Leonardo Times June 2014 by Anouk Scholtes

JUNE 2014

Leonardo Times Journal of the Society of Aerospace Engineering Students ‘Leonardo da Vinci’

page 26

Dutch Space Interview with Arnaud de Jong, CEO

Shock wave – Boundary layer interactions number 2

Micro-ramp vortex generator

Composite forward swept wing Euler corrections in optimization

RVD: Planetary Exploration Symposium

Year 18

Robotic and manned missions to Mars and Jupiter

cover 0614 wit.indd 1

09-Jul-14 10:58:24

A little more room, a lot more comfort Youâ&#x20AC;&#x2122;ll be amazed what extra legroom and a seat that reclines more can do for your journey. Get comfortable with Economy Comfort, available on all KLM flights. Visit klm.com/economycomfort for more information.

031001136 EC tbv LeonardoTimes_210x297mm.indd 1

21-01-14 09:36

Contents Table of contents

Cover articles

Contents

Editorial

From Leonardo’s desk

Current affairs

Weight reductions for the Airbus A380

Student Project - Eco-Runner Team Delft

Shock wave - Boundary layer interactions

LVD - Unmanned Aerial Vehicles: The next big thing?

Student Project - Forze Hydrogen Racing Team Delft

Composite forward swept wing

Internship Report - Bentley Motors

Interview - Dutch Space

Micro-ramp vortex generator

Residual stress effects on fatigue

RVD - Planetary Exploration Symposium

We vlogen met een zucht... - Denver’s Airport of doom

Shock wave-boundary layer interactions are a very common feature in both transonic and supersonic flows. Because of their major importance on the performance and safety of high-speed flight vehicles, they have now been studied for over 60 years, but their control remains challenging.

Response surface based optimization

Windmilling of Turbofan Engine

Design for air-to-air refuelling operations

Column - Low toxicity rocket propellants

Interview with Arnaud de Jong, CEO

Dutch Space, the largest space company of the Netherlands appointed Arnaud de Jong as the new CEO last year. The Leonardo Times sat down with him to discuss his vision on the future of Dutch Space and his outlook on European and International space business.

AkzoNobel

Heerema

Dutch Space

Thales

EPO

NLR

Fokker

Composite forward swept wing

Multidisciplinary design optimization in aeroelasticity is not a new subject, but in order to reduce the costs, the aerodynamic models are usually very much simplified. What can be gained when one can include a more accurate and complex aerodynamic solver?

KLM

Shock wave – Boundary layer interactions

Euler corrections in optimization

Advertisement index 02

Dutch Space

RVD: Planetary Exploration Symposium

Robotic and manned missions to Mars and Jupiter The 28th Space Department had the honor of organizing the annual symposium of the VSV ‘Leonardo da Vinci’. A total of 564 engineering students and aerospace professionals visited the symposium and witnessed seven extremely inspiring talks and interactive discussions.

JUNE 2014 Leonardo Times

index 0614.indd 3

21-Jul-14 18:57:41

Editor’s letter Dear reader, Welcome to another edition of the Leonardo Times. The cover of the magazine features a radar scan of the west coast of the Netherlands. The image is provided by ESA’s Sentinel-1A satellite, which was launched in April this year. The Sentinel-1A is a part of the ambitious Copernicus programme of the European Commission and ESA. Copernicus programme is the new name for the Global Monitoring for Environment and Security programme (GMES). The Sentinel family of satellites is designed for operational aspects of Copernicus. Sentinel-1A will provide all weather, day and night radar images for land and ocean services. The radar can scan earth’s surface in different modes, which enables it to monitor large areas in low resolution or zoom into a specific area. In the image, radar reflections from the ships at sea are visible, looking like stars in nighttime sky. Also visible are cities of Amsterdam in center-right side and Rotterdam with Europe’s largest port to the left, the city of Delft is also seen close by. The satellite is not yet at full operation capacity and the image is one of its early scans.

company in Netherlands, has delivered the solar panels for the Sentinel 1A. A new technology was developed for this project as solar panels open up after satellite launch. Airborne Composites, has supplied the carbon panels and mechanisms which connect the panels and the satellite have also been manufactured by companies based in Netherlands. We, at Leonardo Times caught up with the Arnaud de Jong, CEO of Dutch Space, the leading space company in Netherlands. On page 26, we feature a candid interview, where we discuss various aspects of Dutch Space, which is a subsidiary of EADS, and the space industry in Europe and beyond.

Year eighteen, number 2, June 2014 The ‘Leonardo Times’ is issued by the Society for Aerospace Engineering Students, the VSV ‘Leonardo da Vinci’, of the Faculty of Aerospace Engineering at Delft University of Technology. The magazine is issued four times a year with a circulation of 5500 copies. EDITOR-IN-CHIEF: Sushant Gupta FINAL EDITOR: Jeroen Wink EDITORIAL STAFF: Bob Roos, Jasper van Gorcum, Joris Stolwijk, Jules L’Ortye, Lakshmi Sabbapathy, Lubi Spranger, Nikita Mahto, Prithvi Penumadu, Raphael Klein, Shahrzad Hosseini.

Students at TU Delft are continuously busy in developing new technologies and in this issue, you can find articles from two student teams active in Delft: Eco-Runner Team and Forze hydrogen racing team. Also featured is a special article from RVD, the space department of the VSV ‘Leonardo da Vinci’. The article covers the Planetary Exploration symposium organized by the RVD this year, which was attended by over 500 aerospace students and professionals.

THE FOLLOWING PEOPLE CONTRIBUTED: Jef Michielssen, Ir. Robert Kroese, Frank Rijks, Rogier Giepman, Ir. Ferry Schrijer, dr.ir. Bas van Oudheusden, Lotfy Hassan, Martin Hartvelt, Marius Tase, Felix Minde, Tom Lotz, Nout van Zon, Ir. Floris Huijbregts, Aditya Ramanathan and MO Li.

I hope you enjoy reading the contributions made by students and professionals in putting together, what we hope, is a stimulating accompaniment for the travels in the summer vacation.

PRINT: DeltaHage B.V., Den Haag

Sushant Gupta

ESA

There is a significant contribution coming from the Netherlands into making of the Sentinel satellites and thus this radar scan of Netherlands. Dutch Space, biggest space

Colophon

COVER IMAGE: Radar image of Netherlands. Credits: ESA DESIGN, LAY-OUT: dafdesign, Amsterdam

Articles sent for publishing become property of ‘Leonardo Times’. No part of this publication may be reproduced by any means without written permission of the publisher. ‘Leonardo Times’ disclaims all responsibilities to return articles and pictures. Articles endorsed by name are not necessarily endorsed editorially. By sending in an article and/ or photograph, the author is assured of being the owner of the copyright. ‘Leonardo Times’ disclaims all responsibility. The ‘Leonardo Times’ is distributed among all students, alumni and employees of the Aerospace Engineering faculty. VSV ‘Leonardo da Vinci’ Kluyverweg 1, 2629 HS Delft Phone: 015-278 32 22 Email: VSV@tudelft.nl ISSN (PRINT) : 2352-7021 ISSN (ONLINE): 2352- 703X For more information the website can be visited at www.vsv.tudelft.nl At this website the ‘Leonardo Times’ can also be digitally viewed. Remarks, questions and/or suggestions can be emailed to the following address: LeoTimes-VSV@student.tudelft.nl

Figure 1. Solar panels for Sentinel 1A have been provided by Dutch Space

Leonardo Times JUNE 2014

Ediorial and Colophon 0614.indd 4

09-Jul-14 13:00:55

FROM LEONARDO’S DESK

Dear readers, On July 18, the VSV ‘Leonardo da Vinci’ will celebrate her 69th birthday. For more than 68 years, Aerospace students from our faculty have been working hard to organize a lot of different activities. During these years, the VSV was able to grow into the society we are today due to the combined efforts of our members, our members of honor and board members. This year we had the opportunity to use these experiences to make our activities even better. One example of using this knowledge was the symposium ‘Planetary Exploration’ on March 4, which was the successor to many great symposia. In 1951, the first symposium was organized by the VSV. Due to the great success of the event, the board decided to organize a symposium every year. For over 60 years, the VSV has been organizing a symposium every year. Approximately 500 students and 50 professionals experienced a day full of interesting lectures of prominent speakers, about the manned and unmanned missions to Mars and Jupiter. The Chairman of the day and our member of honor, Prof. Heinz Stoewer, led us during the day. When looking into the history of our society, not only have our students and their experiences brought the VSV where we are today, but our Members of Hon-

our have also played an emminent role. Therefore, we as the board are proud that we had the opportunity to ask Michel Peters, the CEO of NLR, to be our new Member of Honour. Due to his enthusiasm, assistance with contact opportunities for the Study Tour and participating in the CEO Debating Series, we think Michel Peters is a perfect fit next to our other members of honour. Besides our students, our members of honour, previous board members still play an important role. Each period the board and previous board members sit together to discuss the way the society is functioning at this moment by giving advice. During the current year, we could always ask them for help, when it was needed. This could also be seen during the final week of the third exam period, when the 68th board took over Leonardo’s Desk while we were away for our vacation at Morocco. While the 68th board was controlling the society and welcoming a lot of students to Leonardo’s Desk, we were travelling from Marrakesh through the Atlas Mountains to Casablanca by car. One week to sit back, relax and enjoy all the free time we had. But after a week of drinking sweet mint tea instead of alcohol, driving through the mountains, eating couscous and visiting beautiful places it was time to get back to work to the place we call our home, Leonardo’s Desk.

For almost 69 years, the VSV has been growing thanks to the efforts of our members, members of honour and board members. Every five years, the VSV celebrates these efforts and looks back to the early years of the society during the lustrum-year. Next year, our society will be 70 years old, which will be celebrated by one year of large aerospace-themed events, bigger parties and at the end of the year the VSV Seppe Airshow. To prepare the aerospace students for the 14th Lustrum-year, a pre-lustrum week was planned starting from May 13. During the week, the theme and color of the 14th lustrum were presented. On Wednesday, all our members came together to get to know the new committees in which they can show the world once again what we are capable of. Finally, the week ended with a big party where the 70th board was presented and welcomed by all the members. Finally, it was clear that the week was a good introduction for next year. A year where the VSV will show itself in a different way by organizing bigger events using our members, members of honour and the experiences gathered in history. On behalf of the 69th board of the VSV ‘Leonardo da Vinci’, and with winged greetings, Jef Michielssen, President of the 69th board

JUNE 2014 Leonardo Times

from leonardos desk 2013.indd 5

03-Jul-14 23:50:53

Current Affairs

B787 IS SAFE ACCORDING TO FAA

MIRAGE F1: OUT OF BUSINESS

April 10, Mont-de-Marsan, France

March 19, 2014, Washington D.C., USA

he FAA stated that the design and manufacturing of Boeing’s 787 Dreamliner is safe. It ordered a review of the plane after battery problems grounded the entire fleet last year. The FAA said some of Boeing’s suppliers did not follow industry standards for inspection that include training and testing requirement for inspectors. But it said effective processes had been implemented to correct the issues that arose during and after certification. “For manufacturing, early issues with suppliers implementing the new business processes are being addressed, and improvements are in progress throughout the supply chain,” said the report, issued jointly by the FAA and Boeing’s critical systems review team. The report added that while some design issues had occurred, their causes tended to represent “individual escapes in the design or manufacture of the airplane”. (J.L.) BBC

he end of the Dassault Mirage F1 is near, but the French Air Force does not believe in quick goodbyes. The singleengine fighter was a success for Dassault Aviation. During the design phase, the company’s engineers and designers broke with the traditional delta wing design in order to deliver a low-cost Mach 2 capable fighter with low landing speeds and the ability to use rudimentary runways. Since the maiden flight of the first prototype back in 1966, more than 700 Mirage F1s have been produced with more than 470 exported across the world to Africa, the Middle East, Latin America and Europe. However, the Mirage F1 is a dying breed. Budget cuts in Spain forced their air force to retire the type earlier than planned in June last year leaving just France, Morocco, Libya and Iran as the last users of the aircraft. (J.L.) Aviation Week

INTERNATIONAL DEBUT OF F35B

April 16, London, UK

AIRBUS A320neo: UPGRADE

April 17, New York, NY, USA

hirteen years after full-scale development began, the Lockheed Martin F-35 will make its international debut at the Royal International Air Tattoo and the Farnborough air show in July, the UK Ministry of Defense announced on April 16. Lockheed Martin confirms that the short take-off and vertical landing F-35B is the variant selected to support the two shows in the UK, where the Royal Air Force and Royal Navy plan to operate 138 of the aircraft. The debut appearance in Fairford and Farnborough also is reference to the program’s deep links with UK suppliers and affiliated companies, which build 15% of every F-35, including the aft fuselage assembled by BAE Systems. Additionally, the UK made substantial investments in the programme, ranking only behind the USA among the eight national partners in the development programme. (J.L.)

Penton Media

Flight Global

hen it came to the A320neo, Airbus purposely focused on major propulsion and aerodynamic improvements to gain a maximum performance advantage in a relatively short development time. Now, with the A320neo scheduled to start flight tests in October 2014, industry sources claim that Airbus is shifting its focus to avionics, cabin and other system improvements that could sustain the A320 well into the following decades. According to these sources, suppliers will meet with Airbus later this month to discuss various upgrade packages for the A320neo under an effort dubbed the Single Aisle Incremental Development (SAID) program, also known as the A320neo Plus. The initiative could result in a series of systems and interior upgrades that would start being introduced as of early 2016 onto production aircraft. (J.L.)

Leonardo Times JUNE 2014

current affairs june14.indd 6

09-Jul-14 11:12:58

Current Affairs

LAUNCH OF THE DRAGON CAPSULE

CHINA’S MANNED LUNAR ROVER

April 10, 2014, Chongqing, China

April 18, Cape Canaveral, FL, USA

paceX launched its unmanned Dragon capsule to the International Space Station on Friday, its third trip carrying supplies and equipment to the orbiting lab. After three delays, the Falcon 9 rocket and its Dragon capsule finally blasted off as planned from Cape Canaveral in Florida. About ten minutes after lift-off, the capsule separated from the second stage of the rocket before going into orbit. The capsule deployed its solar panels shortly after, ensuring a continuous power supply. It marks the California-based company’s third commercial resupply mission and fourth visit to the ISS, where it is due to deliver 2.2 tonnes of cargo, including food, new spacesuits, components to replace those aboard the ISS and equipment for 150 scientific experiments. SpaceX has a 1.6 USD billion contract with NASA for a series of future supply missions. (J.L.)

NSA, The Chinese Space Agency has unveiled a prototype of their current conceptual design of a manned moon rover, still subject to change. The rover is able to carry two people, and it can be seen on the right of the image. The left vehicle in the image is a model of China’s unmanned lunar rover Yutu. Although the manned rover seems mostly iron frames and wheels, it is being designed to be foldable, so it fits more easily inside the launcher. Astronauts can build the rover themselves on the moon surface. The wheels on the picture will not be the actual wheels but are dummy wheels made of foam, as the actual wheels are kept classified. The conceptual design is expected to be finished some time in 2015. (B.R.) ECNS

NEW EVIDENCE FOR LIQUID WATER

April 3, 2014, Saturn orbit

Space-Travel.com

EARLY COLLISION AVOIDANCE

April 4, 2014, Darmstadt, Germany

fter launching satellite Sentinel-1A on April 3, ESA received word on the evening of April 4 about a danger of collision with an old NASA satellite named ACRIMSAT. The collision could take place the next morning with an estimated distance of 20m between the satellites. ESA’s crew was initially hesitant to perform an avoidance maneuver because Sentinel-1A was still in LEOP (Launch and Early Orbit Phase). Collision avoidance during LEOP was also never simulated. ACRIMSAT ran out of fuel and it was unable to perform any maneuver, so ESA’s night shift performed the first collision avoidance maneuver ever before the spacecraft payload was actually started up. (B.R.)

assini, NASA’s spacecraft that orbits Saturn, has performed several fly-bys at Saturn’s 504km diameter moon Enceladus over the past years to accurately measure the gravity field of Enceladus. The measurements suggest that there is a low-density rocky core, with an icy outer shell, and denser liquid water in-between at southern latitudes, artistically shown in the image. Cassini spotted organic molecules and water vapor pouring from vents near the south pole of Enceladus in 2005, sparking interest in the moon as a potential location in the search for microbiological life. Together with the recent gravitational measurements, the existence of liquid water on the moon appears likely. Liquid water could exist on Enceladus because of the tidal forces caused by its eccentric orbit around Saturn. (B.R.)

ESA

NASA

JUNE 2014 Leonardo Times

current affairs june14.indd 7

09-Jul-14 11:13:09

INDIADINO, TOULOUSE

WEIGHT REDUCTIONS FOR THE AIRBUS A380 Postbuckling of the A380 VTP skin panels

The skin panels of the Vertical Tail Plane (VTP) are the largest single piece composite components assembled on the Airbus A380. By allowing postbuckling to these skin panels might result in severe weight reductions for the VTP of the A380. The goal of the study is to give an indication of possible weight reductions by allowing postbuckling to the skin panels of the VTP. TEXT Ir. Robert Kroese, Airbus Operations GmbH, A380 VTP Stress

or the present study, a distinction between three different types of buckling has been made; local buckling, global buckling and blade crippling. If local buckling occurs, only local areas of the skin between the stringers buckle (see Figure 1a). This does not result in instability of the panel. In the case of global buckling, the buckles spread over several stringer bays (see Figure 1b). These large deformations result in instability of the panel and the panel will not be able to carry the loads after buckling onset. Blade crippling is local buckling of the stringer blades (see Figure 1c). Due to the crippling of the blades, the stringer is not able to carry the loads after buckling onset and therefore this also leads to instable behavior. Both global buckling and blade crippling are not allowed to occur before Ultimate Load (UL). To set a limit for local buckling onset, the Airbus postbuckling margin policy has been used. This policy gives a limit for local buckling onset depending on the local panel thickness. These limits, ranging from 1.2 Limit Load (LL) up to 1.5LL, are

Weight A380.indd 8

based on empirical data and should make sure that skin-stiffener separation does not occur before UL. PART A: THEORY The focus of the first part is on flat rectangular panels. The first goal is to find the correlation between theory and practice and the second goal is to analyze the influence of important parameters on the postbuckling behavior of the panels. Based on the Von Karman equations (Von Karman, 1910) for large displacements of panels and the method presented in the book of Kassapoglou (Kassapoglou, 2010), an analytical model has been created. The boundary conditions are simply supported and the panels are loaded in unidirectional compression. It is assumed that the layup is balanced and symmetric. Abaqus (Dassault SystĂ¨mes) has been used for the numerical models since Abaqus is known to be good with non-linear FEA. A good correlation was found between the analytical and numerical models for the outof-plane and end-displacements.

