《复合材料 Composites》课程教学资源（学习资料）第五章 陶瓷基复合材料_fiber21

Availableonlineatwww.sciencedirect.com BCIENGE DIRECT ACTA ASTRONAUTUGA PERGAMON Acta Astronautica 55(2004)409-420 www.elsevier.comlocate/actaastro Advanced ceramic matrix composite materials for current and future propulsion technology applications S Schmidt,*, S. Beyer, H. Knabe, H. Immicha, R. Meistring, A Gessler- a EADS-Space T ation, Munich, German EADS Domier Research and Technology, Friedrichshafen, Germany eEADS Corporate Research Centre, Munich, Germany Abstract Current rocket engines, due to their method of construction, the materials used and the extreme loads to which they are subjected, feature a limited number of load cycles. Various technology programmes in Europe are concerned, besides developing reliable and rugged, low cost, throwaway equipment, with preparing for future reusable propulsion technologies. One of the key roles for realizing reusable engine omponents is the use of modern and innovative materials. One of the key technologies which concern various engine manufacturers worldwide is the development of fibre-reinforced ceramics--ceramic matrix composites. The advantages for the developers are obvious--the low specific weight, the high specific strength over a large temperature range, and their great damage tolerance compared to monolithic ceramics make this material class extremely interesting as a construction material. Over the past years, the Astrium company(formerly DASA)has, together with various partners, worked intensively on developing components for hypersonic engines and liquid rocket propulsion systems. In the year 2000, various hot-firing ests with subscale(scale 1: 5)and full-scale nozzle extensions were conducted. In this year, a further decisive milestone was achieved in the sector of small thrusters, and long-term tests served to demonstrate the extraordinary stability of the C/sic material Besides developing and testing radiation-cooled nozzle components and small-thruster combustion chambers, Astrium worked on the preliminary development of actively cooled structures for future reusable propulsion systems. In order to get one step nearer to this objective, the development of a new fibre composite was commenced within the framework of a regionally sponsored programme. The objective here is to create multidirectional (3D) textile structures combined with a cost-effective infiltration process. Besides material and process development, the project also encompasses the development of special metal/ceramic and ceramic/ceramic joining techniques as well as studying and verifying non destructive investigation processes for the purpose of testing components c 2004 Published by Elsevier Ltd 1. Introduction diflerent propulsion concepts, the advanced develop- ment of reliable"throwaway items"paying special Within the scope of the national technology pro- attention to the main aspect of low cost, and prepa gramme ASTRA, work is being carried out on two ration for future reusable propulsion technologies for multiple use(30-50 launches). Apart from cutting manufacturing times and costs for" throwaway items for commercial launcher propulsion systems, one of 0094-5765/S-see front matter 2004 Published by Elsevier Ltd. doi:10.1016 ]. actaastro.200405.052

Acta Astronautica 55 (2004) 409 – 420 www.elsevier.com/locate/actaastro Advanced ceramic matrix composite materials for current and future propulsion technologyapplications S. Schmidta;∗, S. Beyera, H. Knabeb, H. Immicha, R. Meistringc, A. Gesslerc aEADS-Space Transportation, Munich, Germany bEADS Dornier Research and Technology, Friedrichshafen, Germany cEADS Corporate Research Centre, Munich, Germany Abstract Current rocket engines, due to their method of construction, the materials used and the extreme loads to which theyare subjected, feature a limited number of load cycles. Various technologyprogrammes in Europe are concerned, besides developing reliable and rugged, low cost, throwaway equipment, with preparing for future reusable propulsion technologies. One of the keyroles for realizing reusable engine components is the use of modern and innovative materials. One of the keytechnologies which concern various engine manufacturers worldwide is the development of 3bre-reinforced ceramics—ceramic matrix composites. The advantages for the developers are obvious—the low speci3c weight, the high speci3c strength over a large temperature range, and their great damage tolerance compared to monolithic ceramics make this material class extremelyinteresting as a construction material. Over the past years, the Astrium company(formerlyDASA) has, together with various partners, worked intensivelyon developing components for hypersonic engines and liquid rocket propulsion systems. In the year 2000, various hot-3ring tests with subscale (scale 1:5) and full-scale nozzle extensions were conducted. In this year, a further decisive milestone was achieved in the sector of small thrusters, and long-term tests served to demonstrate the extraordinarystabilityof the C/SiC material. Besides developing and testing radiation-cooled nozzle components and small-thruster combustion chambers, Astrium worked on the preliminarydevelopment of activelycooled structures for future reusable propulsion systems. In order to get one step nearer to this objective, the development of a new 3bre composite was commenced within the framework of a regionallysponsored programme. The objective here is to create multidirectional (3D) textile structures combined with a cost-e;ective in3ltration process. Besides material and process development, the project also encompasses the development of special metal/ceramic and ceramic/ceramic joining techniques as well as studying and verifying non destructive investigation processes for the purpose of testing components. c 2004 Published byElsevier Ltd. 1. Introduction Within the scope of the national technologypro￾gramme ASTRA, work is being carried out on two ∗ Corresponding author. di;erent propulsion concepts, the advanced develop￾ment of reliable “throwawayitems” paying special attention to the main aspect of low cost, and prepa￾ration for future reusable propulsion technologies for multiple use (30–50 launches). Apart from cutting manufacturing times and costs for “throwawayitems” for commercial launcher propulsion systems, one of 0094-5765/$ - see front matter
c 2004 Published byElsevier Ltd. doi:10.1016/j.actaastro.2004.05.052

