautics - São José dos Campos - Brazil Rogério Pirk - Institute of Aeronautics and Space - São José dos Campos - Brazil Sandro da Silva Fernandes - Technological Institute of Aeronautics - São José dos Campos - Brazil Sérgio Frascino M. Almeida - Technological Institute of Aeronautics - São José dos Campos - Brazil Silvia Giuliatti Winter - State University of São Paulo - Guaratinguetá - Brazil Takashi Yoneyama - Technological Institute of Aeronautics - São José dos Campos - Brazil Valdemir Carrara - National Institute for Space Research - São José dos Campos - Brazil Vera Lúcia Lourenço - Institute of Aeronautics and Space - São José dos Campos - Brazil Vivian M. Gomes - State University of São Paulo- Guaratinguetá - Brazil Volnei Tita - Technological Institute of Aeronautics - São José dos Campos - Brazil William Roberto Wolf - Stanford University - Palo Alto - USA Wilson F. N. Santos - National Institute for Space Research - Cachoeira Paulista - Brazil Wilson Shimote - Institute of Aeronautics and Space - São José dos Campos - Brazil

342

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 330-331, Sep. - Dec., 2011

Instructions to the Authors Scope and editorial policy The Journal of Aerospace Technology and Management is the official publication of the Departamento de Ciência e Tecnologia Aeroespacial (DCTA – Department of Aerospace Science and Technology), in São José dos Campos, São Paulo State, Brazil. The journal is published three times a year (April, August 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, the author should classify it according to the area selected from the following topics: • Acoustics; • Aerodynamics; • Aerospace Systems; • Applied Computation; • Automation; • Chemistry; • Defense; • Electronics;

• Management Systems; • Materials; • Mechanical Engineering; • Meteorology; • Propulsion; • Structures; • Vibration.

The journal uses the double-blind peer review process for evaluation of the manuscript. The submissions will be evaluated by three Editorial board members or ad hoc referees, and they may be selected for publication according to the editorial policy of the journal.

mandatory requirements all papers must include: title, authors’ names, abstract, and keywords (three to six items that should be based on NASA Thesaurus volume 2 – Access Vocabulary). 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.

contents editorial Any researcher may write the editorial on the invitation of the Editor-in-Chief. Review articles They should cover subjects falling within the scope of the journal. These contributions should be presented in the same format as a full paper, except that they should not be divided into sections, such as: Introduction, Methods, Results, and Discussion. However, they must include a 150 to 200-word abstract, keywords, concluding remarks, acknowledgment, and references. The article should not exceed 20 pages. Technical papers These articles should report the results of original research and they must include: a 150 to 200-word abstract, keywords, introduction, methods, results, discussion, acknowledgments, references, tables and/or figures. This kind of article should not exceed 16 pages. Communications Communication articles should report previous results of ongoing research. They should include a 150 to 200-word abstract, keywords, tables and/or figures, acknowledgments, and references. The communication should not exceed eight pages. Thesis abstracts The journal welcomes Masters and PhD thesis abstracts for publication.

paper submission the manuscript should be written in English and it should electronically submitted. If there is any conflict of interest regarding the evaluation of the manuscript, the author must send a declaration indicating the reasons, in order that the review process occur fairly. See the instructions on: www.jatm.com.br/papersubmission. 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 J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 343-344, Sep. - Dec., 2011

343

of reproduction (printed or electronic) without previous authorization of the Editor-in-Chief. If the reproduction is allowed, it is mandatory to mention the Journal of Aerospace Technology and Management. 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 (IAE)/ATTN: Helena Prado/ Praça Marechal Eduardo Gomes, 50 – Vila das Acácias – CEP: 12228-901 – São José dos Campos/SP, Brazil.

references references should be cited 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 four or more names, use the form “Smith et al. (1997)”. If two or more references would have the same identification, distinguish them by appending “a”, “b”, etc., to the year of publication. Acceptable references include journal articles, numbered papers, books, and submitted articles, if the journal is identified. References from private communications, dissertations, thesis, published conference proceedings and preprints from conferences should be avoided. Self-citation should be limited to a minimum. 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 do not need to appear in the reference list. They should be listed in alphabetical order, according to the last name of the first author, at the end of the article. 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., Ferziger, J. H. and Kline, S. J., 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. 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. EMBRAPA, 1999, “Politics of R&D”, Retrieved in May 8, 2010, from http://www.embrapa.br/publicacoes/ institucionais/polPD.pdf. 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.