As part of a parameter study, the influences of the boundary conditions, element types, mesh sizes, layup directions, panel thickness and aspect ratio have been studied. Part of the parameter study has been used to establish the numerical models. The most important conclusion drawn from the parameter study is that the boundary conditions have a major influence on the postbuckling behavior of (unstiffened) panels. It has a large influence on both buckling onset values as well as the deformation behavior of the panel (both out-of-plane and in-plane). Using the parameter study, the influence of every boundary condition component could be analyzed. Finally, the boundary conditions of the numerical model have been adapted to match the boundary conditions of the test panel. Doing so, a good correlation was found between the FE model and the test panel. Finally, design rules for the layup have been formulated which could be used in the final part.

Leonardo Times JUNE 2014

09-Jul-14 11:25:44

Since no skin-stiffener separation occurred during the panel tests, test data from another component has been used to validate the skin-stiffener separation method. The skin-stiffener separation prediction method is based on the QFC (quadratic failure criterion), combining the interlaminar normal stress and shear stress. Using a very refined mesh and the ASC (average stress criterion) at the stringer run out where the separation occurred during the tests, the point of separation could be predicted quite well. However, the process of refining the mesh size and finding converging stresses using the ASC is very time consuming. Analyzing a test panel of two meters in length and a width of six stringer bays with the refined mesh needed for accurate predictions took over 48 hours of calculation time. PART C: A380 VTP SKIN PANELS To reduce calculation costs, local FE models based on the validated FE models of part B have been used rather than the global VTP model. Three local areas of the actual skin panel have been chosen to be represented by the local models to find weight reductions. A tool has been developed for automatic model generation in which stringer geometry and layup are the parameters, which have been altered during the process. The result files from Abaqus have been analyzed using another automated tool. Using a second order

ROBERT KROESE

Figure 1. A. Local buckling, B. Global buckling, C. Blade crippling

ROBERT KROESE

INDIADINO, TOULOUSE

PART B: TEST The second part has again been split into two parts. First, to validate numerical models using test data and second, validation of a method for skin-stiffener separation prediction. To validate the numerical models, actual test data from the A380 skin panel certification tests were used. This includes the test results of three different 2m panels, which have been tested for pure compression, and three different compression/ shear ratios. An example of an Abaqus model is shown in Figure 2. The complete panel including stringers has been modeled using shell elements and the test rig and support structure was modeled using kinematic couplings and boundary conditions. The strain results from Abaqus have been rotated in the correct direction to determine the membrane and bending strains for all measurement positions in the test area of the panels. A good correlation was found between the test data and the numerical models, especially the sudden steep increase of bending strains after buckling onset was very well predicted by Abaqus.

Figure 2. FE model in Abaqus stiffened panel

derivative method, the buckling onset values for all three types of buckling have been determined. With the values of buckling onset known for the original layup, different layups have been analyzed. Different layups were created, by making the original layups thinner with removal of one or more plies. The modified layups have been analyzed for buckling onset. Based on these results, changes have been made to the stringer geometry to increase or decrease buckling onset values. Once the stability criteria had been fulfilled, the results have been checked for strength and skin-stiffener separation. The process of changing the layup and stringer geometry has not been automated on purpose. In this way, the influence of the changes could be studied and analyzed. By changing the stringer foot width, mostly local buckling onset can be influenced while changing the stringer web height has more influence on the global buckling onset. These results can be used in any further optimization process to modify and optimize the complete skin panel. Based on the results of the local areas, up to 20kg can be saved for both skin panels by allowing postbuckling. FUTURE Besides weight savings, other advantages

are found due to the modified layups. Since the overall panel thickness has been decreased, less time and material is needed to produce the skin panels. Therefore, the skin panels can be produced cheaper and faster. However, this was beyond the scope of this thesis and the economic benefits have not been studied in detail. Since certification regulations are very strict, numerous tests and numerical simulations have to be performed to demonstrate that a fully modified skin panel is able to carry the same loads as before. Since these tests are expensive and very time consuming, postbuckling will not be applied at this moment. However, it still is a very interesting option to apply postbuckling in the future when it can be combined with other major modifications. Special thanks to Dr. Christos Kassapoglou and the Airbus A380 VTP stress team led by Ivano Bertolini and my supervisors Henrik Lüttmann and Sören Ohle. References [1] Karman von, T., “Festigkeitsprobleme in Maschinenbau”, Encycl de Math Wiss, 4: 348-351,1910. [2] Kassapoglou, C., “Design and analysis of composite structures: with applications to aerospace structures”, Wiley, 2nd edition, 2010.

JUNE 2014 Leonardo Times

Weight A380.indd 9

09-Jul-14 11:25:54

Student project

ECO-RUNNER TEAM DELFT ‘The most fuel efficient vehicle in the world’

The slogan, ‘Eco-Runner: the most fuel efficient vehicle in the world’ of the EcoRunner Team Delft says it all: designing and building the most fuel efficient vehicle in the world. The Eco-Runner Team is a ‘D:DREAM Team’ where students from various faculties work together to design and build extremely efficient, hydrogen powered vehicles and conduct research in the fields of aerodynamics, structures, electronics, hydrogen fuel cells and driving strategies. TEXT Frank Rijks, Team Manager, Eco-Runner Team Delft, Student Aerospace Engineering

elft University of Technology takes great pride in its so called ‘D:DREAM’ (Delft: Dream Realisation of Extremely Advanced Machines) teams, and rightly so. As one of these teams, the Eco-Runner Team offers students the opportunity to bring theoretical knowledge obtained during their studies into practice. The Eco-Runner Team was founded in 2006 and since then has built three cars: the Eco-Runner 1, the Eco-Runner H2 and the Eco-Runner 3. This year, the team is building the Eco-Runner 4 where knowledge from various fields of dynamics, structural analysis and aerodynamics finally comes to life in the design and production of this new vehicle. “Actually holding the parts you designed in CATIA in your hands for the first time is a great feeling,” says Pieter (Student, Bachelor Aerospace Engineering), responsible for the suspension of the Eco-Runner 4. CUTTING-EDGE As in the Aerospace industry, the EcoRunner Team strives to build their vehicles

as lightweight as possible, using high-end composite materials for the monocoque and wheels, and high quality aluminum alloys for the suspension. For the Eco-Runner 3, this resulted in an impressive total vehicle mass just shy of 40kg, yet still strong enough to transport a person of 50kg at an average speed of 25km/h. The Ecorunner is yet to be finished while writing this article, but the mass is expected to be reduced by at least 7kg with respect to the Eco-Runner 3. This significant weight reduction was achieved through extensive finite element modeling of the body structure. The team competes in the hydrogen category. Fuel cells converting chemical energy from hydrogen into electrical energy have been around for several decades now, but are gaining immense popularity only in the last few years. Several major car manufacturers (e.g. Hyundai and BMW) have already developed hydrogen powered road vehicles. The Eco-Runner Team sees the hydrogen category as the

greatest challenge and wants to prove the potential and capabilities of hydrogen as energy source for future vehicles. The fuel cell for the Eco-Runner 4 is developed in Germany, at the German Aerospace Laboratory (DLR), in close cooperation with the powertrain team. It is a state of the art fuel cell, able to convert more than 60% of the available energy in hydrogen into useable electrical energy. “Being able to cooperate with such a high tech institution on such a unique, custom product is a pretty cool experience”, says Enzo (Student, Bachelor Electrical Engineering), Chief Electronics. The brushless in-wheel DC motor is capable of achieving efficiencies of over 90%, that leads to the total efficiency of the Eco-Runner 4 (from chemical to kinetic energy) exceed by 50%. The electronics in the Eco-Runner 4 wou be more sophisticated than before. A good data acquisition system has been developed to make sure parameters such as speed, fuel cell voltage and fuel cell cur-

Leonardo Times JUNE 2014

Eco-Runner Team Delft.indd 10

03-Jul-14 23:54:33

MINIMIZING RESISTANCE To achieve extreme mileages, keeping the total resistance to a minimum is crucial. The factors that contribute most to the total resistance are rolling resistance and aerodynamic drag (approximately 50% each).

that, the Eco-Runner Team was able to use the university’s excellent facilities, such as the low speed, low turbulence wind tunnel and the open jet facility, to optimize the aerodynamic design even further. All this effort lead to an approximate total resistance of less than 3N at a velocity of 25km/h. COMPETITION The Eco-Runner Team builds their vehicles to compete at the annual Shell Eco-Marathon. In the European edition of this global competition, almost 200 teams from over twenty countries compete for prizes. In 2013, the team was runner-up in the hydrogen prototype category with a mileage of 287km/kWh (2914km/l of petrol). The vehicles are split up into prototype (very futuristic) and urban concept (‘conventional’ car inspired) vehicles. Teams also have a choice of the energy source they use. A few examples are: petrol, solar power, battery electric and hydrogen. The Shell Eco-Marathon is a large event, challenging young engineers to think about solutions for future mobility. “Even though it is a competition, you can feel

that teams are not only there to win, but also to share ideas and find solutions to common problems together”, team manager Alexander says. In the end, the greatest challenge might be for the drivers. Evelijn (Civil Engineering), who drove the Eco-Runner 3 at the Shell Eco-Marathon 2013, explains: “It can get pretty hectic. A lot of vehicles are on the track together at any given time, and the visibility from the cockpit is pretty limited. Good communication is really important.” FUTURE PROSPECTS The challenge of designing and building the most fuel-efficient vehicle in the world is never ending. In the future new technologies may arise that can be used to improve designs and vehicles. If the Eco-Runner Team sounds challenging and fun to you there is also a possibility to join the team. There are both parttime and fulltime positions and it is also possible to do your minor as part of the team.

ECO-RUNNER TEAM DELFT

To reduce rolling resistance, special lowresistance tyres are used and the mass is kept to a minimum. The aerodynamic drag is reduced by keeping the frontal area small and by shaping the Eco-Runner like an airfoil, increasing the amount of laminar flow over the body and reducing drag. The frontal area of the Eco-Runner 4 is only 0.25m2 (compared to a Renault Twingo series, with a frontal area of 2.315m2). The aerodynamic design is tested and optimized using CFD software that was made available to the team. Besides

Figure 1. The 2012-2013 Eco-Runner core team.

Figure 2. Finding the transition point on the Eco-Runner 4.

ECO-RUNNER TEAM DELFT

Besides analyzing data afterwards, real time data communication is needed to implement the correct driving strategy. Right now, the team is trying to implement this data communication into the Eco-Runner 4. This is a big challenge since the distance that has to be covered by the data is around 600m, with buildings in between the transmitter and receiver, and probably a lot of interference from other signals.

ECO-RUNNER TEAM DELFT

rent can be measured and analyzed afterwards. The analysis of these parameters is important because, besides having an extremely efficient vehicle, the driving strategy of the Eco-Runner 4 during the competition plays a large role as well. In order to find this driving strategy it is important to know how the car and in particular the fuel cell will behave during the race. Together with the measured data, a simulation program, developed by the team, calculates the most efficient driving strategy. The next challenge is to bring this theoretical strategy into practice. A good communication with the driver is therefore vital.

Figure 3. Testing the efficiency of the fuel cell.

JUNE 2014 Leonardo Times

Eco-Runner Team Delft.indd 11

03-Jul-14 23:55:13

SHOCK WAVE - BOUNDARY LAYER INTERACTIONS Flow control by micro-ramp vortex generators

Shock wave-boundary layer interactions are a very common feature in both transonic and supersonic flows. They can be encountered on compressor and turbine blades, in supersonic jet inlets, on transonic wings, on the stabilization fins of missiles and in many more situations. Because of their major importance on the performance and safety of high-speed flight vehicles, they have now been studied for over 60 years, but their control remains challenging. This article presents the results of a study on a new type of flow control device: the micro-ramp vortex generator. TEXT Rogier Giepman, PhD student, Aerodynamics, Dr.ir. Ferry Schrijer, Assistant professor, Aerodynamics and Dr.ir. Bas van Oudheusden, Assosiate professor, Aerodynamics at TU Delft

t was April 9, 1945; Guido Mutke was on a high-altitude training mission with his Messerschmitt Me-262 airplane, when he receives an emergency call from a Luftwaffe colleague under attack by an American P-51 Mustang. He goes full throttle in a steep dive and within a couple of seconds reaches speeds of over 1100 km/h. His plane starts vibrating violently, rivets are popping and all control authority is lost. Somehow, however, using just his horizontal stabilizer, he is able to regain control of his airplane and safely bring it back to solid ground. Although not validated, Guido Mutke might have been the first pilot to break the sound barrier. Back then, the knowledge of transonic / supersonic aerodynamics was still very limited and the drastic changes in flight properties when entering the transonic regime could not be properly explained. In the years after the war it was, however, realized that many of these events could be explained by the formation of shock waves on the (locally) supersonic wings. The appearance of shock waves is, however, not restricted to external surfaces like wings. They can also appear on the internal components of an aircraft: the engine’s inlet, on the fast rotating turbine and compressor blades etc. The physics encountered for all these cases is very similar, however, to reduce the scope of this article, we will restrict ourselves to the example of a supersonic jet inlet, as presented in Figure 1. When entering the engine,

flow control.indd 12

the flow experiences an abrupt change in flow angle and an oblique shock wave is formed that reflects off the inlet’s walls multiple times, up to the point where the flow turns subsonic. In itself, this is not a bad thing, the flow needs to be decelerated before entering the compressor stage of the engine, and oblique shock waves are well suited for this job. The shock reflections on the other hand can be problematic, because of the boundary layer developing along the walls of the inlet. Without the presence of this boundary layer (inviscid flow), the shock wave would reflect from the same location as where it hits the surface (see Figure 1). However, in any real life application, there will be friction / viscosity and consequently also an interaction between the oblique shock waves and wall boundary layer. Figure 2 zooms in on this interaction and shows the typical flow topology that is encountered for an oblique shock wave reflection. As is known, disturbances in a flow travel with the speed of sound, therefore they can only travel downstream in a supersonic flow. The incoming freestream and the supersonic part of the incoming boundary layer are, therefore, not aware of the presence of the incoming shock wave. However, close to the wall, the boundary layer is still subsonic, which allows for information to be transferred in the upstream direction of the flow. Via this route, the incoming boundary layer is ‘warned’ that there

is a region of strong adverse pressure gradient (caused by the shock) ahead. In response to this information, the boundary layer will thicken even before reaching the incident shock wave. The supersonic part of the boundary layer is therefore turned into itself and compression waves are created, which usually converge just outside of the boundary layer to form the reflected shock wave. Depending upon the strength of the incident shock wave and the stability of the incoming boundary layer, the boundary layer may separate and an expansion fan and reattachment shock will also be formed. Comparing Figures 1 and 2, clearly simple inviscid flow theory does not provide an adequate description when considering the flow behavior close to the wall. To complicate matters even further, a shock wave-boundary layer interaction with separation is also inherently unstable. The separation bubble will be growing / shrinking over time and as a consequence the reflected shock wave, expansion wave and reattachment shock wave will be moving as well. Shock instability is a wellknown cause of air-intake buzz, which leads to large fluctuations in the thrust output and if not handled properly by the pilot can result in the engine to unstart. Due to the major impact on the performance and safety of high-speed flight vehicles, shock wave-boundary layer interactions have been an active field of

Leonardo Times JUNE 2014

09-Jul-14 11:36:13

ROGIER GIEPMAN

WWW.PBASE.COM ROGIER GIEPMAN

HOLGER BABINSKY (UNIVERSITY OF CAMBRIDGE) & ZHENGZHONG SUN (TU DELFT). MODIFIED BY ROGIER GIEPMAN

WWW.SR71.ORG

ROGIER GIEPMAN

Figure 1. Inlet of a Pratt & Whitney J-58 engine, as used on the SR-71 Blackbird

Figure 3. The purpose of flow control is to increase the ‘health’ of the boundary layer by adding more momentum to the near-wall region of the flow, thus creating a fuller profile.

Figure 2. Flow topology encountered for an oblique shock wave reflection, for the case of a separated boundary layer

research for over 60 years now. And in this period of time, a large body of data has been gathered and a great deal has been learned about the mean and instantaneous flow behaviour of shock waveboundary layer interactions. The effective control of shock wave-boundary layer interactions, however, still remains a very active and open topic. With the development of new experimental and numerical techniques, it is possible to gain a better understanding of the physics involved in shock wave-boundary layer interactions than ever before. At the aerodynamics department of the TU Delft, we have mostly focused our efforts on the experimental side of the problem and approached it using a variety of measurement techniques, with Particle Image Velocimetry (PIV) serving as our main flow diagnostics tool. PARTICLE IMAGE VELOCIMETRY The basic idea of PIV is simple: small tracer particles are added to the flow and a camera is used to track the movement of the particles that travel with the flow. Two images are acquired in short succession (the time delay is typically Δt ~ 1μs in supersonic flows), such that the same particles are recorded in both images although shifted by a small amount. Using computer algorithms, the particle movement from the first to the second image can be obtained. Since the time separation between both images is known, the velocity of the particle and thus of the flow can be computed. To capture these tiny and fast moving particles on film, a large amount of light

Figure 4. Flow development downstream of a micro-ramp vortex generator, showing a conceptual model (Babinsky, 2009) and the instantaneous vortical structures measured by Sun using a tomographic PIV system (Sun, 2012).

is required. This is delivered by a special laser that can produce two high-energy pulses for the given time separation Δt and ‘freezes’ the location of the particles in both frames. FLOW CONTROL DEVICE The goal of flow control for shock waveboundary layer interactions is to reduce the size of the separation bubble and the associated unsteadiness of the interaction region. The size of the separation bubble depends on two critical factors: the shock strength and the ‘health’ of the incoming boundary layer. Weaker shocks would reduce the size of the separation bubble, but unfortunately reducing the shock strength is not always an option. For instance, in supersonic jet intakes, a series of oblique shocks is used to decelerate the incoming flow to a subsonic Mach number. Weaker shocks can be used, but then to reach the same compression ratio, more shocks are needed and consequently also a longer / heavier inlet is required. The shock strength is therefore usually a given, which follows from the early stages of the design process and does not allow for too much tweaking. Therefore, the option that remains for flow control is to manipulate the ‘health’ of the boundary layer. It is well known that fuller boundary layers (see Fig.3) are less prone to separation. Full in this context means that there is high-momentum fluid present close to the wall. In traditional supersonic jet inlets, a full boundary layer profile is usually obtained by means of boundary layer bleed. Slots are introduced in the

wall, which remove the low-momentum portion of the boundary layer close to the wall from the flow (see also Fig.1). The new boundary layer that forms is much fuller and less prone to separation. The mass flow removed from the inlet (~2% of the capture inlet mass flow) is typically not reinjected elsewhere in the engine and is vented overboard. So, to achieve the same mass flow rate through the engine, the frontal area of the engine needs to be increased, which also increases the engine’s weight and drag. Micro-ramp vortex generators form a promising alternative to boundary layer bleed. A micro-ramp is a small wedgelike ramp, with a typical height of half the boundary layer thickness (see Fig.4). When placed in a boundary layer, two counterrotating vortices are formed downstream of the micro-ramp that transport high-momentum fluid from high up in the boundary layer down towards the wall. They act as mixing devices and create a boundary layer profile that is fuller and less prone to separation. Compared to boundary layer bleed, this system is more robust and does not reduce the mass flow, therefore allowing a smaller engine intake. However, before these devices can be used on an actual engine, some fundamental questions need to be answered: 1. Can the size of the separation bubble / interaction unsteadiness be reduced, and if so, by how much? 2. What is the optimal location of the micro-ramp with respect to the shock system? 3. What are the effects of Mach and Reyn-

JUNE 2014 Leonardo Times

flow control.indd 13

09-Jul-14 11:36:16

ROGIER GIEPMAN

Figure 6. Flow field overview for the case of a micro-ramp (h=3 mm) mounted 108 mm upstream of the wall impingement location of the incident shock.