410 S. Schmidt et al./Acta Astronautica 55(2004)409-420 RLIO, built by Pratt Whitney, for the American launcher Delta ll Tarbert ear relaforredt Relics ICRP) System studies undertaken at Astrium as well as structural and thermal analyses promise, thanks to the use of CMCs in thrust chambers of liquid-propellant rocket engines, substantial advantages compared to metal materials, which are currently utilized for Cerami campsites most launcher propulsion systems for the cooled combustion-chamber structures and nozzle extensions The main advantages comprise on the one hand the possible weight reduction and on the other hand the 20040060080010001200140016001800IC high resistance to thermoshocks as well as the stability to chemical attack versus the liquid propellants used. A further significant advantage is the high creep re- Fig. 1. Ratio of strength to weight as a function of temperature [1]. sistance and the extraordinary resistance to high tem- multiaxis states of stress occurring in actively cooled the main challenges comprises implementing a high thrust chambers necessitate a fibre composite that fea- hrust-to-mass ratio, i. e. high thrust (performance) tures sufficient shear strength in as many directions with a low engine mass-kN/kg; this applies to isotropic behaviour. The currently G owaway items as well as to future propulsion sys- available 2-directional fibre composites would proba- ns. In particular, against the backdrop of reusable bly only have a very limited lifetime. For this reason, propulsion system components, modern and efficient some years ago the development of a new material materials for realizing new construction concepts will system and manufacturing process, respectively, was play a decisive role commenced, with the objective of combining multidi Since the early 1990s, the Astrium company has rectional(3D)textile structures with a cost-effective been working on a pacesetting key technology, filtration method. Besides material and process amely ceramic matrix composites(CMCs).An development, the focus is on the development of ive of of cmcs prises carbonfibre-reinforced silicon carbide(C/SiC), concepts as well as the verification of non-destructive which is made using the liquid polymer infiltration test methods (LPI) Process. Fig. I illustrate the excellent ratio of Due to the above advantages inherent in ceramic rength to weight, in particular at high temperatures, fibre composites, currently engine manufacturers and compared to currently utilized metal materials, is research institutes are stepping up their activities just one special feature that makes it attractive as a geared towards the use of ceramics in rocket-engine construction material [1] thrust-chamber components In view of the extreme In the sector of space propulsion systems, to date thermomechanical loads in the combustion chamber ceramic fibre composites have been used primarily of liquid-propellant rocket engines, previous devel for radiation-cooled nozzle extensions and combustion opments concentrated above all on the use of ceramic chambers for small thrusters; the advantage here lies fibre composites in the less thermally loaded noz n the low specific weight (lightweight construction), zle extensions [1]. At Astrium, nozzle extensions dispensing with active cooling and at the same time have been developed to date made of C/Sic for the high service temperatures upper-stage engine AESTUS and successfully tested To date, the high temperature, lightweight mate- on the altitude test bench P4. 1 at DLR in Lampold rial has become an established material in particular hausen. A subscale nozzle on the scale of 1: 5 for the for nozzle extensions. Currently, commercial car- Ariane 5 main engine Vulcain was made of C/SiC and bon/carbon nozzles, manufactured by Snecma in also successfully tested on the research test bench P8 France, are being used for the upper-stage engine at DLR in Lampoldshausen at combustion-chamber