Illustrations All illustrations, line drawings, photographs, and graphs should be referred as “Figure” and submitted with good definition (1 to 2 mega pixels). References should be made in the text to each illustration using the abbreviated form “Fig.”, except in the beginning of the sentences. Explanations should be given in the figure legends, so that illustrations are kept clean.

Tables Authors should take notice of the limitations set by the size and layout of the journal. Therefore, large tables should be avoided. All tables must be numbered and mentioned in the text as “Table”.

Equations equations should be typed on individual lines, identified by numbers 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.

Acknowledgments The financial support received for the elaboration of the manuscript must be declared in this item. 344

J. Aerosp.Technol. Manag., São José dos Campos, Vol.3, No.3, pp. 343-344, Sep. - Dec., 2011

Correspondence

General Information

All correspondence should be sent to: Journal of Aerospace Technology and Management (JATM) is a techno-scientific publication serialized,

JOURNAL OF AEROSPACE TECHNOLOGY AND MANAGEMENT

edited, and published by the Department of Aerospace Science and Technology – (Departamento de Ciência

Instituto de Aeronáutica e Espaço (IAE)

e Tecnologia Aeroespacial - DCTA). It contains articles that have been selected by an Editorial Committee

Praça Mal. Eduardo Gomes, 50- Vila das Acácias

composed of researchers and technologists from the scientific community. The journal is published every

CEP 12228-901

four months, and its main objective is to provide an archival form of presenting scientific and technological

São José dos Campos/ São Paulo/Brazil

Contact

research and generate a greater global exchange of knowledge.

Phone: (55)12-3947-5122/5008 E-mail: editor@jatm.com.br

JATM is added/indexed in the following databases; SCOPUS - Elsevier; CAS - Chemical Abstracts Service;

Order your copy (for free): secretary@jatm.com.br

DOAJ - Directory of Open Access Journals; J-GATE - The e-journal gateway from global literature; LIVRE -

Web: http://www.jatm.com.br

Published by: Departamento de Ciência e Tecnologia Aeroespacial Distributed by: Institute of Aeronautics and Space

DOI numbers attributed by CROSSREF.

Desktop publishing and printing: Zeppelini Editorial Edition: 750

Financial support:

São José dos Campos, SP, Brazil ISSN 1984-9648

Historical Note: JATM was created in 2009 after the initiative of the director of the Aeronautics and Space Journal of Aerospace Technology and Management

Institute, Brigadeiro Engenheiro Francisco Carlos Melo Pantoja. In order to reach the goal of becoming a

Vol. 3, n.3 (sep./dec. 2011) – São José dos Campos: Zeppelini Editorial, 2011

journal that could represent knowledge in science and aerospace technology, JATM searched for partnerships

Four monthly issue

1. Aerospace sciences 2. Technologies 3. Aerospace engineering

The copyright on all published material belongs to CDU:629.73

JATMv3n3capa.indd 2

Departamento de Ciência e Tecnologia Aeroespacial (DCTA).

05.12.11 16:44:57

Vol. 3 N. 3 Sep./Dec. 2011

ISSN 1984-9648 ISSN 2175-9146 (online) www.jatm.com.br

Journal of Aerospace Technology and Management V.3, n. 3, Sep./Dec., 2011

Journal of Aerospace Technology and Management www.jatm.com.br

JATMv3n3capa.indd 1

05.12.11 16:44:31