ROGIER GIEPMAN

Figure 5. Test section of the ST-15, showing the micro-ramp, the 12˚ shock generator and two cameras recording the flow.

Figure 8. The separation probability Psep with a micro-ramp (h = 4 mm, x = -17.3δ) upstream of the interaction (a) and without a micro-ramp (b).

olds number of the micro-ramp’s effectiveness? 4. How severe are the 3D effects introduced by the ramp? This article touches upon the first question, for more information on the first and second question please refer to [3]. The third question is currently being investigated by an MSc Student from the Aerodynamics department (Aabhas Srivastava) and the fourth question still remains to be investigated. RESULTS The measurements were performed in the ST-15 supersonic wind tunnel of the TU Delft at a Mach number of 2.0 and a freestream velocity of U∞ = 524m/s. The oblique shock wave is created by a 12˚ wedge spanning the entire test section (see Fig.5). Two cameras were used for the measurements; one was focused on the shock wave-boundary layer interaction and the other on the flow between the micro-ramp and the interaction. The tunnel wall boundary layer is approximately δ = 5.2mm thick and micro-ramp heights of 2, 3 and 4mm were tested. In Fig.6, the data from both cameras was stitched together to create the average flow field from micro-ramp to interaction. For the particular case presented, a microramp height of 3 mm was used and measurements were performed in the symmetry plane of the ramp. The region close to the micro-ramp is blanked, because laser reflections were too strong to allow for an accurate measurement of the seeding particles. Behind the ramp, a wake can be distinguished which lifts off the surface and weakens when moving downstream. Around x = −20 mm, the wake interacts with the shock system, and a low-velocity

flow control.indd 14

Figure 7. Three uncorrelated snapshots of the low-speed bubble for a 4 mm micro-ramp located at x/δ = −17.3 (x = -90 mm). The solid black line presents the zero-velocity iso-line. (a) No separation (b) Small separation bubble (c) Large separation bubble.

pocket is formed. Also clearly visible is the low-speed bubble that is formed at the wall and which is the result of the strong adverse pressure gradient imposed by the shock system. On average there is no separation taking place for this particular configuration, however, instantaneously there can still be separated regions. The unsteady behaviour of the low-speed bubble is presented in Fig.7, which shows three instantaneous snapshots of the flow in the interaction region. The solid black line presents the zero-velocity isoline. For Fig.7 (a) no reversed flow is detected, for 7(b) a small pocket of 6 mm2 is present and for 7(c), a large pocket of 44mm2 is measured. Given the unsteady nature of the interaction, it makes sense to talk about the separation probability Psep of the flow, which is defined as the probability that a certain point (x,y) shows reversed flow. For example, when Psep(x,y) equals 50%, then half of the time the flow is reversed and the other half of the time the flow it is attached at the location (x,y). Fig.8 compares the separation probability with (a) and without a micro-ramp (b). Without a micro-ramp, flow reversal is observed in a region spanning the entire field of view, which is approximately 9δ. By placing a 4 mm micro-ramp 17.3δ upstream of the interaction, this is reduced to 5.5δ. The probability of encountering reversed flow is also reduced. Without a micro-ramp, there are regions close to the wall that shows flow reversal 75% of the time. With a micro-ramp, this peak value has been reduced to 41%. The smaller separation bubble also translates into a reduction of the reflected shock unsteadiness by ~50%. These measurements highlight the great potential that micro-ramp vortex genera-

tors can have as flow control devices for shock wave-boundary layer interactions. However, as mentioned before, there are still a number of steps to take before these ramps can find their way into the engine of a jet fighter. Up to this point, only planar PIV measurements have been performed on the interaction region. The flow structures introduced by the ramp are, however, inherently three-dimensional and the same holds for the separation bubble. In the near future, tomographic PIV measurements will be carried out on the interaction region, which will deliver the velocity field inside an entire volume instead of just a plane. In case you are interested / would like to be involved with these measurements, please contact one of the authors of this article. CONTACT INFORMATION Rogier Giepman – R.H.M.Giepman@tudelft.nl

References [1] H. Babinsky, Y. Li, and C. W. Pitt Ford, “Microramp Control of Supersonic Oblique Shock-Wave/Boundary-Layer Interactions”, AIAA Journal, 47(3):668– 675, March 2009 [2] Z. Sun, F. F. J. Schrijer, F. Scarano, and B. W. van Oudheusden, “The threedimensional flow organization past a microramp in a supersonic boundary layer”, Physics of Fluids, 24:1–22, 2012. [3] R.H.M. Giepman, F.F.J. Schrijer, and B.W. van Oudheusden, “Flow control of an oblique shock wave reflection with micro-ramp vortex generators: Effects of location and size”, Phys. Fluids 26, 066101, 2014

Leonardo Times JUNE 2014

09-Jul-14 11:36:22

GREAT IDEA

to help

YOUR CAREER

TAKE OFF

Meet Désiree She joined AkzoNobel after a Masters in Strategic Management and currently works at our Aerospace Coatings business. In her role as Market Segment Manager she acts as a portfolio manager servicing all segments of Aerospace Coatings, like OEM, Maintenance and Defence. “I have direct contact with different functions in the organization, such as RD&I, Marketing, Sales and Production. We are truly customer focused and everyone puts their heads together to find the best solution for our clients”, she says. “There’s a good reason that one out of every three airplanes worldwide is coated with our products.” She encourages Masters students to explore their personal boundaries. “If your career takes off at AkzoNobel, you will find a dynamic and open platform where you are free to maximize your abilities. And we are a multicultural organization present in more than 80 countries worldwide, including mature and emerging markets. So there is plenty of scope for travel and international work.” Désiree Oldenburger Market Segment Manager

To find out about our career opportunities, please visit our website www.akzonobel.nl/careers

501.10.162 AN Desiree-aerospace-ad.indd 1

07462_171013

Where your ideas go far

20-02-14 15:42

UNMANNED AERIAL VEHICLES: THE NEXT BIG THING?

WWW.MYNEWS13.COM

LVD

The benefits and detriments of military and commercial UAVs With law enforcement agencies around the world investing in Unmanned Aerial Vehicles (UAVs, also known as drones) and small commercial drone ventures popping up like weeds, the debate on the morality of drone usage has intensified in the recent months. What are the advantages of UAVs? Are they worth the disadvantages? TEXT Lotfy Hassan, Student Aerospace Engineering, President 21st Aviation Department

BENEFITS OF MILITARY UAVs The benefits of unmanned attacks over manned attacks are not limited to risks involving a pilot’s life. Over the past few years, defense contractors like Northrop Grumman have demonstrated that UAVs are capable of in-air refueling, dramatically increasing the endurance (Figure 1). Where pilots get tired after a few hours, drones do not lose their focus due to fatigue. They will be able to stay on station until their mission is fulfilled, which may take even up to hundred hours. Probably the biggest argument against drone warfare is the risk of substantial collateral damage. Professor Avery Plaw, a political scientist at the University of Massachusetts, investigated the ratio of collateral deaths in Pakistan using drones and conventional methods. Plaw considered four studies of drone deaths that estimated the proportion of civilian victims at 4%, 6%, 17% and 20% respectively. He also found that the proportion of civilian deaths using conventional military methods in the last twenty years ranged

LVD.indd 16

from 30% to more than 80%. Prof. Plaw concluded: “A fair-minded evaluation of the best data we have available suggests that the UAV program compares favorably with similar operations and contemporary armed conflict more generally” (Shane, 2012). However, the promise of precision killing and lower collateral damage does raise the question whether drones threaten to lower the threshold for lethal violence. CAUTIONS OF MILITARY USE The advantage of capturing a terrorist or target is that the military can avoid creating a martyr and cangain access to the bulk of intelligence. Drone attacks eliminate the possibility of arresting and interrogating targets, thus finding out what the terrorist was planning, how and with whom – or whether the individual was even a terrorist to begin with. Al-Qaeda, for instance, is a resilient global organization that was never singularly dependent on Osama bin Laden. It has proven over and over again that it is capable of

replacing dead operatives. The long-term goal of ending the group will not be accomplished by the killing of its members, if Al-Qaeda’s ability to spread its messages continues. BENEFITS OF COMMERCIAL UAVs Commercial UAVs can be used in a broad number of areas: maritime monitoring, telecommunications, transportation, search and rescue, oil and gas exploration to name a few. However, agriculture would be one of the industries that would benefit the most from commercial drones. Some estimate that 80% of the industry’s growth could come from precision farming, where farmers could use drones to monitor crops in a more efficient and timely manner. A good example of such an UAV is the Trimble UX5 (Figure 2). The Mesa County Sheriff ’s Office in Colorado has recently proven the benefits of using drones on search and rescue missions in the search for lost hikers in the mountains. Helicopter drones have also been used to help crews battle forest fires. “Flying

Leonardo Times JUNE 2014

09-Jul-14 11:40:51

manned planes or helicopters would cost at least $600 an hour. We fly [drones] for less than $25 an hour. It is just a new way to put a camera up that is affordable.” Ben Miller explains, who heads the program (Bennet, 2013). In general, commercial UAVs are less expensive than alternative solutions and are easy to use in rough terrains. CAUTIONS OF COMMERCIAL USE Because of the specific features of UAVs, the right to privacy may be compromised. Drones are capable of watching someone without that person knowing he is being watched or why this is happening. But what if it comes to monitoring houses or registering vehicles? Are people innocent until proven guilty? To what extent can law enforcement agencies monitor the general public? According to civil liberties groups, “the main problem with UAVs is that the technology is evolving faster than the law” (Bennet, 2013). Drone related laws have yet to be passed by court, so at the moment it is unclear whether drone surveillance would violate privacy laws more than manned aviation. Drone operators are not held responsible for criminal trespassing, stalking or harassment. Should the public be better informed? E.g. by indicating areas where drones are used for supervision, similar to areas where CCTV is being used to monitor the streets. A CCTV camera is visible on the streets and could be avoided if wanted. However, the UAVs soar through the sky almost invisibly and could be flying above your head without you knowing. Furthermore, a drone in the air involves risks. The issue surrounding UAVs is currently being discussed in the House of Representatives

TRIMBLE UAS

Figure 1. The Northrop Grumman Global Hawk provides a broad overview and systematic surveillance with long loiter time over target areas.

Figure 2. The Trimble UX5 allows for in-depth prospecting and visual inspection of crops and plantations. By combining visual and near infrared images, drought and other stresses can be detected long before it is detectable at visible wavelengths

in the Netherlands, and it seems to be more of a question of when than if, when it comes to drone regulations. CONCLUSION The benefits of both commercial and military UAVs are apparent. Military gains are not limited to not risking a pilot’s life. By taking away the human fatigue factor, the military will be more efficient in fulfilling their missions because drones are capable of staying in the air for hundreds of hours if needed. The biggest argument against drone warfare is the risk of extensive collateral damage. However, Professor Avery Plaw found that the proportion of civilian victims range from 4-20%. To put this in perspective, the proportion of civilian deaths using conventional military methods ranges from 30% to more than 80%. Nonetheless, drone attacks preclude one of the most effective means of undermining a terrorist group, namely the possibility of arresting and interrogating terrorists thereby gathering intelligence needed to ensure that the tactical strike could be converted to strategic advantage. In general, commercial UAVs are less expensive than alternative solutions and are easy to use over various terrains. They are therefore used in a wide variety of areas, such as agriculture, search & rescue and transportation. Currently, the main drawback with respect to drones is the level of regulations. Drone related laws are yet to be passed so at the moment it is unclear whether drone surveillance is violating privacy laws more than manned aircraft and helicopters.

place and the government develops the enabling long-term strategy regarding the technology, UAVs could very well be the next big thing. References [1] Kreps, S., Kaag, J., “The Use of Unmanned Aerial Vehicles in Contemporary Conflict: A Legal and Ethical Analysis”, Polity v44, Northeastern Political Science Association, pp. 260-285, 2012 [2] Heck. W., “Bevestiging na kamervragen: Nederlandse drone-politie bijna een feit”, NRC Binnenland, http:// www.nrc.nl, 2013 [3] EJure, “Drones en Privacy”, http:// www.ejure.nl/drones-en-privacy/, 2014 [4] Shane. S., “The moral case for drones”, New York Times Sunday Review, http://www.nytimes.com/, 2012 [5]Friedman, N., “Opinion: Evolving UAVs Allow U.S. to Stay Ahead of the Enemy”, USNI news, http://news.usni. org/, 2013 [6] Bennet. B., Rubin, J., “Drones are taking to the sky in the U.S.”, Los Angeles Times, http://articles.latimes.com/, 2013

Aviation Department The Aviation Department of the Society of Aerospace Engineering Students ‘Leonardo da Vinci’ fulfills the needs of aviation enthousiasts by organising activities, like lectures and excursion in the Netherlands and abroad.

When laws regulating use of UAVs are in JUNE 2014 Leonardo Times

LVD.indd 17

09-Jul-14 11:40:56

RICHARD VAN ‘T HOF, RH FOTOGRAFIE

Student project

FORZE HYDROGEN RACING TEAM DELFT

TU Delft students develop hydrogen race-car ‘Working towards a sustainable, yet exciting future’ is one of the big challenges in engineering nowadays. A group of students from the TU Delft accepted this challenge and designed a zero-emission hydrogen powered race-car. Combining green technology with racing, Forze wants to show the potential of hydrogen as an alternative fuel. Find out how Forze has been developing hydrogen powered racecars since 2007. TEXT Martin Hartvelt, MSc Student Aerospace Engineering and Chief, Vehicle Dynamics at Forze Delft

orze was founded in 2007 by a small group of students from the TU Delft with a dream: the dream to combine motorsports with green technology and to reach the same speed and intensity that is found in current races, but in a way that fits the future. THE POWER OF HYDROGEN Hydrogen fuel cell technology is a good way to make this possible. By using hydrogen from the tank and oxygen from the air, electricity can be generated with only water as a side product. After entering the fuel cell system, the hydrogen molecule splits into a proton and an electron. An electrolyte allows the proton to pass, but the electron is forced to take a different route, causing a current. The proton and electron then react with oxygen on the other side of the electrolyte and form pure water. Many sensors, pumps and other components are part of the balance of plant, which is needed to operate a fuel cell. In order to operate the system as efficiently as possible, a lot of testing is required, as can be seen in Figure 1.

Forze team.indd 18

In this way, the fuel cell can produce continuous power as long as hydrogen is supplied. However, in the highly dynamic environment of a race, continuous power is not what is required. Therefore, a buffer of ultra-capacitors is added to the system between the hydrogen fuel cell and the electric motors. This buffer allows regenerative braking and the generation of high bursts of power, used for acceleration. The energy buffer consists of 156 ultra-capacitors of 2300F each. The size of this buffer is roughly 800kJ, which is enough to boost the power to 190kW for approximately eight seconds. HISTORY OF FORZE Forze is one of the few teams in the world to design, build and race hydrogen electric race-cars. Currently, about seventy students play an active role in the team. The team is very multi-disciplinary with students from almost every faculty at the TU Delft. In 2007, the team started to build their first hydrogen-powered go-kart to compete in the Formula Zero competition. Forze won the world’s first hydrogen fuel cell powered race, called the Formula Zero World Championship, in 2008 with

the Forze I. After this success, Forze continued building go-karts for another two years, wining many more prizes. The last go-kart, the Forze III, managed to beat a conventional kart and showed the power of hydrogen against combustion engine karts. After three years in the Formula Zero competition, the team scaled up to the Formula Student competition, to compete against 130 other student teams from all over the world. After setting the official FIA record on the one-eight mile with the Forze IV of 10.45 seconds, the Forze V was the first hydrogen-powered car to ever compete in an official Formula Student competition. FORZE VI With five years of experience in racing with hydrogen fuel cell race-cars, Forze decided to scale up and build a full size track racer on hydrogen. In 2012, Forze started the development of their sixth race-car, the Forze VI. The Forze VI has a powerful fuel cell with a nominal power output of 100kW, which is enough to power thirty households. To generate this power output, 3kg of hydrogen is stored in two tanks at 350bar, which is enough

Leonardo Times JUNE 2014

09-Jul-14 11:43:12

RICHARD VAN ‘T HOF, RH FOTOGRAFIE

Figure 1. Testing the fuel cell system with the balance of plant.

Figure 2. The gearbox of the Forze 06.