410 S. Schmidt et al. /Acta Astronautica 55 (2004) 409 – 420 Fig. 1. Ratio of strength to weight as a function of temperature [1]. the main challenges comprises implementing a high thrust-to-mass ratio, i.e. high thrust (performance) with a low engine mass—kN/kg; this applies to throwawayitems as well as to future propulsion sys￾tems. In particular, against the backdrop of reusable propulsion system components, modern and eIcient materials for realizing new construction concepts will playa decisive role. Since the early1990s, the Astrium companyhas been working on a pacesetting keytechnology, namelyceramic matrix composites (CMCs). An in￾teresting representative of the group of CMCs com￾prises carbon3bre-reinforced silicon carbide (C/SiC), which is made using the liquid polymer in3ltration (LPI) Process. Fig. 1 illustrate the excellent ratio of strength to weight, in particular at high temperatures, compared to currentlyutilized metal materials, is just one special feature that makes it attractive as a construction material [1]. In the sector of space propulsion systems, to date ceramic 3bre composites have been used primarily for radiation-cooled nozzle extensions and combustion chambers for small thrusters; the advantage here lies in the low speci3c weight (lightweight construction), dispensing with active cooling and at the same time high service temperatures. To date, the high temperature, lightweight mate￾rial has become an established material in particular for nozzle extensions. Currently, commercial car￾bon/carbon nozzles, manufactured bySnecma in France, are being used for the upper-stage engine RL10, built byPratt & Whitney, for the American launcher Delta III. System studies undertaken at Astrium as well as structural and thermal analyses promise, thanks to the use of CMCs in thrust chambers of liquid-propellant rocket engines, substantial advantages compared to metal materials, which are currentlyutilized for most launcher propulsion systems for the cooled combustion-chamber structures and nozzle extensions. The main advantages comprise on the one hand the possible weight reduction and on the other hand the high resistance to thermoshocks as well as the stability to chemical attack versus the liquid propellants used. A further signi3cant advantage is the high creep re￾sistance and the extraordinaryresistance to high tem￾peratures compared to metal materials. However, the multiaxis states of stress occurring in activelycooled thrust chambers necessitate a 3bre composite that fea￾tures suIcient shear strength in as manydirections as possible, i.e. isotropic behaviour. The currently available 2-directional 3bre composites would proba￾blyonlyhave a verylimited lifetime. For this reason, some years ago the development of a new material system and manufacturing process, respectively, was commenced, with the objective of combining multidi￾rectional (3D) textile structures with a cost-e;ective in3ltration method. Besides material and process development, the focus is on the development of metal/ceramic joining techniques, engine analyses and concepts as well as the veri3cation of non-destructive test methods. Due to the above advantages inherent in ceramic 3bre composites, currentlyengine manufacturers and research institutes are stepping up their activities geared towards the use of ceramics in rocket-engine thrust-chamber components. In view of the extreme thermomechanical loads in the combustion chamber of liquid-propellant rocket engines, previous devel￾opments concentrated above all on the use of ceramic 3bre composites in the less thermallyloaded noz￾zle extensions [1]. At Astrium, nozzle extensions have been developed to date made of C/SiC for the upper-stage engine AESTUS and successfullytested on the altitude test bench P4.1 at DLR in Lampold￾shausen. A subscale nozzle on the scale of 1:5 for the Ariane 5 main engine Vulcain was made of C/SiC and also successfullytested on the research test bench P8 at DLR in Lampoldshausen at combustion-chamber

S. Schmidt et al./ Acta Astronautica 55(2004)409-420 411 ==冒 Fig. 3. 5-axis winding machines. Fig. 2. CMC production by infiltration and pyrolysis of polymers 2. 2. Production capabilities pressures of up to 80 bars. Further, very success- ful test campaigns with radiation-cooled combustion Based on the current production by EADS in chambers were carried out in the small-thruster sec Friedrichshafen of carbon fibre reinforced plastics tor, whereby the material was able to demonstrate its (CFRP)for the protective payload fairing of Ariane 5 (Speldra and Syldra), liquid-polymer infiltration was versus the propellants and combustion products developed by DaimlerChrysler Research for, amongst other things, space components, hot structures, and 2. Manufacturing, process technique and non re-entry technologies destructive investigation (NDD) methods In order to build axisymmetric components such as nozzle extensions. combustion chambers, etc 2. LP- two 5-axis winding machines with dimensions 3200×10,000mm2and2000×5000mm2(Fig.3) mer route. The coated C-fibre bundle is impregnated sions 500/800 x 5000 mm? machines with dimen- The C/SiC is made as shown in Fig. 2 via the poly- as well as two 4-axis windi with a powder-filled polymer and laminated to form are available, in particular for 3D components prepregs Analogously to the manufacturing technique For the autoclave hardening of the components for fibre-reinforced plastics, the structure is laminated, order to provide them with the so-called green body, compacted in an autoclave and cross linked, and then various autoclaves are available for smaller compo- pyrolized without pressure and without moulding nents(1000 x 3000 mm")as well as for large-space tools at temperatures of 1300-1900 K in inert ga structures(3500 x 8500 mm2)in the Friedrichshafen For further compacting(depending on the desired production centre(Fig 4) porosity ) re-impregnation is effected with a suitable For high-temperature treatment, two pyrolysis polymer followed by pyrolysis. The component may furnaces for component sizes of up to 2.5 m in diame- then be coated optionally with an protection layer ter and 3 m in height are available(the Munich-based hanks to the consistent advanced development ECM company ) of the LPI Route over the past years, using a ew reinfiltration polymer permitted reducing the 2.3. NDI methods re-impregnation cycles by 50% while retaining the mechanical strength characteristics. Conse- In view of, compared to metals, significantly ly, it was possible to cut the production costs, in anisotropic ceramic composite structure, the non- icular for large-scale structures, by approx. 25% destructive testing of C/SiC components already