Figure 3. Laminating the bodywork of the Forze 06

Figure 4. The suspension of the Forze 06.

to race on full power for about thirty minutes. The power that is produced by the fuel cell will be used to actuate the two electric motors, which drive the rear wheels. The electric motors, which are shown in Figure 2, have a maximum torque of 400Nm at 7500rpm and the torque is linear for the whole range of rpms. This results in a fast and agile car, which can accelerate from 0-100km/h in less than four seconds and has a top speed of 210km/h. The advantage of having electric motors is that the motors can be controlled separately, which makes it possible to use torque vectoring. This smart differential uses multiple inputs from an IMU and two antennas in the car. APPLICATION OF AEROSPACE KNOWLEDGE With over thirty BSc and MSc Aerospace engineering students in the team, Forze has a lot of systems in the car for which knowledge from Aerospace Engineering is put into practice. Applying the theoretical knowledge in the design of a hydrogen race-car gives the students a lot of practical knowledge and experience. This is very valuable to become a more professional and better engineer. A great example for the application of aerospace knowledge in the design of the Forze VI is the Aerodynamics department. Students in this department learn to work with Computational Fluid Dynamics (CFD) software, which is based on mathematical equations learned in the Bachelor. With

the use of CFD software, the design of the bodywork can be iterated to create the best aerodynamic shape for the car. When racing with a velocity of over 200km/u, Aerodynamics becomes very critical for the performance of the car. When the final external shape is determined, a scale model of the car will be put in the low turbulence wind tunnel at the TU Delft, to validate the simulations and to get more insight in the flow around the car. Additionally, the radiators for the cooling circuit will be tested in the wind tunnel to validate whether they provide enough cooling in the car. Apart from the aerodynamic behaviour of the car, the mass of the car also has a big influence on the performance of the car, especially for the acceleration. In order to get the car as lightweight as possible, aerospace students work on different systems to apply their knowledge to save weight in the design. One example of this lightweight engineering is the use of composites in the car. The bodywork is completely made of carbon and glass fiber composites. Laminating the bodywork will give the students some handson experience, as can be seen in Figure 3. On top of that, a lot of structural components in the car are made of Aluminium 7075-T6, which is a common material in aerospace applications due to its lightweight material properties. These lightweight materials are used in the design of the suspension, as can be seen in Figure 4. FUTURE PLANS In the coming months, Forze will be test-

ing a on the runway of the former airport of Valkenburg near Leiden, and on the circuit of Zandvoort. If the car proves to be reliable enough, Forze will be ready to show the power of the Forze VI to the world. The main goal of the Forze VI is to beat the lap record at the famous German circuit Nürburgring Nordschleife. This is a very challenging circuit with over a hundred corners in a twenty-kilometer track. The current record is set by a Nissan X-Trail FCV with a time of 11:58 in 2008. Next to setting the lap record on the Nürburgring, Forze will be present on a lot of other events to maximize the exposure of hydrogen racing. Racing against combustion-powered race-cars has always been one of the ambitions of Forze. Therefore, Forze wants to compete in the Caterham Cup: a race-competition in the BeNeLux for combustion powered Lotus Super Seven race-cars. Apart from all the dynamic events, every year Forze is participating in about thirty static events like fairs or open days. If the team can continue making the big steps that were made over the past few years, it might just be possible to start dreaming about a formula one car racing on hydrogen. GETTING INVOLVED Are you interested to get involved or support the development of hydrogen electric race-cars? Or do you want to have more information about the project? Please send an email to bestuur@forzedelft.nl or visit our website: www.forzedelft.nl JUNE 2014 Leonardo Times

Forze team.indd 19

09-Jul-14 11:50:17

DLR

COMPOSITE FORWARD SWEPT WING Euler corrections in the optimization of a composite forward swept wing.

The idea of increasing commercial aircraft’s efficiency is still an engineering challenge. Increasing efficiency can be achieved through reducing the aircraft’s weight, reducing the aircraft’s drag or other methods. Accounting for more of these objectives in the same aeroelastic optimization has proven to be a challenging effort. Recent development in aeroelasticity and the availability of more computational resources has made such preliminary design processes more accurate and meaningful. TEXT Marius Tase, MSc. Graduate Aerospace Structures and Computational Mechanics Chair

ultidisciplinary Design Optimization (MDO) has become more popular in aeroelasticity, as more disciplines are put together in order to ensure a more meaningful and detailed preliminary design. MDO processes can be described as being represented by four major components: the structural model, the fluid-structure coupling techniques, the aerodynamic model and the optimization process. These four are connected with each other in an iterative process. The idea of using composite materials and choosing the stiffness of each individual wing-box panel by using composite materials has become more attractive in the past years. The advantages are not just achieving a lighter design, but also choosing the desired aeroelastic behaviour of the wing. This article presents the results of a master thesis, where an MDO set-up has been im-

proved with a better aerodynamic model. CURRENT TRENDS By 2020, aircraft should have become 50% more efficient. One way to partially achieve this is to increase the wing surface area that exhibits laminar flow. DLR (The German Aerospace Centre), under the LamAiR project, developed a commercial aircraft concept designed with a forwardswept wing. There are several mechanisms that trigger the transition between laminar and turbulent flow and the sweep angle is one of them. The increase of the wing laminar flow area through designing a forward-swept wing can significantly reduce turbulence, and thus reduce drag and increase efficiency. The main drawback of forward-swept wings is the undesired twist-bending stiffness coupling leading to aeroelastic stability problems, such as wing divergence. Furthermore,

the use of composite materials have made it possible to reach lighter and more flexible wing designs, potentially making divergence even more critical. The severity of this phenomenon made it clear that there is the need of a preliminary wing design and optimization process that accounts as well for the aeroelastic problem. The optimization of the composite laminates can yield twist-bending coupling terms that could compensate for the wing divergence. Initially, such a tool has been developed at DLR while using the DLM (doublet lattice method) as an aerodynamic tool. Unfortunately, the default DLM cannot account for the wing airfoil thickness and wing twist, and it has an inaccurate compressibility approximation and cannot model the shock waves that commercial aircrafts usually encounter. These drawbacks of the DLM can be corrected by a CFD analysis and include the correction in the optimization process.

Leonardo Times JUNE 2014

Composite forward swept wing_2.indd 20

09-Jul-14 11:51:43

THE FEM The FEM model represents the wing-box as a shell model comprised of the wingbox skin, ribs, spars and stringers. The beam-like structures, such as stringers and spar caps, are modeled as beam elements. Each rib contains three spanwise points used for splining: on at the leading edge, one at the trailing edge and one in the center of the rib. They are connected between themselves through interpolation elements and they ensure a proper introduction in the structure of the wing twist moment induced by the DLM. The model also includes non-structural mass points (such as fuel or other masses). THE OPTIMIZER The optimizer, ALDO, is developed and provided by TU Delft and it is a gradientbased optimizer that is highly oriented towards optimizing composites with respect to lamination parameters. The optimization objective is the wing-box skin mass reduction with both structural and aeroelastic constraints. The constraints are the wing-box skin panel strain failure and buckling failure, wing tip twist, the divergence pressure and the aileron effectiveness. The design variables are the thickness and the A and D matrices (membrane and bending stiffness) for each wing-box skin laminate. The optimization’s stopping criterion is the wing skin mass, which is checked after each iteration for any absolute difference smaller than 10-3. The loading cases are chosen as such to be representative for a commercial aircraft, several cruising and rolling load cases, but also four loading cases chosen from the flight envelope. These last four loading cases are all symmetric trims: LC#1 has M=0.59 at h=0m with n=-

MARIUS TASE

PREVIOUS WORK The DLR’s previous research and development of the MDO process that can design a composite wing-box with aeroelastic constraints for a given set of loading cases, has already been published (Dillinger et al. 2012, 2013). The wing resembles an A320 wing, with a wing semispan of 17.89m a wing surface of 122.6m2, a swept back leading edge of -16.8° and with several supercritical airfoils that have a relative thickness ranging from 11.5% up to 14%. The aeroelastic solver used here is NASTRAN, which can solve the FEM model, and also provides a DLM code. The two models are fundamentally different and they require a transfer of information in between. The DLM mesh is deformed with respect to the FEM deformations, while the forces yielded by the DLM are then introduced in the FEM structure. This fluidstructure coupling method is also known as splining. In this case, splining is done internally by NASTRAN.

Figure 1. (left view) Chord-wise ΔCp of the DLM and the CFD (TAU) solvers with respect to different spanwise coordinates, applied to load cases 1 and 2 of the optimized configuration. (right view) Spanwise wing-box lift and twist moment resultants of the DLM and CFD.

1g, LC#2 is the same, but with n=+2.5g, LC#3 has M=0.87 at h=6,700m with n=-1g and LC#4 is the same as LC#3, but with n=+2.5g. All four are expected to drive the structural design and are mentioned here because the CFD solver for the aerodynamic corrections will use these cases.. THE DLM The DLM is one of the most basic and rudimentary tools one can use in steady and unsteady aerodynamic applications and its use can be extended up to compressible flow, as long as the free stream remains subsonic. The DLM equations yield a solution in the frequency domain, but for the steady case, where the frequency is zero, the DLM is reduced to the VLM (the vortex lattice method). This panel method is discretized through a plate-like mesh, hence it is unable to account for any wing thickness effects, wing twist or airfoil camber. As well, the equations use a PrandtlGlauert compressibility correction, which exhibits an unrealistic singularity effect as the Mach number is reaching the value of 1. The unpredictability of wing shock waves, characteristic to transonic flow, is also not accounted for. All these inaccuracies made it important to find a correction method to the default DLM. Fortunately, later, NASTRAN developed a simple correction method that accounts for the wing twist and, to a certain extent, for the airfoil camber. EULER CFD The Euler equations are a simplified approach for aerodynamic calculations,, where the viscosity and heat conduction terms are neglected. The Euler equations can approximate all the aforementioned effects that DLM cannot approximate,

but there are also some known drawbacks. The Euler CFD does not predict any boundary layer, while the shock waves are approximated as stronger and closer to the trailing edge, when compared to the real phenomenon. The consequence is the introduction of a stronger negative wing twist moment in the structure. Here, the available CFD platform is TAU, a modern software system developed by the DLR that can predict viscous or inviscid flow about complex geometries. In this case, the splining of displacements and forces has to be done separately, outside both NASTRAN and TAU. This is achieved by using a MATLAB code designed especially for this purpose, of coupling NASTRAN with TAU. VALIDATION AND COMPARISON Before the implementation of any correction method in a costly and complex system, such as a MDO process, one needs to quantify the differences between DLM and CFD with respect to both wing geometry and flow parameters. A simple rectangular wing is constructed, where several geometrical parameters are varying, in order to separate the influence of each parameter on the results. The means of comparison are spanwise lift and twist moment distribution. After a successful validation of the aeroelastic set-up, with DLM and CFD being in a very good agreement for simple and very thin wing geometries, one can notice that the DLM-CFD differences are mainly driven by the thickness effect. The CFD solver yields a 3-9% higher lift force, depending on the airfoil thickness, while the twist moment is still in good agreement. For cambered airfoils, the camber correction from NASTRAN has only a small inJUNE 2014 Leonardo Times

Composite forward swept wing_2.indd 21

09-Jul-14 11:51:45

accuracy with respect to the lift, but the wing twist moment is underestimated by the DLM with up to 22%. The next step is represented by the investigation with respect to the flow conditions. With the geometrical influences already known, one can now use the forward-swept wing geometry for four different increasing Mach numbers: incompressible (M=0.25), the critical Mach number (M=0.65) and two transonic cases (M=0.75, M=0.87). The first case confirms the results of the previous investigation. The second shows a steady increase of the lift yielded by CFD, due to a more realistic compressibility approximation. The third case exhibits weak shocks, which induce a higher negative wing twist, while the fourth case shows strong shocks with a significantly negative wing twist. Figure 2 illustrates an example for M=0.87. THE CORRECTION METHOD With the CFD-DLM differences known, one can proceed to implementing the aerodynamic correction inside the optimization. Basically, the correction is represented by the difference of the force vectors (X, Y, Z forces) between the DLM and CFD forces after splining them in the FEM model. These forces are then introduced in the optimization loop after which the process is restarted. The optimization without the CFD correction takes 29 iterations to reach convergence, but the solution is already close to being optimized after ten

iterations. This information is used in the new configuration, where the structure is quasi-optimized for ten iterations using DLM, then the TAU module is started, the DLM-CFD force difference is computed and reintroduced into the structure, then the entire loop is restarted. In this case, a separate investigation showed that six loop repetitions are sufficient to achieve DLM-CFD force difference convergence. For the last series of structural optimization, the optimizer is not constraints to stop after ten iterations, but it follows its own wing-box mass convergence criteria. THE RESULTS The optimization with the Euler corrections finished after a total of three days. Figure 3 shows the mass development of the wing-box skin, with the starting design weighing 1,663kg. The vertical black lines show where the CFD solver is started. The optimized structure is 5.7% lighter than the uncorrected solution. The same figure shows the development of the divergence pressure per iteration. Unlike the uncorrected optimization, the divergence pressure is an active constraint (35,000Pa), meaning that the structure is used in a more efficient way. Wing tip twist and aileron effectiveness are not active constraints in this case. Figure 1 and 2 show the comparison of the DLM-CFD ΔCp of the optimized solution for the four relevant load cases mentioned before. Figure 2 show the effect

MARIUS TASE

Figure 2. (left view) Chord-wise ΔCp of the DLM and the CFD (TAU) solvers with respect to different spanwise coordinates, applied to load cases 3 and 4 of the optimized configuration. (right view) Spanwise wing-box lift and twist moment resultants of the DLM and CFD.

Figure 3. Wing-box skin mass and divergence pressure development per iteration. The vertical black lines represent the moment where the CFD solution is computed.

of the shock wave and it’s consequences in the wing-box twist moment. Basically, the higher negative wing twist moment of the corrected solution influences the stiffness in the trailing edge spar area, where more panels have their main stiffness oriented as such to compensate for twisting and buckling. Here, the stiffer trailing edge makes divergence an active constraint. Generally, the structural constraints and the failure modes follow the same trends as before, with the wing-box tip being driven by buckling failure, while the rest of the wing-box is mostly driven by strain failure. As a concluding remark, the Euler corrected optimization does not fundamentally change the results, but the additional wing twist has an influence on the wing skin stiffness orientations, which by themselves influence the wing’s aeroelastic behaviour. References [1] J. K. S. Dillinger, M. M. Abdalla, T. Klimmek and Z. Gürdal. “Stiffness Optimization of Composite Wings with Aeroelastic Constraints”. 14th AIAA/ ISSMO Multidisciplinary Analysis and Optimization Conference. (September 2012), Indianapolis, Indiana. [2] J. K. S. Dillinger, M. M. Abdalla, T. Klimmek and Z. Gürdal. “Static Aeroelastic Stiffness Optimization and Investigation of Forward Swept Composite Wings”. 10th World Congress on Structural and Multidisciplinary Optimization, (May 2013), Orlando, Florida, USA.

Leonardo Times JUNE 2014

Composite forward swept wing_2.indd 22

09-Jul-14 11:51:46

careersathmc.com

Is it really the blue sky youâ&#x20AC;&#x2122;re dreaming of? At Heerema Marine Contractors (HMC), we see a lot of

Quite frankly, the circumstances to cope with at 10,000

blue. Blue seas around the world, and blue skies on the way

ft depth may be a lot more extreme than what youâ&#x20AC;&#x2122;ve

there. What we do is not exactly rocket science, but we

prepared for at 30,000 ft up in the sky. So if you are looking

come pretty close. We design, construct and install mega

for a sky high career perspective, you may want to consider

structures for the offshore oil and gas industry. Platforms

landing at HMC. The oil and gas industry is booming and

and oilfield infrastructure for operation in shallow waters or

provides new, magnificent engineering challenges every

at depths of several kilometers on the seabed. The deeper

year. Plus a tremendous responsibility for the safety of man

the oilfield, the more high-flying engineers we need to

and the environment. And the tools we use are pretty cool

conceptualize, calculate, design and innovate towards future

too. Our largest crane vessel Thialf has a lifting capacity of

groundbreaking and record-setting projects.

about 50 Airbus A380s. Just so you get the picture.

A HEEREMA COMPANY

Check in for a test flight at: careersathmc.com

Internship report

INTERNSHIP AT BENTLEY MOTORS

Driving Luxury Performance Bentleys are amongst the most sought after luxury cars of the world. They are known for their performance on the road, the luxury in the cabin, their reliability and the artisanship involved in making every single one of them. It was an honor to be part of the Quality Department of a company with such a rich history in British car manufacturing as part of my six-month internship. TEXT Felix Minde, MSc Student Aerospace Engineering

hen I started studying Aerospace Engineering in Delft in 2009, I participated in the OWEE, the welcoming week of the University. I remember biking down the campus one day during the OWEE, hearing the deafening sound of a petrol engine. It was the first time I saw a Formula Student car and the day I decided to try joining the team in the future. I did so in my second year and found it fascinating that a team of students can design and build amazing cars in less than a year’s time. This is why after spending one year with the team; I decided to do my Minor at the same place, which meant spending half a year full time with the DUT Racing Team. Following my increasing interest in cars I chose to do the master track “Design and Production of Composite Materials”, which would give me the opportunity to work in the aerospace industry as well as the automotive industry due to the ever growing demand in lightweight structures. I decided to do the mandatory internship at one of Europe’s car manufacturers to get a better insight into whether this field of work is right for me. My focus was here on the smaller companies because I thought that getting an overview of the entire business would not be easy at a mega-company like Volkswagen. One of my applications went to the quality department of Bentley Motors and even though the offered position was not directly relevant to composite materials, it would get me into contact with most of the business.

The available internship was in the quality department and more specifically the field quality team, which deals with manufacturing, logistics, engineering, aftersales and the customer. My application was successful and a week after the telephone interview I got the great news that I got the job. This process might sound quick and easy, but believe me it was not. All in all, four months passed between sending the application and getting the Ok.

holds true as Bentley strives to become the world’s leading luxury car manufacturer. All Bentleys are designed, styled, tested and built by a total workforce of about 4,000 employees on the main site in Crewe, which only covers an area of about 500x500 m. Currently, there are four different models in production: A coupe, the Continental GT, a convertible, the Continental GTC, a sedan, the Flying Spur and Bentley’s Flagship the limousine Mulsanne. Soon, Bentley will introduce their fifth model, with which it will enter the popular SUV market. A number of facts and figures about the cars can be found in Table 1.