S. Schmidt et al. /Acta Astronautica 55 (2004) 409 – 420 411 Fig. 2. CMC production by in3ltration and pyrolysis of polymers. pressures of up to 80 bars. Further, verysuccess￾ful test campaigns with radiation-cooled combustion chambers were carried out in the small-thruster sec￾tor, wherebythe material was able to demonstrate its long-term stabilityand high chemical compatibility versus the propellants and combustion products. 2. Manufacturing, process technique and non destructive investigation (NDI) methods 2.1. LPI-process The C/SiC is made as shown in Fig. 2 via the poly￾mer route. The coated C-3bre bundle is impregnated with a powder-3lled polymer and laminated to form prepregs. Analogouslyto the manufacturing technique for 3bre-reinforced plastics, the structure is laminated, compacted in an autoclave and cross linked, and then pyrolized without pressure and without moulding tools at temperatures of 1300–1900 K in inert gas. For further compacting (depending on the desired porosity), re-impregnation is e;ected with a suitable polymer followed by pyrolysis. The component may then be coated optionallywith an protection layer. Thanks to the consistent advanced development of the LPI Route over the past years, using a new rein3ltration polymer permitted reducing the re-impregnation cycles by 50% while retaining the same mechanical strength characteristics. Conse￾quently, it was possible to cut the production costs, in particular for large-scale structures, byapprox. 25%. Fig. 3. 5-axis winding machines. 2.2. Production capabilities Based on the current production byEADS in Friedrichshafen of carbon 3bre reinforced plastics (CFRP) for the protective payload fairing of Ariane 5 (Speldra and Syldra), liquid-polymer in3ltration was developed byDaimlerChrysler Research for, amongst other things, space components, hot structures, and re-entrytechnologies. In order to build axisymmetric components such as nozzle extensions, combustion chambers, etc., two 5-axis winding machines with dimensions 3200 × 10; 000 mm2 and 2000 × 5000 mm2 (Fig. 3) as well as two 4-axis winding machines with dimen￾sions 500=800×5000 mm2 and 200=800×2000 mm2 are available, in particular for 3D components. For the autoclave hardening of the components in order to provide them with the so-called green body, various autoclaves are available for smaller compo￾nents (1000 × 3000 mm2) as well as for large-space structures (3500 × 8500 mm2) in the Friedrichshafen production centre (Fig. 4). For high-temperature treatment, two pyrolysis furnaces for component sizes of up to 2:5 m in diame￾ter and 3 m in height are available (the Munich-based ECM company). 2.3. NDI methods In view of, compared to metals, signi3cantly anisotropic ceramic composite structure, the non￾destructive testing of C/SiC components already

412 S. Schmidt et al./ Acta Astronautica 55(2004)409-420 Fig. 5. Testing various specimen plates by means of thermography and ct Fig. 4. Autoclave hardening of huge space structures. CT measurement lies here in the exact localization in particular the visualization of the depth position during production is a decisive criterion as regards the of the defect, as well as in the simple estimation of lifetime and reliability of highly stressed components In particular defect interpretation and the correla the size of the defect in all three spatial directions by tion of the various methods are not yet completel means of the reconstructed images understood. Currently, at Astrium diverse standard procedures for the non-destructive testing of C/Sic 2.3.1. Impulse thermography components, such as thermography, X-ray technology A mobile and proven method for determining com- and ultrasonic technology, are in use. With the aid of ponent qualities is impulse thermography, which has the NDI methods, possible production defects such already been very successfully tried and tested in the as delaminations, pores, and cavities, etc. as well as development of nozzles and combustion chambers.In component conditions before and after testing are to the case of impulse thermography, the component re- be detected. In order to improve the prediction poten mains stationary, and the surface of the component to tial and minimize risks, a comprehensive investigation be tested is warmed very homogeneously with sp programme was launched a short while ago at As cial flashbulb heat in the milli-to microsecond range trium. The focus and objective of such investigation by some few degrees. If no diflerences in material or is to prepare a so-called defect catalogue which is to tructural damage such as, for instance, delaminations serve as a reference for the application of the differ- occur, this thermal impulse penetrates uniformly into the material. If for instance. there are delaminations or procedures, alternative methods such as computer other defects in the composite material, at this spot the and neutron tomography were studied. With respect thermal conductivity is disturbed and visualized via a to the later qualification of the individual methods, special software by means of differing colour codings first of all various C/SiC specimen plates with de- fined, built-in defects at different depth positions and 2.3.2. Computer tomography with different production statuses were made and CT makes it possible to visualize the interior struc- then tested applying the NDI methods thermography, ture of objects non-destructively and without contact. ultrasonic testing and computer tomography(CT). By applying the latest technologies and faster alge ig. 5 shows as an example the test result of two rithms, a spatial resolution of up to l um and less difterent plates measured on the one hand using ther is achieved. As. for instance. the C/SiC combustion mography (left-hand image )and CT(right-hand im- chambers represent 3D axisymmetric bodies, the in- age). In the thermography image (left-hand image ), dustrial 3D CT method is extremely advantageous the differing depth position of the artificial defects is The system permits detecting changes in density as also clearly to be seen. a decisive advantage of the well as defects, together with a characterization with

412 S. Schmidt et al. /Acta Astronautica 55 (2004) 409 – 420 Fig. 4. Autoclave hardening of huge space structures. during production is a decisive criterion as regards the lifetime and reliabilityof highlystressed components. In particular defect interpretation and the correla￾tion of the various methods are not yet completely understood. Currently, at Astrium diverse standard procedures for the non-destructive testing of C/SiC components, such as thermography, X-ray technology and ultrasonic technology, are in use. With the aid of the NDI methods, possible production defects such as delaminations, pores, and cavities, etc. as well as component conditions before and after testing are to be detected. In order to improve the prediction poten￾tial and minimize risks, a comprehensive investigation programme was launched a short while ago at As￾trium. The focus and objective of such investigation is to prepare a so-called defect catalogue which is to serve as a reference for the application of the di;er￾ent methods. Besides the above-mentioned standard procedures, alternative methods such as computer and neutron tomographywere studied. With respect to the later quali3cation of the individual methods, 3rst of all various C/SiC specimen plates with de- 3ned, built-in defects at di;erent depth positions and with di;erent production statuses were made and then tested applying the NDI methods thermography, ultrasonic testing and computer tomography(CT). Fig. 5 shows as an example the test result of two di;erent plates measured on the one hand using ther￾mography(left-hand image) and CT (right-hand im￾age). In the thermographyimage (left-hand image), the di;ering depth position of the arti3cial defects is also clearlyto be seen. A decisive advantage of the Fig. 5. Testing various specimen plates bymeans of thermography and CT. CT measurement lies here in the exact localization, in particular the visualization of the depth position of the defect, as well as in the simple estimation of the size of the defect in all three spatial directions by means of the reconstructed images. 2.3.1. Impulse thermography A mobile and proven method for determining com￾ponent qualities is impulse thermography, which has alreadybeen verysuccessfullytried and tested in the development of nozzles and combustion chambers. In the case of impulse thermography, the component re￾mains stationary, and the surface of the component to be tested is warmed veryhomogeneouslywith spe￾cial Oashbulb heat in the milli- to microsecond range bysome few degrees. If no di;erences in material or structural damage such as, for instance, delaminations, occur, this thermal impulse penetrates uniformlyinto the material. If, for instance, there are delaminations or other defects in the composite material, at this spot the thermal conductivityis disturbed and visualized via a special software bymeans of di;ering colour codings. 2.3.2. Computer tomography CT makes it possible to visualize the interior struc￾ture of objects non-destructivelyand without contact. Byapplying the latest technologies and faster algo￾rithms, a spatial resolution of up to 1
m and less is achieved. As, for instance, the C/SiC combustion chambers represent 3D axisymmetric bodies, the in￾dustrial 3D CT method is extremelyadvantageous. The system permits detecting changes in density as well as defects, together with a characterization with

S. Schmidt et al./ Acta Astronautica 55(2004)409-420 separation, side load) Qualification of measurement technology(pressure sensors at wall) e Investigation into material behaviour under extreme thermal-mechanical conditions Film/ Detector Manufacturing of complex contours with adapted Radiation Holder stifFener rings for buckling loads Demonstration and verification of the metallic/ Fig. 6. Measurement principle of 3D CT. ceramic joining technique 3.1.. Manufacturing and design respect to their type, geometry and position in the com Based on the thermal and structu ponent. It is therefore possible to visualize material analyses, two Vulcain scaled nozzles were made ap- defects in the component volume, to effect local reso- plying the LPI method. The required fibre angle and lution and hence to make a comprehensive statement the wall-thickness progression of the nozzle compo- as regards quality. In addition, a dimensional measure- nent were set via the winding technique so as to be ment,i.ea complete, quantitative coverage of the con- tailor made. Due to the side loads calculated, special tour, can be effected Downstream data processing can stiffener elements were necessary in order to pre- thus serve to determine wall thicknesses and represent vent buckling of the nozzle. By laminating on ring nominal-actual contour comparisons. Fig. 6 illustrates elements and subsequent ageing and pyrolysis, an this principle. Through the continuous advanced de- integral positive compound between nozzle and stiff- elopment of the industrial CT systems, in particular ening element was generated For mant facture of the detectors, components that are 800 X 800 mm two nozzles, a newly developed polymer system was in size can be tested used which permitted reducing the manufacturing time by approx. 30% compared to the old polymer Both nozzles were coated for the hot-firing tests with 3. Development and test of CMc components a CVD-SiC layer. One of the challenges involved the interface design between ceramic nozzle and metal 3. Vulcain subscale nozzle extension combustion chamber. In particular the high tempera- tures occurring at the interface in the case of an area Within the framework of the tekan and astra ratio of 5 represented a particular challenge. Thanks Programme, two Vulcain subscale nozzles on the scale to an angular flange design, the use of flexible high of 1: 5 and with an area ratio of s=5-45 were de- ure seals and special clamping ring problem could be solved. Fig. 7 shows the two coated igned, made using the LPI technique and subjected C/Sic nozzle extensions to hot-firing testing on the Astrium test bench F3 (Ottobrunn )as well as on the dlr test bench P& The development and test objectives of the C/Sic 3. 1.2. Hot-firing tests nozzle extension were The Vulcain subscale nozzle extension was tested in two test se he with a maximum cham To study the compatibility and function of oxida- ber pressure of Pc =40 bars, and a se tion/erosion protection coatings for different mix- quence which comprised one single load point, with ture ratios(O/F=5-8) Pe=80 bars and O/F=6, for the entire test duration of Investigation into nozzle flow, flow separation 32 s, which was suficient to have full-flowing condi- (transient, steady) tions in the nozzle extension installed the 40-bar load Comparison with Vulcain (full-scale lateral case was specially performed to visualize the transi- loads/separation data. tion process from free to restricted shock separation