BUILD A FAST CAR, A GOOD CAR, THE BEST IN ITS CLASS Before continuing with my experience, I will try to give you a short impression of what Bentley Motors is like. Bentley Motors Limited is a luxury car manufacturer located in the heart of England in Crewe, Cheshire. It was founded by W.O. Bentley in 1909 in London with the goal to “Build a fast car, a good car, the best of its class” and even today this famous quotation still

For the first month of the internship, I was on a placement. Instead of starting in the Field Quality Team right away, I was a member of a task force team for the problem resolution in manufacturing. The idea was that, this position, even though it felt like being thrown into cold water at first,

Table 1: Current Bentley Models MODEL

Continental GT

Continental GTC

Flying Spur

Mulsanne

TYPE

Coupe

Convertible

Sedan

Limousine

ENGINE

3993cc Twin Turbo V8/ 5998cc Twin Turbo W12

5998cc Twin Turbo W12

6752cc Twin Turbo V8

MAX POWER

500/625 PS

625 PS

512 PS

MAX TORQUE @ 1700 RPM

660/800 Nm

800 Nm

1020 Nm

0-100 KM/H

4.8/4.2s

5.0/4.4s

4.6 s

5.3 s

MAX SPEED

303/330 km/h

301/325 km/h

320 km/h

296 km/h

BASE PRICE [1]

€198,700 / 291,600 €219,900 / 317,600

€279,100

€431,804

Leonardo Times JUNE 2014

internship bentley 2013.indd 24

09-Jul-14 11:55:17

FELIX MINDE

BENTLEY MOTORS LIMITED BENTLEY MOTORS LIMITED

Figure 1. Rear seats of the Mulsanne

Figure 2. Interior of the Continental GT Supersports

would be a good start to get to know the business and spend a lot of time with the products in the assembly hall. Our task was to pick up problems detected at the end of line checks, feed them back to the responsible zones on the assembly line and solve the problem at its root. The task turned out to be the perfect introduction to Bentley. I got to know the four different models, the steps involved in assembling a car, the colleagues on the assembly line and I was involved with the Quality department as well if it came to quality-related issues. It was amazing to see how a car slowly came to life all the way from the bare chassis to the engine being fired up for the first time. Virtually the entire process, except for placing the rubber bead on the windscreens, is done in the absence of robots. There is still a lot of craftsmanship and experience involved in the production process, especially in the wood and leather pieces. I also learned to appreciate the uniqueness of a Bentley. It is close to impossible to spot two Bentleys which are the same, except if a customer ordered two Bentleys at a time for example to have one at each of his houses - but then still the colours might be different to match the colour of each house. A GT alone can be ordered in 55 million variations. FIELD QUALITY TEAM EXTERIOR & INTERIOR After four weeks, I moved to the Field Quality Team Exterior & Interior, where I stayed for the remaining five months. Examples of a Bentley’s Interior can be found in Figure 1 and Figure 2. As the name suggests, the Field Quality Team is responsible for improving the field quality of Bentleys. Information about issues with the cars is gathered from a variety of sources such as dealers, workshops, customer feedback, mileage cars and the factory. All problems are then prioritized based on faults per

Figure 3. The Scottish Highlands during a weekend trip in November

1000 vehicles, cost, severity and risk. The top issues are handed to field quality engineers of the responsible field and the investigation for a root cause hypothesis is started. These investigations can involve further data analysis, part testing, laboratory trials and part measurements etc. Once the root cause hypothesis is established we can pin down the part and part owner. The problem is then handed over to the engineer who ‘owns’ the part, the part supplier or manufacturing, where the root cause will be confirmed or defined in more detail and a solution developed and eventually implemented. The Field Quality Team takes a managing role while the problem is with the problem owner and ensures that the issue is fixed within a certain time frame. The main task for all investigations is to have a temporary fix in place quickly such that the customer is protected. This temporary fix will later be replaced by the permanent solution. Within this framework, I was given a few field issues to investigate, find the root cause and eventually hand them over to the problem owner. I really enjoyed this work because as expected I got to deal with all kinds of different departments within Bentley and improved my problem solving skills. Apart from the investigations, I was also given a few projects in Excel. As such, I worked on improving the system of ranking incoming field problems and updated a corporate learning database that is available to the entire business. The most rewarding task was however to create two search tools for the team to check a large number of cars for their vehicle specifications or log files from the ECUs. My codes reduced the amount of time to check 100 Bentleys from more than an hour to just 25 seconds. DRIVE EVENTS AND FREE TIME In general, I noticed that the hierarchy at

Bentley is more relaxed compared to other companies. For example, all levels up to the head of the department are seated in the same office and are always available for a quick chat. Highlights of the internship were certainly the Drive Events when I got to be driven in a Bentley. Unfortunately, driving the product was not possible because Bentley requires a special in-house driving license for it. During the drive events though, I had the honour of experiencing a Bentley like a customer, driving through England. Every single time it brought a smile to my face to see how effortless they accelerate and move through corners while offering such a luxurious feel, which is exactly Bentley’s identity: Driving Luxury Performance. There are always around 40-50 interns at Bentley, most of them from Germany, such that there was always something to do during the weekends. For shopping, sightseeing and nights out we went to destinations close by, such as the twin cities Liverpool and Manchester that are just a 30 min train ride away. A couple of times, we also went for weekend trips to destinations such as London, Glasgow, Edinburgh or Dublin. One big advantage to facilitate these trips was the fact that working hours at Bentley are only from 8-12 on Fridays in contrast with full working days otherwise. Overall I had a great time in Crewe and really enjoyed the working climate, work and products at Bentley. I highly recommend everyone looking for an internship at a true British car manufacturer to have a good look at Bentley’s website for open positions.

References [1] Prices and specifications Bentley, http://www.autowereld.com/nieuweautos/prijzen-specificaties/bentley, assessed 02/2014

JUNE 2014 Leonardo Times

internship bentley 2013.indd 25

09-Jul-14 11:55:40

DUTCH SPACE

Interview

DUTCH SPACE Interview with Arnaud de Jong, CEO Dutch Space, the largest space company in the Netherlands and part of Airbus Defence and Space, appointed a new CEO last year. The Leonardo Times sat down with the CEO Arnaud de Jong for an interview. We discuss his career, developments in Dutch Space, his take on competition in the commercial space domain and his future outlook on European and International space markets. TEXT Jeroen Wink and Lakshmi Rathina Sabbapathy, Editors Leonardo Times, Students Aerospace Engineering

LT: Could you tell us about your education and about your professional journey to Dutch Space? “I studied Economics and Mathematics at the University of Groningen, thereafter I went abroad to Canada to study further and do a traineeship. The idea of living and working in Canada was an interesting experience for a year. I came back to Europe to study as an Erasmus student for one year, with a postdoctoral study in Montpellier, France, and went to military service, which was mandatory at that time in Holland. I had a great time there actually. When I started to work, like many students with their first jobs, I was not sure of what I wanted to do and had a broad variety of applications, ranging from very technical companies to banks. Finally, I started to work at a consultancy company and did further studies as a registered accountant. The aim was not to establish a career there, but to gain more work experience, which is very helpful for what I am doing today. I then went to Germany, where I did a consultancy job for DASA, which at that time was Daimler-Benz Aerospace, and

later became DaimlerChrysler Aerospace. They hired me afterwards and I entered DASA’s Mergers and Acquisitions department. During that time I was working on a deal called Astrium, which was a big merger of three space entities in Europe, the German, French and the British businesses; this was a very International operation. I found myself to be very fortunate to be working in that department as a young professional in those very interesting times. Subsequently, I went to Paris at the creation of EADS (European Aeronautic Defence & Space Company), which was a merger of all aeronautics business that the Germans, Spanish and the French had. There was still a lot of consolidation going on in Europe: MBDA, the missile company was created. Airbus was also created and emerged as a company. Astrium was created for the satellite business. I lived in Paris and worked in the headquarters located there for three years. Then I came back to Germany to work as the head of the office of the CEO of the defence business, and carried on my work for another three years. As a next step, I joined the former Cassidian, the defence business of

EADS at that time, as the head of technology, the head of the Eurofighter program and a UAV program.” LT: As the CEO of Dutch Space, what is your view regarding the competitiveness of Dutch Space compared to European companies like OHB of Germany, RUAG of Switzerland and Thales Alenia of France? “First of all, I have to mention that Dutch Space is a 100 percent subsidiary of Airbus Defence and Space. So, we are not positioned in a system integrator role like, for instance, the former Astrium in Toulouse and in Germany obviously are. OHB is definitely positioning itself as the prime integrator of the big systems at the same level where Airbus is, so they are competitors with Airbus Defence and Space. Although we are a 100 percent affiliated company of Airbus Defence and Space, I do not see them as a competitor at all: OHB is our customer we deliver to, like we do to all space primes in Europe. So, we have to be careful with defining who our competitors are, because competition on that side is not present.

Leonardo Times JUNE 2014

Interview Dutch Space.indd 26

21-Jul-14 19:16:33

DUTCH SPACE

Figure 1. VEGA interstages in the Dutch Space clean room

Thinking about our competition on the subsystem level where we are, e.g. instruments, structures and solar panels, we have genuine competitors who are not always comparable companies to Dutch Space, as they might have a different origin or specialization. It is a fact that we have delivered a tremendous number of solar arrays in Europe, despite competition and that we have managed and will manage again to export to the United States, where another big competitor for solar arrays is located. The market is really tough in that sense. If we manage to win these contracts again, of which we are convinced, it confirms that we are really competitive. It’s quite a transparent market, so apart from meeting the requirements and delivering exactly the product that the customer needs; the only thing that determines if you win a contract is the price. For commercial contracts, a Dutch company can compete with American companies and win with a better price, as there are no politics involved. What you do need to demonstrate is your reputation and capability of delivering. One needs to convince the customer that you have the right qualifications to be in this market.”

LT: Regarding competitiveness, are you concerned about the rise of private companies like SpaceX? “We have already started acknowledging that this market is becoming much more commercial. SpaceX is of course a good example. Commercialization is a trend, which is here to stay. At Dutch Space, we

to be able to deliver at a good price. And delivery also concerns being in time, within cost limits and delivering good quality. If not, the customer will not accept it. The commercialization is already going on.” LT: Dutch Space is mainly active in Europe but also has projects in the USA. Is Dutch Space also involved in the “emerging” space markets like India, South Korea or China?

“ Dutch Space is a 100 percent

“First of all, we are closely monitoring the activities in the emerging countries. We want to focus more on product services and recurring product delivery. Of course, we see China as a huge market. A market with its challenges, but also one with huge opportunities. We are talking with Chinese representatives on certain products, but of course we have to be careful not handing-over our technology. That does not mean that their market is closed - we are discussing options with them. In China, we were talking about delivering somewhat standard products, although nothing is standard in space, but we would still need to adapt certain features to meet the specifications of the Chinese customer. Currently, the US is indeed our biggest export market, thanks to our good connections with the large primes there.”

subsidiary of Airbus Defence and Space. However, we deliver our Dutch high-

”

tech to all space primes in Europe have seen it coming, and we are responding to it, because if you cannot be competitive at a commercial level, it is not possible to survive. However, this is not new for us to be honest. People tend to think that only SpaceX is commercial, whereas ESA is also very commercial. Just because there are no competitors from the US, it does not mean that there is no competition in Europe. Fortunately, we have enough competition here. And you need that environment to be on the edge of the technology. It is a continuous race of being in front of others in certain innovative issues, while on the other hand you need

JUNE 2014 Leonardo Times

Interview Dutch Space.indd 27

21-Jul-14 19:16:39

DUTCH SPACE

“We are not active in nano satellites, because it is not a big volume market, so the effort is not worth the cost. There is certainly a trend towards smaller satellites with a large power request. It is one of the areas where we are successful; we won tenders in that field. It is positive to see that the electrical requirements of satellites are increasing. More and more, it comes down to the question how much power you can deliver, hence how good your solar arrays are. Size of the satellite is of course an indication on how much power it consumes, but even the smaller satellites are getting a higher power density, which is more important for us than the mass and volume. We are still meeting requirements for satellites down to one metric ton in mass. We have won some commercial contracts for these sizes. Of course, the panels are smaller but they are also more of a recurring product, which is what we are aiming at. The good thing is that all the experience we have built up over all these years’ fits with the market needs. For a commercial need, you just need good technology, maybe not even topnotch since it needs to work and there is always a price cap on it. Of course, it is still high-tech, not obsolete technology in any way. But in the commercial market, the price is the only a discriminating factor.

Figure 2. The European Robotic Arm at Dutch Space

LT: Are you worried about Intellectual Property problems in some countries? “To put it bluntly, we have a core technology that makes our solar arrays successful, we are truly amongst the top players in the world, for which have protected certain features that guarantee the success of the overall system. As you can imagine, we are not interested in selling these core technologies. I am not willing to sell just one. Even if the technology is protected, it is not about copying as such, I am here to do business. Not to hand-over our blueprints even if we make some bucks from it.”

LT: New players like China and India are building their own launchers and satellites, they are of course very interested in these technologies. “Yes, we understand it if they order a solar array, because they are looking for certain technologies for an interesting price. We will reply for sure, under the condition that we can sell it as an entire subsystem. In that case, there would be no difference for us whether we sell it to India or to the US or UK. So there is no limitation on that.” LT: In the space industry, there is a trend in miniaturization with nano and micro satellites. For example at SSTL (Surrey Satellite Technology), is Dutch Space redesigning their systems for these smaller satellites?

“People tend to think that only SpaceX is commercial, whereas

ESA is also very commercial” 28

We are certainly not aiming at providing solar arrays for cubesats, although we did deliver them for the Delfi satellite. Looking at the added value of those satellites and the amount of effort it takes, for us the focus is to just help and support them. For comparison, it is not comparable to for instance the COTS CRS ISS freight spaceship, equipped with two solar arrays (Cygnus capsule-red), because then you are talking business - in terms of complexity, volume and money.” LT: If you look at the customers of Dutch Space, What is the distribution between commercial, institutional and defence? “It is always fluctuating. A few years ago, we had a bit more of commercial clients. And now about two thirds of our revenues are ESA related, 25% is commercial and something like 10% is from defence. That is roughly it. The two-thirds revenue from ESA is quite steady but the commercial revenue is a bit fluctuating. Defining more clearly the commercial revenue, it simply means that there is no intervention of ESA in it. The 10% from defense is a fact for a couple of years already, which is basically the embedded training system of the F-35 for Lockheed Martin. Currently at Airbus, the defence and space segments are being merged together. That also means that our focus on defence will not get

Leonardo Times JUNE 2014

Interview Dutch Space.indd 28

21-Jul-14 19:16:40

DUTCH SPACE

Figure 4. Picture of Cygnus taken by the ISS on 29th September 2013

DUTCH SPACE

lesser. Defence is also a business where you need to have a long breath, not something you can take on in one year.”

DUTCH SPACE

Figure 3. Rosetta’s self-portrait during it’s Mars swingby (flyby)

LT: What are your plans for Internationalization of Dutch Space as it is an Airbus Defence and Space company? “We are now in the new business line of electronics, which is one of the four business lines of the new Airbus Defence and Space organization. We are focusing on products services at a subsystem level, so not the system level where Toulouse and Ottobrunn are. This subsystem level gives us a broad market. For us Airbus is like any other customer, which allows us to also export to others primes like OHB. The second step is that we have combined the solar array business that we have here in Leiden, with the solar array expertise we have in Ottobrunn. These are being integrated under my lead. That also means that we are combining strengths, by not re-inventing the wheel but by approaching customers at a clever level. Of course, we also have the brainpower and supply chain power that helps us to adapt to exactly what the customer wants. We can deliver from Germany and from The Netherlands. These are big advantages because we are not comparable. Ottobrun is not comparable to Leiden from the structure point of view but also not from the customer and the technical design point of view. There is a lot of integration going on. It’s a clever approach towards the market.” LT: What kind of new proposals for Ariane 6 is Dutch Space putting forward to enter negotiations with? “This I’m not going to answer, because we are in competition! But what I can say, which is not a secret, is that we have a long-lasting reputation from various Ariane and Vega projects. It is a clear task to gain a high added value in Holland and

Figure 5. Advanced Solar array hinges including the synchronization system

here in Leiden with our offers. And of course we are not departing completely from what we have done and built-up in the past for the Ariane project. We are still talking about a launcher. It will have all the features of a launcher. Of course, the design will be different, the material will be different, and maybe the timing of the throughput and the output will be different. But we are still talking about a launcher, it will still be launched with a

tors clever by telling exactly what we are doing. But of course the choice of the technology you are using, will very much influence whether you are successful or not.” LT: What plans do you have for the ERA (European Robotic Arm)? “Our plan is to finally launch it, but that is not up to Dutch Space of course. It was scheduled for this year, but it was postponed. We truly hope that this time the launch will succeed in 2015. I cannot add anything else; ERA is a brilliant ERA project and product (Figure 2), it works but it needs to be launched. Its ready to be delivered and now we are waiting for the launch. It is not under our responsibility but of course, it’s why we are doing it. To be launched - that is the final target.”

“There is certainly a trend towards smaller satellites

with a large power request” certain target, with certain specs like mass and volume we need to carry etc. and we will not depart from the heritage that we have. Today, we build launcher thrust frames, which are one the heaviest loaded part of the Ariane 5. The same holds true for the interstage 1/2 of Vega (Figure 1). We’ll look at these parts very carefully in the new design to assess what we can add into that. But I am not going to answer since we have all the sections as we call it and I do not want to make our competi-

LT: The arm was supposed to build the ISS, now it’s finished so what is the intent for ERA now? “The good news is that ISS will be occupied for much longer than some people thought. There will be repairs on the exterior of the space station to be done and JUNE 2014 Leonardo Times

Interview Dutch Space.indd 29

21-Jul-14 19:16:43

DUTCH SPACE

Figure 6. Sentinel-1A in the clean room at Dutch Space

LT: Will Dutch Space hire more young people with the expansion of product lines? Would you like to have more collaboration with the Delft University of Technology? “On your first question, that is definitely one of the targets that we are looking at. But not only young people, although be assured that I have nothing against young people. We need certain characters and certain spirit matching with the products and the experience that we have; a team should have the right mix of young and experienced people; also young in terms of mentality. It is indeed one of the targets and it is also one of the success factors we have to keep in mind. However, at the moment we also steering to become a healthy company, in terms of the capacity we need and in terms of how we are doing as a business. On your second question; I am in very good contact with Hester Bijl (Dean, Faculty of Aerospace Engineering, TU Delft) on exactly these questions and in the future, this is not 100% confirmed yet, I intend to play a more active role in

giving advice to the university, because that is where the talent pool is.” LT: Where do you see the future of the European space industry in another 10 years in terms of Space Exploration? “That is a very complicated question. You can have multiple answers on that. I still think we are just starting to explore the universe. We are just starting to explore what we know about the place where we live. On Earth, there is continuous, high demand on everything, from the trace gases that you want to measure, to taking measures for a better environment. We, all the billions of people on this planet, still create issues for which space could have the best or cheapest, or most constant solution. Exploring the solution is

another answer. We just started; look at Rosetta and our solar arrays (Figure 3). The wakeup call confirmed that our product delivered power and that our solar panels are working under the most severe conditions you can imagine. It is only a small contribution to exploring the universe. Fortunately this quest will continue, because I think we have just begun to understand what needs to be explored. I see a bright future for that. The universe is immense and I think it’s just becoming more and more understood, also taking into account that space adds to the quality of our lives on earth. The more and the better we can understand space and use it for our global challenges, the better it is for our planet, for space industry and of course for Dutch Space.”

DUTCH SPACE

ERA will play an important role in that. It’s also the risk for cosmonauts of going outside the ISS. It is a fatiguing and potentially hazardous task, from which the cosmonauts would be relieved. One could also imagine, that the older the ISS gets, the more you would need a servicing system like ERA. (It would be used for the services on the outside and for payload). The work is there. It just needs to be launched, that’s the point.”