S. Schmidt et al. /Acta Astronautica 55 (2004) 409 – 420 413 Fig. 6. Measurement principle of 3D CT. respect to their type, geometry and position in the com￾ponent. It is therefore possible to visualize material defects in the component volume, to e;ect local reso￾lution and hence to make a comprehensive statement as regards quality. In addition, a dimensional measure￾ment, i.e. a complete, quantitative coverage of the con￾tour, can be e;ected. Downstream data processing can thus serve to determine wall thicknesses and represent nominal-actual contour comparisons. Fig. 6 illustrates this principle. Through the continuous advanced de￾velopment of the industrial CT systems, in particular of the detectors, components that are 800 × 800 mm2 in size can be tested. 3. Development and test of CMC components 3.1. Vulcain subscale nozzle extension Within the framework of the TEKAN and ASTRA Programme, two Vulcain subscale nozzles on the scale of 1:5 and with an area ratio of  = 5–45 were de￾signed, made using the LPI technique and subjected to hot-3ring testing on the Astrium test bench F3 (Ottobrunn) as well as on the DLR test bench P8. The development and test objectives of the C/SiC nozzle extension were: • To studythe compatibilityand function of oxida￾tion/erosion protection coatings for di;erent mix￾ture ratios (O=F = 5–8). • Investigation into nozzle Oow, Oow separation (transient, steady). • Comparison with Vulcain (full-scale) lateral loads/separation data. • Upgrading/verifying of design tools (heat transition, separation, side load). • Quali3cation of measurement technology(pressure sensors at wall). • Investigation into material behaviour under extreme thermal–mechanical conditions. • Manufacturing of complex contours with adapted sti;ener rings for buckling loads. • Demonstration and veri3cation of the metallic/ ceramic joining technique. 3.1.1. Manufacturing and design Based on the thermal and structure-mechanical analyses, two Vulcain scaled nozzles were made ap￾plying the LPI method. The required 3bre angle and the wall-thickness progression of the nozzle compo￾nent were set via the winding technique so as to be tailor made. Due to the side loads calculated, special sti;ener elements were necessaryin order to pre￾vent buckling of the nozzle. Bylaminating on ring elements and subsequent ageing and pyrolysis, an integral positive compound between nozzle and sti;- ening element was generated. For manufacturing the two nozzles, a newly developed polymer system was used which permitted reducing the manufacturing time byapprox. 30% compared to the old polymer. Both nozzles were coated for the hot-3ring tests with a CVD-SiC layer. One of the challenges involved the interface design between ceramic nozzle and metal combustion chamber. In particular the high tempera￾tures occurring at the interface in the case of an area ratio of 5 represented a particular challenge. Thanks to an angular Oange design, the use of Oexible high￾temperature seals and special clamping rings, the problem could be solved. Fig. 7 shows the two coated C/SiC nozzle extensions. 3.1.2. Hot-4ring tests The Vulcain subscale nozzle extension was tested in two test sequences, one with a maximum cham￾ber pressure of pc = 40 bars, and a second test se￾quence which comprised one single load point, with pc=80 bars and O=F=6, for the entire test duration of 32 s, which was suIcient to have full-Oowing condi￾tions in the nozzle extension installed. The 40-bar load case was speciallyperformed to visualize the transi￾tion process from free to restricted shock separation