Figure 7. Inspecting a Solar Panel in the clean room at Dutch Space

Leonardo Times JUNE 2014

Interview Dutch Space.indd 30

21-Jul-14 19:16:48

Jouw universum, ons dagelijks werk www.dutchspace.nl

RESIDUAL STRESS EFFECTS ON FATIGUE A consistent experimental method to quantify residual stress effects

As a product of manufacturing and assembly, residual stresses are present in an aircraft before it has flown a single mile. It is of no surprise that those internal stresses affect the fatigue life of the aircraft. The surprising reality is, that residual stresses are not taken into account in fatigue life predictions at all. One of the main reasons for this is a lack of correlation between experiment and predictions. Clearly, fundamental research is required in this field. TEXT Tom Lotz, Master Student Aerospace Engineering, Structural Integrity and Composites

he damage tolerance design philosophy is widely adopted in the aircraft industry nowadays. This concept is based on the fundamental assumption that initial defects will always be present in the structure. As such, these defects will likely develop into fatigue cracks and might continue until a critical crack length is reached if no action is undertaken. Consequently, the crack growth time must be accurately predicted to ensure safe aircraft operations. Although the effects of loading, environment and structural geometry on the fatigue life of a component can be accounted for in design, residual stress effects are less pronounced. One of the reasons for this is the difficulty of measuring residual stresses; they leave no outward sign and change in magnitude or direction during assembly. Furthermore, they evolve during aircraft life because of overloads, micro crack formation and stress relaxation. Next to undesired elusive stresses that result from manufacturing processes, the aircraft industry makes use of well-controlled methods like cold hole expansion and shot peening to introduce compres-

sive residual stress at the structural surface (where fatigue cracks are most likely to initiate). Even though these methods have been successfully used for decades already, their beneficial effects may not be taken into account [Skinn, 1994]. The main reason is that there has been limited correlation between experiment and prediction in the past. This is most likely due to confounding factors in the experiments [Stuart, 2011]. The goal of this thesis study was therefore to quantify residual stress effects on crack growth by consistently introducing residual stresses into lab coupons, measuring the stress profile and performing crack growth tests on them. STRESS INTRODUCTION The fundamental principle behind residual stress is an incompatible strain field within the structure volume. This incompatibility will lead to internally ‘captured’ strains (Eigenstrains) and consequently to internal stresses. Plastic bending was chosen for this study in a four point bending configuration, as shown in the visual. 12.7mm thick 7050-Aluminium coupons were used to avoid buckling during bending. After

the maximum load had been reached, the spring back generated a Z-shaped residual stress profile in the coupons without having confounding factors like heat affected zones (welding) and through thickness stress variations (cold hole expansion, quenching). STRESS MEASUREMENTS During the four points bending operation the coupon superficial strains were measured by strain gauges and digital image correlation (DIC). Figure 1 shows the horizontal strain distribution over one coupon surface at maximum applied load as measured by DIC. Those strains were transformed to residual stresses by using the compressive and tensile stress-strain data. A second method called the ‘slitting method’ was employed to measure the residual stress profile of three coupons. For this method a cut had to be made incrementally over the desired path. As the cut was increased in depth, the coupons deformed as result of the redistributing residual stress. This deformation was measured and calculated back to the original residual stress profile. As shown in Figure

Leonardo Times JUNE 2014

Residual stress.indd 32

09-Jul-14 12:07:32

TOM LOTZ

Figure 2. The residual stress profiles as measured by strain history and slitting.

TOM LOTZ

Figure 1. The horizontal strain distribution on the surface of a coupon as measured by Digital Image Correlation.

Figure 3. A close-up of one coupon clamped in the hydraulic grips. The small starter notch is visible on the left edge of the coupon.

2 both methods agreed to within 20MPa, which is considered very acceptable according to the residual stress standards. This confirmed the correctness of the measurements and the consistency of the bending operation. CRACK GROWTH TESTS Now that the residual stress profiles were known, crack growth tests could be performed on stress-free (baseline) and stressbearing coupons. A very small (0.1mm thick) starter notch was cut on one edge of every coupon from which the fatigue crack propagated. In the stress-bearing coupons the notch was made at the compressive residual stress side. At a crack length of about 8mm the crack would then grow into the tensile residual stress field. This mimics real-life residual stress profiles such as from cold hole expansion and shot peening. As shown in Figure 3, the coupons were clamped into a fatigue bench by hydraulic grips. Because of small imperfections in the clamping and the load train, non-symmetric crack growth occurred on the first two coupons. To avoid this, the next coupons were fatigued in a three-point-bending configuration. Against all expectations, there was a larger amount of non-symmetric growth. After extensive discussions with supervisor Marcias Martinez and lab technician Hans Weerheim, it was decided that a compromise had to be made by testing all remaining coupons from the residual tensile side. This greatly accelerated crack growth and gave more symmetrical cracks.

Figure 4. Crack growth rate versus applied stress intensity factor at different stress ratios.

FINITE ELEMENT MODEL In addition to the experimental work a Finite Element Model (FEM) was developed that mimicked the four-point-bending and fatigue tests. The bending was modeled such that the strain distribution over the midline was equal to the average of all tests. As expected the residual stress profile from the FE model fell within the experimental extremes. To mimic the fatigue tests, the applied force was introduced by a distributed load and crack growth was simulated by manually incrementally deleting constraints on the crack path. Stress intensity factors (K) were obtained at each crack increment by using the J-integral technique. EXPERIMENT VS. PREDICTION Crack growth predictions were made by superimposing the residual stresses on the applied loading. To do this, the baseline crack growth data was interpolated to find growth rate data at all intermediate stress ratios. In Figure 4, the predictions are shown together with the experimental data. It can be seen that crack growth rates are initially accelerated due to the tensile residual stress and later on decelerated when the crack enters the compressive stress zone. All predictions are within a growth rate factor two, which is within the amount of experimental scatter that can be expected in fatigue tests on metallics. It was concluded that superposition is the correct method to take residual stresses into account. If the residual stress distribution is one-directional and is accurately measured, the crack front is one-dimen-

sional and multiple cracking is absent. THE ROAD AHEAD It is obvious that there is no aircraft that contains deal residual stress states. Usually the structure is more complex, residual stresses are multi-directional and prediction a priori is extremely difficult. In order to design with residual stresses an integrated approach is therefore required that takes every manufacturing step of the parts into account. More research should lead to standardized residual stress profiles for deliberately introduced residual stresses. For bulk residual stresses from manufacturing, critical crack paths should be identified for each component by engineering judgment. Only if enough effort is made to arrive at this level of knowledge, the inheritance from this thesis can be employed to refine damage tolerance calculations.

References [1] D.A. Skinn, J.P. Gallagher, A.P. Berens, P.D. Huber, J. Smith. “Damage tolerant design handbook”, Wright Patterson Air Force base, Ohio: University of Dayton Research Institute, prepared for air force wright aeronautical laboratories. 1994 [2] D.H. Stuart, M.R. Hill, J.C Newman Jr., “Correlation of one-dimensional fatigue crack growth at cold expanded holes using linear elastic fracture mechanics and superposition”, Engineering fracture mechanics 78, p.p. 1389-1406, 2011.

JUNE 2014 Leonardo Times

Residual stress.indd 33

09-Jul-14 12:07:42

RVD

PLANETARY EXPLORATION SYMPOSIUM HAALBEELD FOTOGRAFIE

Robotic and manned missions to Mars and Jupiter

This year, the 28th Space Department has the honor of organizing the Annual Symposium of the VSV ‘Leonardo da Vinci’. On March 4, 2014, a total of 564 engineering students and Aerospace professionals visited the symposium and enjoyed seven extremely inspiring talks and interactive discussions. This day, themed ‘Planetary Exploration’ focused on future and planned robotic and manned missions to Mars and Jupiter. TEXT Nout van Zon, BSc Student Aerospace Engineering, Symposium Affairs of the 28th Space Department

Humankind was once drawn into unknown lands and across vast oceans. Now is the time to set our sights on the heavens, to discover, explore and seek answers to the many questions we have about our solar system. The innovations needed to one day set foot on Mars are the challenges that today’s students, the future generation of Aerospace engineers, will develop on. Man has always been driven to discover new worlds and explore the unknown whilst pushing the boundaries of modern day scientific and technological limits as far as possible. Space exploration is one of the Faculty of Aerospace Engineering’s main goals for the future and our contributions to these missions are growing. Furthermore, delivering value to the public is essential in maintaining a sustainable and lasting human presence in space exploration endeavors. Robotic and manned exploration of Mars and Jupiter will bring the world together by sharing the excitement that was once

Leonardo Times JUNE 2014

achieved by the Apollo program. It is in our nature to explore the unknown, to question the known and to expand our human presence across the next frontier. EXPLORATION OF JUPITER, EUROPA AND OTHER ICY MOONS Opening the day was Prof. Roger-Maurice Bonnet, former Director of Science at the European Space Agency. Currently, he is the president of the Committee on Space Research (COSPAR) and executive director of the International Space Science Institute (ISSI) in Bern. Professor Bonnet gave the audience a very informative and interesting introduction to the discovery of Jupiter, the scientific revelations already made about this planet and the finding of the icy moons Ganymede, Callisto, Europa and Io. As a conclusion, he noted the scientific value of exploration of the icy moons and the importance of the upcoming Jupiter mission JUICE. The chairman of the day Prof. Heinz Stoewer,

VSV ‘Leonardo da Vinci’ member of honor, seamlessly introduced our next speaker Dr. Bert Vermeersen, co-investigator of the GALA and PRIDE instruments on the JUICE mission. Dr. Vermeersen is an associate professor at our own faculty here in Delft and following the introduction by Prof. Bonnet continued to dive into the JUICE mission. He gave a very clear summary of the habitability of the four icy moons concluding that Europa had the four crucial aspects required: liquid water, stable environment, essential elements and chemical energy. He then continued to discuss the two JUICE instruments, of which he is co-investigator. These are the Ganymede Laser Altimeter (GALA) and the Planetary Radio Interferometer & Doppler Experiment (PRIDE). Finally he concluded by showing a slide of the current JUICE team and emphasizing the need for young engineers to join as “[Bert Vermeersen and his team] will be retired by the time the JUICE mission arrives in

JURRIAAN BROBBEL

Figure 1. Prof. Jean-Pierre Bibring explains what we know about Mars

2030”. LONG DURATION MANNED FLIGHTS Together with the Jovian system, Mars is also one of the most exciting and interesting targets for planetary exploration in our solar system. The Dutch ESA Astronaut André Kuipers was the next speaker and combined his medical background and space flight experience to discuss the human aspects of flying a long duration mission to Mars. Kuipers strongly believes that one day, we will eventually set foot on the red planet, however that many technical hurdles will first need to be overcome. By comparing the human aspects of space flight on the International Space Station and on a long mission to Mars, he was able to make some very convincing conclusions. First and foremost, André Kuipers mentioned the psychological aspects of a manned mission to Mars. During his six months stay on the ISS he was always able to look at the Earth, our home. Furthermore, the presence of the Soyuz spacecraft ready to evacuate the entire crew and bring them safely back to our planet within a matter of hours was reassuring. On a mission to Mars, home is just a pale blue dot in the vastness of the universe and there’s no turning around enroute. This brought him to his next conclusion being the risk of an inflight failure

of a subsystem or a medical emergency of one of the crewmembers. Once again, there is no turning around or Soyuz to bring the crew safely back to Earth. Furthermore, the ESA astronaut mentioned the issue of consumables including food and drinking water. The ISS has proven technology in recycling water however; it is also continuously visited by resupply vehicles with fresh food. On a mission to Mars, all food must be taken with or it must be grown onboard. Finally, Kuipers believes the biggest technical hurdle to overcome for a manned flight to Mars is protecting the crew against the extremely strong radiation experienced once outside of the protective radiation belt surrounding the Earth. MARS, A MOST EXCITING DESTINY In the afternoon, the Symposium continued to look at Mars. Th next speaker was Prof. Jean-Pierre Bibring from the Institut d’Astrophysique Spatiale. Prof. Bibring has an extensive experience with missions exploring the solar system, in particular Mars, including being responsible for the OMEGA instrument that was launched as part of the Mars Express mission. He gave a very clear and intriguing introduction to what is already known scientifically about the red planet. He discussed the origin of the planet’s red color and its very interesting surface features. Following was Richard Cook from the NASA Jet Propul-

sion Laboratory in the United States. As deputy program director of the Curiosity mission, he is an excellent expert on this operation and shared the very exciting engineering story of the rover. From the design hurdles to the spectacular landing of the Curiosity on the Mars, he guided us through the entire engineering process. Concluding the Mars segment was Vincenzo Giorgio, Vice President Domain Exploration and Science for Thales Alenia Space from Torino, Italy. He gave the audience insight into the status of the ExoMars 2016 and 2018 missions. His focus was on the 2018 landing mission that will deploy a European rover on the red planet. GLOBAL EXPLORATION ROADMAP Concluding the symposium was Prof. Johann-Dietrich Wörner, chair of the Executive Board of DLR and chair of the ESA Council. Heading the ESA Council Professor Wörner discussed the future global exploration roadmap of the world’s space agencies and the politics involved in the space industry. “Space is not just about science, it is also about politics. Whatever we do in space has to be legitimate, since everything we do is funded by tax money. Therefore, the first and most important question we always ask is the ‘why-question’ as Wörner explained during his presentation. In the end, if we are unable to generate public interest for going to Mars JUNE 2014 Leonardo Times

JURRIAAN BROBBEL

Figure 2. Prof. Johann-Dietrich Wörner discusses the necessity for international cooperation in Space endeavors.

it will be difficult to achieve this goal, “you have to convince people to such an extent that they come to you and ask you to please go to Mars to look for water.” In his talk and discussion of the ‘why-question’, he addressed the current top-down approach in the space industry: looking at what the agency wants and convincing the public it is important. “That has to change, in the future we should ask what the end users want and need, whether it be the general public or scientists, and the executives should make a strategy around that.” Jokingly Wörner continues to discuss the political role in the space industry by referring to the Apollo program “I believe that this nation should commit itself to achieving the goal, before this

decade is out, of landing a man on the moon and returning him safely to the earth”, President Kennedy stated on May 25, 1961. “Politically this is no longer correct, nowadays we would have definitely included ‘and a woman’.” So why was it possible to put a man on the moon in under a decade and have we not already set foot on Mars? Politics have changed according to Wörner, “in the past we often did something and explained only after the mission why we did it, we go to space to understand the past, the present and the future of the universe. This is a motivation that politicians don’t share, because within the four years of an election period, the change in the universe won’t be that big.” Ultimately, the ESA

Chairman is confident we will one day set foot on Mars, but not for another 30 years and not without a global cooperation of which the International Space Station is a great example. SPONSOR ATG EUROPE ATG Europe is the market leader in providing specialized engineering, scientific and technical services to Europe’s Aerospace, High Tech, Energy and Offshore industry. Exceptional engineering talent is hard to find, recruiting them even harder. We have a unique method to continuously map and recruit the brightest minds worldwide. We keep them close through our talent hubs; we know exactly where they are located, what motivates

The VSV ‘Leonardo da Vinci’ would like to thank the following companies for their contributions.

Leonardo Times JUNE 2014

SPONSOR OHB The German-based OHB AG is as European space and technology group one of the most important independent forces in European aerospace with more than 30 years of experience in developing and executing innovative space technology systems and structures. With its range of specific aerospace, aviation and telematics products, OHB is with 2,400 employees one of the most interesting employers in Europe. The “Space Systems” business unit focuses on developing and executing space projects. It is responsible for developing and fabricating Low-Earth Orbit and small geostationary satellites for navigation, research, communications and Earth observation. Its manned space flight activities focus on the assembly and fitting of the International Space Station (ISS). The exploration segment works on studies and models for exploring our solar system, primarily focused on Mars and the Moon. In addition, efficient reconnaissance satellites and broadband wireless transmission of image data form core technologies of our security and reconnaissance capabilities. The “Aerospace + Industrial Products” business unit is responsible for fabricating aviation and space products as well as other industrial activities. In this area, OHB has established itself as a significant supplier of aerospace structures. OHB is the largest German supplier

JURRIAAN BROBBEL

Figure 3. Richard Cook discusses the NASA missions to Mars, focusing on Curiosity

JURRIAAN BROBBEL

them and how to reach them. We provide this excellence to our excellence centers, but also to customers who recognize that competitive advantage starts with the right people. Medialabs graphic output dominates Europe’s Aerospace and High Tech visual communication. Understanding the technical complexity in detail, combined with a strong feel and talent for design is the key to our success. But the visuals can speak for themselves: www.atg-europe.com/medialab. ATG Projects provides high-end thermal, structural and flow engineering services to European High-Tech industries. We are involved in research and development programs focusing on composites and their application in the Aerospace, Semiconductor and Maritime industry. By combining a research level expertise with a down-to-earth problem solving approach, we are able to offer innovative solutions to complex issues on time, within budget. Using the high quality standards we already meet for over 15 years in the Aerospace industry and the experience in design, analysis and optimization we have built up, we support our customers throughout the whole development process, from the early stages of conceptual design up to and including full qualification and certification of the product.

Figure 4. Dr. Bert Vermeersen discusses the habitability of Jupiter’s icy Moons

of components for the Ariane 5 program and produces critical components for aircraft engines. In addition, OHB is an experienced vendor of mechatronic systems for antennas and telescopes and is involved in several major radio telescope projects. OHB is always looking for strong additions to our team, so if you are looking for a new challenge, do not hesitate and send an email to career@ohb-system.de.

SPACE DEPARTMENT The Space Department promotes astronautics among the students and employees of the faculty of Aerospace Engineering at Delft University of Technology by organizing lectures and excursions.

JUNE 2014 Leonardo Times

“We vlogen met een zucht...”

DENVER’S AIRPORT OF DOOM The story behind the world’s possibly most controversial airport

Big airport projects have often been prone to controversy. Schiphol’s Polderbaan project, Heathrow’s expansion, and the construction of Berlin’s new airport all suffered from political, financial or technical issues. However, Denver International Airport (often referred to as DIA) caused uproar in so many ways that it easily tops any other major airport project in history. TEXT Joris Stolwijk, Editor Leonardo Times, BSc student Aerospace Engineering

ight from the initial planning phase until decades after it’s opening, the airport made headlines with cost overruns, technical failures, provocative art displays and even doomsday conspiracies. What caused all the construction problems and how did it turn into a source of inspiration for conspiracy theorists? As one could expect, the issues are not completely unrelated from each other. THE POWER OF AIRLINES Denver International Airport was opened in 1995 after years of delays to replace the ageing Stapleton Airport that had been serving the Denver metropolitan area in Colorado, USA since 1932. Stapleton Airport suffered from typical old airport issues: it had an inconvenient runway layout, there was no more room for expansion and it received many complaints about noise from local residents. Forecasts for the air traffic demand in Denver predicted that a major capacity increase would be needed in order to meet future demands, and to achieve this it was decided to build a completely new airport, far outside the city limits on a land site twice the size of Manhattan. However, it

quickly turned out that the used forecasts were overly optimistic, and actually, the demand for flights went down after the construction of the new airport had commenced. How did this happen? At a major airport like Denver, demand is heavily dictated by the presence of major hub carriers. Stapleton was a hub for three carriers: Frontier, Continental and United. Frontier went bankrupt in 1986 and was taken over by Continental. Continental and United were not convinced at all by the new airport plans, but because it was vital to get these airlines on board to be hub carriers for the new international airport, both airlines had the power to demand very ambitious design changes before they would agree with the construction plans. Continental demanded that a two-level bridge, which would span dual taxiways, would be built to connect the main terminal with the Continental concourse, in addition to the planned tunnel system. Also, it was requested that a maintenance hangar and flight kitchen would be built and paid for by the airport. United demanded the same facilities, as well as that the bridge which connected

to the Continental concourse would be covered with opaque glass, such that passengers on their way to the United concourse would not be able to see the Continental aircraft looking shiny in the bright Colorado sun. All these requirements drove up the construction costs beyond limits, especially since the cost estimations were kept deliberately low for political reasons. Within a year after Continental signed a lease, it went into ‘Chapter 11’ bankruptcy, which allowed the airline to walk away from the signed contracts. In 1994, Continental decided not to operate from Denver at all. Thus, United remained as the only carrier to operate a hub at the new airport. Funnily enough, Continental and United were merged in 2010 as a result of yet another bankruptcy! THE BAGGAGE SYSTEM FROM HELL! United also demanded a baggage handling system with extremely ambitious specifications, and it demanded that the airport would only be opened once the baggage handling system functioned as specified. United only requested to have

Leonardo Times JUNE 2014

We vlogen met..0614.indd 38

03-Jul-14 23:08:33

this system implemented for its own concourse, but the Denver city government decided to build it for the entire airport. This made it a project of size and complexity never seen before, and there was not a single company that offered a design proposal, which would meet the requirements. Finally, BAE was given a practically blank check and unrestricted access to the building site to realize this system. However, the originally planned $183 million construction costs grew by $1 million per day, and even after years of delays, the system never really worked. During a demonstration on national TV, the system failed dramatically, literally eating baggage and throwing it around. Only United decided to use the system for outgoing flights, and in 2010, the system was completely removed, saving United $1 million per month on maintenance and repairs of the system. CONTROVERSIAL ART AND CONSPIRACY THEORIES While cost overruns, delays and technical problems are not that uncommon, what really makes this airport stand out is the bizarre, sometimes daunting art displays that are spread around the terminal. The first thing that greets you when driving towards the terminal entrance is a sculpture of a giant blue mustang with red glowing eyes (See Figure 1). This sculpture has its own scary story: the artist who created it, died because the horse structure fell on top of him when a cable snapped. It had to be completed by his family. After driving past the horse, the main Jeppesen terminal building catches the eye with its circus tent-like rooftop structure (See Visual), resembling the white caps of the Rocky Mountains nearby. Inside this terminal building, several mural paintings can be found, each depicting rather disturbing scenes. For instance, there is one mural where a giant military figure wearing a gas mask is shown, holding a machine gun and a sword, with scared children around him who are suffering in darkness (Figure 2). Another mural shows three dead women in a coffin, each of different ethnicity and with a burning city in the background. According to the artist, the whole collection of murals represents a story where man-made environmental destruction and genocide is eventually turned into a world coming together in harmony and peace. Conspiracy theorists think otherwise, saying that the murals show signs of an upcoming New World Order, where one large government formed by the world’s elite rules the world, after the world population has been reduced to two billion. They believe that the airport was merely built as a cover-up for an underground secret military base, with many facilities to even run an underground city. In these underground bunkers, chosen ci-

Figure 1. The horse statue at the airport entrance

Figure 2. One of the murals found in the terminal building.

vilians and the world elite can seek cover from an inevitable massive doomsday scenario. According to the conspiracy theorists, it is also not a coincidence that the runway layout resembles a swastikalike shape, and even the failing baggage system was simply a cover-up so that the tunnel systems can be used for other evil purposes. Absurd ideas can be found everywhere on the internet about what is hiding deep underneath the airport buildings, and as far-fetched as they seem, they do provide some food for thought. A DECENT AIRPORT IN THE END In the end though, for the average traveller, Denver International is just an airport like any other. And right now the airport functions perfectly fine, with a manually operated baggage system, and being a hub for United as well as a newly formed Frontier Airlines. Between 2005 and 2010,

readers of Business Traveller Magazine awarded “Best Airport in North America” to the airport for five years in a row. One might wander by a mural and wonder why on earth these scenes are shown at an airport – as if flying isn’t already scary enough –, but most will not really care as long as they don’t miss their connecting flights. References [1] Flyvbjerg, B. Buzelius, N. Rothengatter, W. (2003). Megaprojects and Risk: An Anatomy of Ambition. Cambridge: Cambridge University Press [2] “Five Years in a Row”. Wingtips (City & County of Denver Department of Aviation) 1 (10). January 2010. Retrieved July 11, 2012. [3] Jacang Maher, Jared (August 30, 2007). “DIA Conspiracies Take Off ”. Denver Westword. Retrieved July 11, 2012

JUNE 2014 Leonardo Times

We vlogen met..0614.indd 39

03-Jul-14 23:08:55

FLICKR / JON OSTROWER

RESPONSE SURFACE BASED OPTIMIZATION Gradient-Enhanced Kriging of a high-pressure turbine

In the Aerospace industry, before the construction of a new aircraft or engine is even realized, its performance and characteristics are already known. This knowledge is obtained through simulations in a wind-tunnel or using computational fluid dynamics. But with complex aircraft and engines, the use of these simulations can be limited by the computational means that are available. For this master thesis, the application of surrogate modeling, with the design of turbines of aero-engines is studied. This master thesis was performed with Rolls-Royce Deutschland. TEXT Ir. Floris Huijbregts, MSc Graduate, Flight Performance and Propulsion

f computational simulations are used to optimize a design, this optimization will always be a compromise. The improvement of a particular design can go on for a long time, even if the gains become very small. Over time, these increments in performance will no longer be worth the effort or computations. At that point, a design is considered to be optimized, although it can remain a sub-optimal design. A method to deal with these computational simulations is to employ a so-called surrogate model. A surrogate model imitates the behaviour of its original function, and can provide predictions in doing so. In this definition, the computational simulations are regarded as the original function (true value) and expensive to obtain, e.g. they take a long time to be solved. The surrogate model is cheap to evaluate and thus allows a quick prediction of the function value. KRIGING During the thesis work, a modeling

method known as Kriging was applied. Kriging finds its origins in the mining industry (Cressie, 1990). When a particular mineral-field is discovered, it is interesting to know what quantity of minerals can be expected in this field. In order to get an estimate, test-drills can be performed. There are only a limited number of testdrills available, because they are expensive. Based on these test-drills, the entire lay-out of mineral deposits â&#x20AC;&#x201C; and thus the total amount, could be determined using Kriging. Rolls-Royce Deutschland is a company that uses surrogate modeling in the design optimization of their turbines. The surrogate model does not replace the simulations entirely, but is used in conjunction with it. In the research, the computational simulations are equivalent to the expensive test-drills. The amount of minerals is similar to the turbineâ&#x20AC;&#x2122;s efficiency. The goal is now to find the highest efficiency, using only a few computational simulations. The efficiency of a turbine is constrained

by the mass-flow through the turbine as well as the reaction. This last constraint is an indication for the amount of energy that is subtracted from the airflow through the turbine. For the constraints, a surrogate model can be built. Now, using the one surrogate model for the efficiency and two surrogate models for the constraints, the objective function to be optimized is defined. GRADIENT-ENHANCED KRIGING Kriging uses only the function value (efficiency) of the computational simulations. The faculty of Aerospace Engineering, however, developed a promising extension to this method; the use of gradientinformation (Dwight, 2012). This is referred to as Gradient-Enhanced Kriging (GEK). See it this way: you are out skiing and you need to go home, but your view is obstructed by a snowstorm. It is interesting to know your altitude, but it can be more useful to know the slope of your position (the gradient) to find your way back

Leonardo Times JUNE 2014

Gradient Kriging.indd 40

04-Jul-14 00:02:36

FLORIS HUIJBREGTS

FLICKR / JON OSTROWER

Figure 1. Visualization of a prediction based on the Kriging and GEK method, for a one-dimensional function

TURBINE DESIGN Its blade shape determines the turbine characteristics –efficiency, mass flow and reaction. All blades in one row are similar, and the shape is defined by a set of parameters. The entire three-dimensional shape is given by fourteen parameters; thus, a fourteen-dimensional surrogate model has to be constructed. Using the experience of Rolls Royce, the fourteen parameters are defined as a deviation from an existing turbine shape. This shape would be further optimized using the surrogate model. The initial surrogate model is built based on the observations of fourteen simulations.

0.8

0.6

0.4

0.2 0 0

0.5 x 1

Figure 2. The Rosenbrock function, used as example for optimization problems.

FLORIS HUIJBREGTS

OPTIMIZATION One could stop here and merely use the surrogate model to determine the opti-

mal shape of the turbine. However, because these are predictions, there is a need to improve this prediction as well; and to optimize the surrogate model. In order to do so, the predicted optimum is located. At this optimum, one can evaluate the true value; and perform a simulation with that particular turbine shape. Adding this piece of information is referred to as adaptive-sampling. The initial fourteen observations are an example of non-adaptive sampling. When one repeats this exercise, the prediction of the optimum can be improved. This method is only applicable when the problem has one local optimum, the global optimum. With multiple local optima, it would be possible to get trapped at a local optimum, not the global optimum. However, the turbine with which the exercise started was already a near-optimal turbine, based on Rolls Royce’ experience. In addition, the domain over which the variables could vary was relatively small. OUTCOME Because of a limitation in calculating the gradients, the gradient-information was added to the center point of the domain only. Subsequently, several iterations adding adaptive samples are performed. These adaptive samples did not contain 1

0.8

0.6

0.4

0.2 0 0

0.5 x 1

Figure 3. The GEK-prediction of the Rosenbrock function, using the gradient-information at the centre observation.

FLORIS HUIJBREGTS

Using the gradient-information in the surrogate model, the hope is to get a better prediction and thus find a more optimal turbine design. Of course, this gradientinformation has to be evaluated as well, using the computational simulations. It is, therefore, also of interest, if the increase prediction-accuracy weighs out the increase in computations to obtain the gradients.

home, located in the valley.

any gradient-information. When Kriging was compared to GEK, it was possible to see that the latter not only converged quicker towards an optimal prediction, the predictions itself were also more accurate. Although the construction of the surrogate model showed to be very sensitive, a clear improvement of GEK was observed over kriging, using the same number of computational simulations. OUTLOOK In the end, the addition of a small piece of information showed significant effects. Not only Rolls-Royce, but other design organizations can as well benefit from surrogate modeling with gradient-information. The interesting thing is, the surrogate model can reflect any objective function or quantity. With a proper problem definition, limited observations of this quantity, GEK can be used in many different areas. Possible improvements in constructing the surrogate model, using the GEK approach, will further boost its possibilities. PERSONAL EXPERIENCE Before starting my thesis research, I figured I wanted to work with an aerospace company. Not only for the experience in the industry, but also to see how research in a company compares to that at a university. As an FPP-student, it required some hard work to get used to this more statistical type of assignment. All in all, it is safe to say that I learned many lessons with Rolls-Royce, and not just on engineering! References [1] Cressie, N., “The origins of Kriging”, Mathematical Geology, vol. 22, pp/239252, April 1990 [2] Dwight, R.P., De Baar, J., Azijli, I., “A Tutorial on Adaptive Surrogate Modelling”, 2012

JUNE 2014 Leonardo Times

Gradient Kriging.indd 41

04-Jul-14 00:03:01

WINDMILLING OF TURBOFAN ENGINE Calculation of Performance Characteristics of a Turbofan Engine under Windmilling. The turbofan is a type of air breathing jet engine that finds wide use in aircraft propulsion. During the normal operation of a turbofan engine installed in aircraft, the combustor is supplied with fuel, flow to the combustor is cut off and the engine runs under so called Windmilling conditions being driven only by the ram pressure ratio by producing drag. In-depth analysis is done to study the performance characteristics at this state. TEXT Aditya Ramanathan, MSc Student Aerospace Engineering, ISAE Toulouse

turbofan is a modified version of a turbojet engine. Both share the same basic components, but the turbofan has an additional turbine to turn a fan located at the front. This is called a “two-spool” engine. Some of the air from this fan enters the engine core for combustion while around 90% of it goes through a duct called the bypass duct. When the fuel supply to the combustion chamber is cut off, then the engine is said to run in the windmilling condition, i.e., the spool rotation is only due to the ram pressure ratio. It is an extreme operating condition and is usually avoided. During windmilling, the compressors and turbines work far away from their design points with poor efficiencies and the fan operates in a very different way as if it was self-windmilling. The objective of this project is to create a model to predict the mass flow rate and the speed of the two spools at the windmilling condition. This requires the coupling of the 3-D component simulations with the simulation tools. The development of these

models relies on the use of characteristic fields, which are formulated using the map fitting tools or the artificial map parameter (β) and focus on the issue of extrapolation of these fields, particularly at the fan stages. Testing on the test bench then validates these extrapolation techniques. From these models, we study the performance compared to the operation in windmilling in terms of the mass flow rate, dilution and stability range of the combustion chamber. A parametric study will evaluate the area of convergence of the models on the flight envelope of the engine. Later, an analysis will be made on the influence of the secondary nozzle section and/or on estimating losses in the bearings of windmilling. EXPERIMENTAL SETUP The development is carried on an engine with a high bypass ratio, unmixed flow geared turbofan by Price induction. The engine is used for a two- to four seater private light aircraft for a maximum takeoff weight between 1550 and 2550 kg.

Modern materials such as composites and light alloys are used to achieve an optimized weight. The diameter of the fan is less than fourteen inches and consists of fourteen blades and forty for the Outlet Guide Vanes, which is driven by a single stage low-pressure turbine. The engine core consists of a centrifugal compressor and a single stage high pressure turbine. The engine is provided with instruments to measure the steady temperature and pressure at different locations. Tests are carried out at the test facility with minor changes in order to simulate the windmilling conditions at the ground. During operation a flexible air tight tube is used to connect the engine with the fan duct in order to reduce vibrations from the fan that would affect the engine. SIMULATION SOFTWARE The simulation uses software called PROOSIS propulsion object oriented simulation software developed by Empresarios Agrupados for the engine cycle analysis and the CFD code for accurate 3-D com-

Leonardo Times JUNE 2014

Windmilling of turbofan.indd 42

03-Jul-14 23:23:59

ADITHYA RAMANATHAN

PRICE-INDUCTION.COM

Figure 1. Mach number on Fan (left) and OGV (right) of hub (top) and shroud (bottom) ADITHYA RAMANATHAN

PRICE-INDUCTION.COM

Figure 2. Steady state windmilling

Figure 3. Stagnation pressure at the rotor exit

ponent simulations. It also models both continuous and discrete model systems. It impacts different configurations and preliminary dimensioning of equipment, mono-point and multi-point design, parametric studies, sensitivity analyses, customer deck generation, optimization studies, multi-fluid models, maps handling, etc. The entire model is created by linking the different component models in a graphical user friendly interface. There are simple averaging techniques available to handle the 3D-0D-component data exchange though the boundary conditions of the whole engine model remain the same. While the boundary conditions of the 3-D simulations are automatically fed by the PROOSIS to the CFD software. The figure shows the schematic view of the DGEN 380 model in PROOSIS. PROOSIS primarily performs the simulations based on the thermodynamic gas turbine cycles using averaged variables to describe the flow properties. The different component characteristics are fed through the maps obtained from either the CFD simulation or the test results. The modeling is determined by the thermodynamic and the fluid functions. The calculation of the performance at the required operating point is done by solving a set of linear mass and power balance equations for all the components using NewtonRaphson method. COMPONENT MFT MAPS The Map Fitting Tool is a representation based on similarity parameters. These

Figure 4. PROOSIS Model

tools are very accurate and result in much smoother maps. The similarity parameters used are scalars of the efficiency, mass flow and the rotational speed calculated separately for each of the rotating components. With the help of the steady experiment, PROOSIS was able to simulate a large envelop of conditions up to low pressure ratios using the MFT maps in order to run the steady state calculations. The graph in the figure shows steady state windmilling i.e. when the fuel is cut off and the pressure ratio drops and reaches a state where Π<1 but when the pressure ratio increases with higher mass flow rate the engine operates under normal condition. The CFD simulations are first run for the fan blade alone and large separations were seen near the tip. Adding the Outlet Guide Vane blades and using the mixing technique for the fan and OGV interface had little effect on the flow across the fan. The turbine produced little work: roughly of the order of 10-15% of the design work. The high pressure compressor operates at a very low value of the pressure ratio and the fan operates with a pressure ratio of less than one with a very high increase in the bypass ratio. All the results suggested that the fan is the most important component to simulate with CFD. The CFD simulations were achieved for a rotational speed of 20% of the design speed. The fan and OGV behaved properly at the hub whereas separations appear at the top, especially for the OGV. The hub portion of OGV appears to be occupied with high Mach numbers whereas the shroud

portion is not much affected compared to the hub part. Further, the lower part of the fan blade compresses a little while the upper part expands the flow with an overall aerodynamic load of zero. CONCLUSION FUTURE WORK The present work has been dealt with the behavior of the fan stage of a high bypass ratio turbofan engine-out conditions by reproducing windmilling operation in a ground level test bed. The results demonstrate the challenges that arise in characterizing the flow due to the extremely low temperature and pressure variations. Work is in progress to complete the database with unsteady measurements to characterize the turbulent and unsteady components of the separated flow and provide a reference validation test case. Further work is constantly going on in improving the PROOSIS model. The extrapolation model is also being studied and improved to meet the new challenges. A detailed study of the MFT map methodology is also being studied for a step by step calculation procedure. References [1] W. Braig, H. Schulte et C. Riegler, « Comparative analysis of the windmilling performance of turbojet and turbofan engines », [2] N. García Rosa, J. Pilet, J.-L. Lecordix, R. Barènes et G. Lavergne, « Experimental analysis of the flow through the fan stage of a high-bypass turbofan in windmilling conditions

JUNE 2014 Leonardo Times

Windmilling of turbofan.indd 43

03-Jul-14 23:24:10

The world deserves a bright future. Just like yours. Locations Internships and graduation assignments

Corporate Head office Hengelo, Huizen, Eindhoven and Enschede

80 till 100 internships and graduation assignments every year in different fields of work

High Tech Jobs Domains

Thales is always looking for colleagues with who would like to work in a High Tech environment

o.a. Defence, Transportation Systems and Cyber Security

Leonardo Times JUNE 2014

To find out more, scan the QR code or visit thalesgroup.com/nl Air to air.indd 48

Together. Safer. Everywhere. 21-Jul-14 19:07:02

DESIGN FOR AIR-TO-AIR REFUELLING OPERATIONS New passenger and tanker aircraft design for AAR scenarios Air-to-air refuelling is a way to improve fuel efficiency of the overall transport system without waiting for the improvement of basic aviation technology. To take full advantage of such an operation, both passenger aircraft and tanker aircraft (which deliver required fuel to the passenger aircraft in halfway) must be designed accordingly. Moreover, the issues of safety and reliability must also be considered besides fuel economy. TEXT MO Li, AWEP Department, PhD Candidate at Flight Performance and Propulsion Chair

ir-to-air refuelling (AAR) has a long history in military operations, with extensive applications since the late 1940s and it soon became an indispensable part in modern air power. In the beginning, the motivations of AAR operations were quite straightforward: extending the range of combat aircraft in conditions where a suitable airbase was absent or too vulnerable. With the accumulation of operational experience, the doctrine of AAR operation has been enriched based on luxuriant advantages found in various aspects. Besides the great improvement in strategic and tactical flexibility, AAR is also found to be beneficial for many other perspectives, such as reducing the effect of burning fuel to transfer fuel (Nangia, 2006). Also, it is possible to achieve a lighter take-off weight for a given payload-range mission, which would in turn reduce the take-off field requirements and generate less noise emissions. Furthermore, when airborne, it is feasible to increase an aircraftâ&#x20AC;&#x2122;s fuel load to a gross weight significantly greater than its maximum take-off gross weight (Dougherty, 1996). Another perspective is a reduction in the required fuel reserves. In emergency situations (even in adversary air zone), tankers have