414 S. Schmidt et al./ Acta Astronautica 55(2004)409-420 E=45 E=5 =340mm =152mm 0:DD29 762000 Length=332 mm Fig. 9. Vulcain subscale nozzle during 40 bar hot-firing test(F3) Fig. 7. Vulcain subscale C/SiC nozzle extensions (1: 5 ). CRC-02 0009 10/11/2gg Fig. 10. Vulcain subscale nozzle on dlr test bench P Fig 8. Vulcain subscale nozzle on test bench F3 in Ottobrunn cap-shock pattern is clearly visible. Additionally, the thermographical imaging lens system is shown, poin ing nearly perpendicularly to the outer surface of the which is typical of parabolic rocket nozzles, i.e. the nozzle. The nozzle extension withstood the transition Vulcain or SSME type. Figs. 8 and 9 show the noz- process without structural damage zle extensions on test bench F4 and during the 40-bar hot-firing test. 3.2. Summary During the F3 hot-firing test(40 bars), particularly in the transition range from free to restricted shock Design challenge involving ceramic subscale separation, temperatures of up to 2300 K were mea- nozzle for first- or booster-stage application success- sured on the hot-gas side by means of thermogra- fully demonstrated phy. In addition to the high wall temperatures, large thermal gradients occurred, especially in the sector Side-load case during transient start-up of the stifFening elements. Temperature measurements shut-down yielded gradients of up to 650 K. Fig. 10 illustrates the Maximum buckling load case due to integral pres- combustion-chamber pressure test with P=80 sure difference between inner and outer nozzle wal bars at H0+ 10 s. The exhaust plume with the typical with strongly overexpanded core flow

414 S. Schmidt et al. /Acta Astronautica 55 (2004) 409 – 420 Fig. 7. Vulcain subscale C/SiC nozzle extensions (1:5). Fig. 8. Vulcain subscale nozzle on test bench F3 in Ottobrunn. which is typical of parabolic rocket nozzles, i.e. the Vulcain or SSME type. Figs. 8 and 9 show the noz￾zle extensions on test bench F4 and during the 40-bar hot-3ring test. During the F3 hot-3ring test (40 bars), particularly in the transition range from free to restricted shock separation, temperatures of up to 2300 K were mea￾sured on the hot-gas side bymeans of thermogra￾phy. In addition to the high wall temperatures, large thermal gradients occurred, especiallyin the sector of the sti;ening elements. Temperature measurements yielded gradients of up to 650 K. Fig. 10 illustrates the high combustion-chamber pressure test with pc = 80 bars at H0 + 10 s. The exhaust plume with the typical Fig. 9. Vulcain subscale nozzle during 40 bar hot-3ring test (F3). Fig. 10. Vulcain subscale nozzle on DLR test bench P8. cap-shock pattern is clearlyvisible. Additionally, the thermographical imaging lens system is shown, point￾ing nearlyperpendicularlyto the outer surface of the nozzle. The nozzle extension withstood the transition process without structural damage. 3.2. Summary Design challenge involving ceramic subscale nozzle for 3rst- or booster-stage application success￾fullydemonstrated: • Side-load case during transient start-up and shut-down. • Maximum buckling load case due to integral pres￾sure di;erence between inner and outer nozzle wall with stronglyoverexpanded core Oow

S. Schmidt et al./ Acta Astronautica 55(2004)409-420 Function of metallic/ceramic joining technique under real conditions Manufacturing of tailor-made ceramic structures with adapted stiffeners based on thermal and struc tural calculations Nozzle interface C/SiC material withstood high wall temperatures . Diameter: 490 mm 2300 K)and thermal gradients, especially for Wall thickness: 3.6 mm highest heat flux measured with restricted shock separation. Test results show excellent scalability of flow Length: 1360 mm phenomena to full scale, e.g o transition from free to restricted shock separation fo Nozzle exit: tion, cap-shock pattern and Mach disk Diameter: 1330 mn all thickness: 1. 4 mm To the best of our knowledge, world record test with ceramic nozzle extension, with regard to combus- tion chamber pressure and operational load case Knowledge gained at Astrium in designing, manu- Fig. 11. Section drawing of Aestus engine with c/Sic nozzle. facturing and testing of ceramic subscale Vulcain nozzle can be applied directly to nozzle design for Comparison between Aestus test results and stan- further first- or booster-stage nozzles, e.g. SSME dard construction(metal nozzle) type Flow features, i. e. free and restricted shock sepa Verification of reproducibility with respect to the ration. for sSme are identical to vulcain. due to manufacture of complex large-scale structures made of C/SiC applying the LPI pi Dimensioning load cases for SSME-type ce Optimization of the process route by using newly ramic nozzle are identical with successfully tested developed polymer systems. Vulcain-type ceramic nozzle. Implementation of a material characterizati gramme to determine material characteristics tests, ILS, thermal conductivity of the 33. Aestus nozzle extension nozzle laminate In 1998, development work on the C/sic extension Fig. 1l shows a section drawing of Aeatus engine nozzle commenced, the objective being to demonstrate with C/Sic nozzle extension the basic feasibility of making full-scale components by means of the LPI Process. A main focus was the 3.3.1. Manufacturing and design design(FEM, thermal)of fibre composite structures The leap from subscale to large structures such as taking into consideration the loads(vibration, etc. the Aestus nozzle represented a particular problem occurring in the case of Ariane 5. Further development Especially the process-induced component shrinkage and test objectives were occurring during manufacture and as a function of the fibre orientation had to be solved during production Preparation of future developments in the area of development with a special emphasis on adherence to quid propulsion the geometrical tolerances. Verification of interface design--joining technique Based on the FEM and thermal analyses, the re- between ceramic nozzle and metal combustion quired fibre angle and the wall-thickness progression of the Aestus nozzle were set via the winding tech- Optimization of the design tools(FEM and thermal) nique, as for the Vulcain subscale nozzle. Thanks to