Air to air.indd 46

been proven to be able to support the warplanes with the fuel for their diversion range. Finally, considerable time savings can be achieved when compared to making multiple stops. AAR FOR CIVIL APPLICATIONS? With these advantages, aerial refuelling can also be beneficial for civil applications, though it has yet to be developed. Why is AAR not used in current civil aviation? One explanation is that unlike military aviation, airliners operate their aircraft with carefully planned routes. If they want to operate in a long-range route, they use long-range aircraft. This also encourages manufacturers to build jets, which can cover the longest potential route in the market. This statement explains the situation, but it does not mean the current transport system and aircraft design are optimal. There are some long-range routes, but in fact, 90% of all routes are shorter than 1,500nm (Kenway, 2010). These routes are far shorter than the design range of most aircraft. For example, the Airbus A320 has a design range of 2,600nm; wide-body aircraft are designed to fly over 4,0009,000nm. These aircraft are operated in a

very inefficient way with greatly oversized fuel tanks, wing areas, landing gears, etc. Regional jets are designed for shorter range. But in terms of passenger capacity and cabin comfort level, they cannot meet all the needs from markets. In East Asia for example, some airliners operate Airbus A380s and Boeing 747s in domestic flights, which cover less than one fourth of their design range. Such inefficient operations greatly reduce the profit of airliners, and could even result in no profit at all. At the same time, extra CO2, NOx and noise emissions are produced. It turns out that civil aviation does need the operational flexibility like the military does. Flexibility in civil applications is used for more profit and less emission rather than providing means for greater intimidation and destruction. Moreover, AAR in civil operations requires higher standards in terms of safety and comfort level. On the other hand, civil aviation manufactures have been experiencing a flattened technology progress curve for more than thirty years. After the massive application of supercritical airfoils and high bypass ratio turbofan engines, the rate of performance improvement went down. Recent advancements in aerodynamics, mate-

Leonardo Times JUNE 2014

09-Jul-14 12:10:18

rial and propulsion improve the overall airplane efficiency in the order of 1- 5%. Meanwhile, large aircraft are operated in inefficient ways as stated earlier, which may result in an efficiency decrease of up to 14% to 30%, depending on the route. But long range requirements do have a reason: if the aircraft is designed with relatively short range, even though it can be very efficient, the airliners are less likely to buy it. This is mainly because it lacks the flexibility to be operated in longrange routes. To overcome this dilemma, air-to-air refuelling is a possible solution, which provides airliners with the required range flexibility. This scenario is studied in a European project called RECREATE (REsearch on a CRuiser Enabled Air Transport Environment). In this project, a passenger aircraft specially designed for AAR operations is the “cruiser”, which is refueled by another special designed tanker aircraft. REFUELLING CONFIGURATION The first attempts at AAR operations were not feats of engineering prowess, but of aerobatic and acrobatic skill. There were numerous styles of connecting two aircraft in the sky. When the fancy stunt became a serious military operation, aviation engineers put a lot of effort in designing and testing many potential configurations. New series of technologies have been developed, including communication, navigation, fuel-transferring and flight control. Besides, also the standard for operation has been established and enriched. Nowadays, the AAR approach features staggered formation, with the tanker positioned above and in front of the receiving aircraft, which in turn performs the entire manoeuvre. To reduce the risk of civil AAR operations, it is reasonable to use the mature military experience and equipment as much as possible. But if the differences of two operations are overlooked, the effectiveness of civil AAR operations could be compromised. In order to have a proper understanding, all the possible fuel transfer systems were listed in a design option tree for the preliminary assessment and selection. The

flying boom system was chosen because a high fuel transfer mass rate is required for such large aircraft. The tanker aircraft should carry the boom and most of the associated systems, since the cruiser’s fuel consumption is more sensitive to weight increments because of their longer range. After that, all the possible relative cruiser and tanker positions were considered. Some infeasible options were discarded first, while the remaining ones were graded in a trade-off matrix, which includes twelve independent criteria covering important aspects. Surprisingly, it turned out that the current military approach got the lowest total score and the reversed approach (receiving aircraft flies above and ahead, while tanker performs approach manoeuvre) received the highest score. This innovative approach has great advantages in safety, the lowest requirements for the pilot training level, a good passenger comfort level and low costs for deploying the whole system, which are all crucial aspects in civil aviation. The main challenge comes from the fuel transfer system. Because the tanker should carry the major part of the fuel transfer system and the tanker stays below and behind the cruiser, the refuelling boom should be deployed against the incoming flow, which is not a mature technology and may encounter aeroelastic issues. As a spin-off, the forward extending boom had become an interesting and challenging research topic. A preliminary study has already provided several solutions to fight the static aeroelasticity. No evidence indicates the infeasibility of the forward extending boom. In next step, dynamic aeroelasticity like flutter will be analysed. IMPACT ON PASSENGER AIRCRAFT DESIGN When selecting the refuelling configuration, it is beneficial for the overall airtransport system to put the most part of the refuelling system on the tanker aircraft. Besides, the tankers, rather than the cruisers, should also carry out the formation manoeuvre. It results in minor

Figure 1.The configuration of civil AAR formations is different to the military approach, which is favourable in terms of many crucial aspects in civil aviation.

modifications in terms of the cruiser’s refuelling system. The major changes to the cruiser are the design parameters. To design the aircraft, a series of aircraft design tools have been developed and adapted to the scenario. Generally speaking, AAR operations allow a cruiser aircraft to be designed with much less fuel capacity, thus less structure weight and less engine thrust, which would reduce the required wing area. Again, a smaller wing area would reduce the structure weight and engine thrust as well. For a mission length of 5,000nm with one refuelling operation in the middle, the snowball effect would result in about 25% reduction in the operative empty weight. Weight reduction is a main reason why AAR could result in fuel savings. Actually, the smaller wing-area will cause some decrease in aerodynamic efficiency because of size effects but the overall gaining is dominant. Besides the weight reduction, AAR also allows to reduce fuel reservations. Nowadays, the passenger aircraft flies over ocean with relatively straightened paths, which is achieved by extending diversion time in the ETOPS regulations. Because of these regulations, these aircraft carry more reserved fuel for a potential diversion flight, as they are further away from backup airports. To achieve a good overall efficiency in AAR operations, it is beneficial to put the refuelling rendezvous close to the tanker base. So the flight paths of cruisers are close to the tanker base already, which can be used as backup airports, allowing the cruiser to carry a much lower amount of reserve fuel, from 16% down to 9% of the total mission fuel. In general, refuelling once in a 5,000nm mission range will guarantee a cruiser to burn about 15% less fuel than a comparison direct flight aircraft with the same payload. If a tanker can refuel the cruisers with its fuel consumption less than the fuel saved by the cruisers, then the overall AAR operation is beneficial in terms of fuel consumption compared to the direct flight.

Figure 2. CFD methods are extensively used to create response surfaces, which form the foundation of the aerodynamic part of the design tool. Formation aerodynamics are also studied during the selection of refueling configurations.

JUNE 2014 Leonardo Times

Air to air.indd 47

09-Jul-14 12:10:25

Figure 3. 3D-view of the cruiser aircraft resembling a state-of-the-art aerodynamic layout.

Figure 5. Specially designed tanker aircraft are the key factor to achieve high overall fuel benefits. The new tanker has a joint-wing configuration, results in a lower operational empty weight.

TANKER AIRCRAFT DESIGN The winning or losing margin of AAR operations relies on the tanker efficiency, which is measured as fuel transferred to the cruiser divided by fuel consumed by the tanker itself. Like the cruiser aircraft, tankers are also designed with the same design tools. To make a comparison, current military tankers are also analyzed in the typical tanker mission of 250nm refuelling radius with the capability of refuelling three cruisers in one-flight. Each refuelling process takes about 20min and there are also 20min loiters between two refuelling operations. Current military tankers are conversions from existing transport aircraft. They are versatile in terms of the capability to carry some cargo, but they consume too much fuel since they are too heavy and have too much empty volume, which produces extra form drag. In this mission profile, the KC- 135 and A310-MRT have tankerefficiency of only 3 to 3.5. If they are used

Air to air.indd 48

for a civil AAR operations, almost all the benefits gained by the cruisers will be lost. It would be more reasonable to convert smaller aircraft like the A320 to be a specialized tanker. The maximum take-off weight should be increased by more than twenty tons to accommodate the required fuel capacity. Landing gear, maybe even wings should be strengthened; engine thrust should be increased too. After a large amount of modifications, it could achieve a tanker efficiency of 5.5. With this A320-ST tanker in place, the overall fuel savings of AAR operations could be around 7%. Completely new tankers are also to be designed. By doing this, the tanker efficiency could be increased at least to 7.5, making the overall AAR fuel savings go up to 10%. SYSTEM OF SYSTEM LEVEL The tanker efficiency also varies with the design refuelling radius and the number of refuelling counts per flight. Increasing

Figure 4. Sharing the same fuselage and the 5,000 nm design range with direct flight aircraft, the AAR cruiser can be much smaller and lighter, thus consumes less fuel.

the refuelling radius would allow for less devious cruiser paths, and thus shorter travel distances for the cruiser, but more fuel is consumed by the tanker. This is a trade-off on the system of system level. When a tanker is designed to refuel more cruisers per flight, the tanker efficiency tends to go down. But this also means fewer tankers should be deployed and operated to feed a certain fleet of cruisers, relieving the burden of flight management. According to the real geologic map, the tanker bases do not always sit in the middle of all the cruiser routes. The off-design AAR scenarios should be studied. The results could also change the inputs for aircraft design. In general, the RECREATE project opens more design topics than it closes.

References [1] Nangia, R.K., “Operations and aircraft design towards greener civil aviation using air-to-air refuelling”, The Aeronautical Journal. 705-721, Nov., 2006. [2] Dougherty, S.J., Colonel, L., “Air Refueling: The Conerstone of Global Reach— Global Power”, Maxwell Air Force Base, Alabama, AU/AWC/RWP076/96-04, 1 April 1996. [3] Kenway, G.K.W, Henderson, R., “Reducing Aviation’s Environmental Impact Through Large Aircraft For Short Ranges”, 48th AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition, AIAA 20101015, 4 - 7 January 2010.

Leonardo Times JUNE 2014

09-Jul-14 12:10:28

NASA

Column

LOW TOXICITY ROCKET PROPELLANTS Hydrazine (N2H4) and its hypergolic mate nitrogen tetroxide (N2O4) are used on virtually all spacecraft and on a large number of launch vehicles. In recent years however, there has been an effort in identifying and developing alternatives to replace hydrazine as a rocket propellant. TEXT Jeroen Wink, Student Aerospace Engineering, Editor Leonardo Times

HYDRAZINE AND ITS DERIVATIVES Hydrazine (N2H4) is the most widespread rocket propellant. Since its development in the late fifties and early sixties, variations like Monomethyl Hydrazine (MMH) and Unsymmetrical dimethyl Hydrazine (UDMH) have been the standard propellant for nearly all spacecraft. On a technical level, hydrazine derivatives combine long-term storability with relatively good performance. Hydrazine is easily decomposed using a catalyst, simplifying monopropellant thruster designs. On top of that, hydrazine is hypergolic (ignites on contact) with a large range of storable oxidizers like Nitrogen Tetroxide and Nitric Acid variations, greatly simplifying the bipropellant thruster design. TOXICITY AND REACH REGULATION These merits come at a price. Hydrazine derivatives, including hypergolic rocket propellant, are highly flammable, toxic and carcinogenic. Furthermore, most variations have a vapor pressure higher than ambient air, which means that if a leak occurs during ground operations, the hydrazine will quickly evaporate and mix in the air, increasing the risk of inhalation by ground personnel. These adverse properties induce the need of strict safety precautions and regulations while handling

the propellants. Any operation involving hydrazine requires all personnel to wear so-called S.C.A.P.E. (Self Contained Atmospheric Protective Ensemble) suits, which greatly reduces mobility of the personnel. The environmental impact is also severe, resulting in strict regulations for the design and operations of ground equipment. These factors induce great risk and complexity in the development and operation of hydrazine-based systems, which in turn results in a high cost. Apart from the operational aspects, future legislation might complicate matters even more. The European Union’s REACH (Regulation conserving the Registration, Evaluation, Authorization and Restriction of Chemicals) regulation puts hydrazine on a list of substances that are candidates for complete abolishment in the near future. If this is implemented, no European country can allow its industries to work with hydrazine anymore. This will render further development of hydrazine-based spacecraft propulsion in Europe practically impossible. These factors have lead to a recent interest in low toxicity (or “green”) rocket propellants. Multiple ESA and EU projects are trying to identify and increase the tech-

nology readiness level of different potential hydrazine replacements. ENERGETIC SALTS: ADN, HAN AND HNF In Europe, most focus has been on a group of energetic, which include Ammonium Dinitramide (ADN), Hydroxylammonium nitrate (HAN) and Hydrazinium Nitroformate (HNF). These energetic salts form solid crystals and can be used as a low toxicity oxidizer in solid rocket motors. However, the crystals can also be dissolved in water for monopropellant applications acting as a hydrazine replacement. The combustion products of the propellants have relatively large molecules, but the high operating temperatures compensate this. ADN is the ammonium salt of dinitraminic acid and decomposes into nitrogen, oxygen and water. It was originally developed in the Soviet Union. In the early 1990s, the Swedish Defense Research Institute (FOI) started development on ADN as a nontoxic and smokeless solid rocket oxidizer for missile applications. Currently, multiple parties including ESA, NASA, Nammo Raufoss and Swedish Space Company are further developing ADN based liquid rocket applications. Dissolved in water,

JUNE 2014 Leonardo Times

column 0614.indd 49

04-Jul-14 00:06:00

ADN has a slightly lower performance but higher specific impulse density as hydrazine. In 2010, the Swedish Prisma mission was the first spacecraft to feature ADN based thrusters. Hydroxylamine Nitrate (HAN) was originally invented as a reducing agent for plutonium ions for nuclear energy applications. It also attracted great interest as green rocket oxidizer. Much like ADN, HAN is being developed as a smokeless propellant for tactical missiles by Raytheon. Dissolved in water, it potentially has a slightly higher performance compared to ADN. NASA is planning a demonstration mission (GPIM or Green Propellant Infusion Mission) by the end of 2015. Hydrazinium nitroformate (HNF) was developed in the late 1990s in cooperation between the TU Delft, TNO and APP. Like ADN and HAN, its initial application was to replace Ammonium Perchlorate as solid oxidizer. However most solid propellant combinations of HNF where plagued by a high-pressure exponent (which is unwanted in solid rocket motors). However in theory it has the same merits as ADN and HAN when dissolved in water. Additionally, it is the most energetic of the described salts, which might offer a performance increase compared to HAN and ADN. However, currently no spacecraft mission is planned to fly HNF based systems in the near future. HYDROGEN PEROXIDE AND NITROUS OXIDE The ionic liquids described above feature large combustion product molecules and

column 0614.indd 50

high operating temperatures. This imposes significant challenges for the development of durable catalytic reactors and often requires the combustion chamber and nozzle to be regenerative cooled. Liquids like Hydrogen Peroxide and Nitrous Oxide do not have these problems. Hydrogen peroxide is a wide spread substance, mostly used as a bleaching agent. For rocket applications purity between 85 and 95 percent is needed. The development of this so-called High Test Peroxide (HTP) took place in Germany in the late 1930s and 1940s. After the Second World War, the technology spread across the globe. It gained some popularity in the 1960s as oxidizer and featured on the British Black Arrow missile. It was also used to power the main engines of NASAâ&#x20AC;&#x2122;s X-15 spaceplane. Furthermore, it is used in the gas generators of the Russian Soyuz rockets. It delivers relatively good performance while featuring storability; high density and non-toxic exhaust products. It does however have its challenges. In its high purity form, HTP can spontaneously decompose if subjected to pollution. Moreover, its viscosity is very low, making feed systems very prone to leakage problems. Numerous accidents have occurred throughout history due to these factors, making further development difficult Nitrous oxide, also known as laughing gas is considered to be the most safe of the propellants described in this review. It is being used as an anesthetic agent as well as an aerosol propellant in whipped cream canisters. It is also used to enhance car performance when injected

ROSCOSMOS

NASA

Figure 1. Crew wearing S.C.A.P.E suits for fueling NASA spacecraft

Figure 2. Proton Launch, toxic yellow fumes originating from the UDMH and NTO propellants

into the engine. Its wide spread nature and low toxicity makes it an attractive rocket propellant. It can be used both in monopropellant as in bi-propellant applications. It has a very high vapor pressure, which gives it self-pressurizing capabilities. This can greatly simplify the propulsion system. However, it also has by far the lowest performance of the described propellants. Dissolving hydrocarbons in the Nitrous oxide will increase the performance but also increase its explosive risks. The US Company Firestar has patented a number of nitrous oxide fuel blends dubbed NOFBX, where Nitrous oxide is mixed with fuels like ethane, ethene and acetylene as well as with a number of stabilizing agents to get a high performance monopropellant. An additional complication of nitrous oxide based systems is the fact that nitrous oxide is a very effective dissolving agent. This makes the development of durable catalyst a challenge. Furthermore, it makes development of composite tanks difficult since it will dissolve the epoxy of the composite. CONCLUSION The replacement of hydrazine by low toxicity alternatives will continue to be a challenge in the near future. Each alternative has its own specific drawbacks and since virtually all spacecraft used hydrazine derivatives, it will require a lot of effort of the space industry to gain similar experience and confidence with any alternative. On the other side, the development of green rocket propellants will have the potential to significantly reduce the environmental impact and cost of spaceflight.

Leonardo Times JUNE 2014

04-Jul-14 00:06:11

Global environmental concerns call for future innovative products. Currently, the aircraft industry is seriously considering to install Contra-Rotating-Open-Rotors (CROR) on mid-range 150-200 seater aircraft by the year 2020. Today, NLR (National Aerospace Laboratory) specialists work in close coรถperation with aircraft & engine manufactures to investigate noise, vibration and safety aspects of these novel aircraft concepts.

www.nlr.nl Advertentie_NLR.indd 1

8-12-2009 15:59:05

Come on board.

11017 SPR ADV Stork Fokker - Come on board - A4 - ENG.indd 1

07-11-11 15:29