S. Schmidt et al. /Acta Astronautica 55 (2004) 409 – 420 415 • Function of metallic/ceramic joining technique under real conditions. • Manufacturing of tailor-made ceramic structures with adapted sti;eners based on thermal and struc￾tural calculations. • C/SiC material withstood high wall temperatures (≈ 2300 K) and thermal gradients, especiallyfor highest heat Oux measured with restricted shock separation. • Test results show excellent scalabilityof Oow phenomena to full scale, e.g. ◦ transition from free to restricted shock separation. ◦ plume pattern for overexpanded core Oow opera￾tion, cap-shock pattern and Mach disk. – To the best of our knowledge, world record test with ceramic nozzle extension, with regard to combus￾tion chamber pressure and operational load case. – Knowledge gained at Astrium in designing, manu￾facturing and testing of ceramic subscale Vulcain nozzle can be applied directlyto nozzle design for further 3rst- or booster-stage nozzles, e.g. SSME type. – Flow features, i.e. free and restricted shock sepa￾ration, for SSME are identical to Vulcain, due to parabolic nozzle contours. – Dimensioning load cases for SSME-type ce￾ramic nozzle are identical with successfullytested Vulcain-type ceramic nozzle. 3.3. Aestus nozzle extension In 1998, development work on the C/SiC extension nozzle commenced, the objective being to demonstrate the basic feasibilityof making full-scale components bymeans of the LPI Process. A main focus was the design (FEM, thermal) of 3bre composite structures taking into consideration the loads (vibration, etc.) occurring in the case of Ariane 5. Further development and test objectives were: • Preparation of future developments in the area of liquid propulsion. • Veri3cation of interface design—joining technique between ceramic nozzle and metal combustion chamber. • Optimization of the design tools (FEM and thermal). Fig. 11. Section drawing of Aestus engine with C/SiC nozzle. • Comparison between Aestus test results and stan￾dard construction (metal nozzle). • Veri3cation of reproducibilitywith respect to the manufacture of complex large-scale structures made of C/SiC applying the LPI process. • Optimization of the process route byusing newly developed polymer systems. • Implementation of a material characterization pro￾gramme to determine material characteristics (creep tests, ILS, thermal conductivity) of the original nozzle laminate. Fig. 11 shows a section drawing of Aeatus engine with C/SiC nozzle extension. 3.3.1. Manufacturing and design The leap from subscale to large structures such as the Aestus nozzle represented a particular problem. Especiallythe process-induced component shrinkage occurring during manufacture and as a function of the 3bre orientation had to be solved during production development with a special emphasis on adherence to the geometrical tolerances. Based on the FEM and thermal analyses, the re￾quired 3bre angle and the wall-thickness progression of the Aestus nozzle were set via the winding tech￾nique, as for the Vulcain subscale nozzle. Thanks to

416 S. Schmidt et al./ Acta Astronautica 55(2004)409-420 Fig. 13. Prepreg manufacture (on the left )and laminated stiffening ring(on the rig Fig. 12. Production of the Aestus nozzle structure by means of the winding technique at the Friedrichshafen Production Centre. duction and process techniques, production time was this tailor-made composite layup, the nozzle contour cut by approx. 30%. The first hot-firing test took place could be made so as to be near net shape as well as in the year 2000 weight-optimized. Fig. 12 illustrates how the fibres are laid up by means of the winding technique at the 3.3.2. Hot-firing test Friedrichshafen production centre Within the framework of the test programme, The loads occurring on launching the Ariane ne- funded in-house, the structural integrity of and the cessitate providing a stiffening ring at the end of the thermal load on the C/SiC nozzle extension were ver- nozzle. In contrast to the nozzle structure, the stiffen ified in a sine-load vibration and vacuum hot-firing ing ring was made by means of the prepreg technique. test. The test was performed on the DLrs P4.2 test Fig. 13 shows prepreg manufacture on the left-hand facility in Lampoldshausen in the year 2000 side, on the right-hand side the stiffening ring is shown The scheduled and realized test time amounted to on the layup tool. The ring and nozzle were joine 150 s at a combustion-chamber pressure of 1l bars and with a mixture ratio [O/F] of 2.05. Fig. 14 shows Since development was launched, altogether 5 noz- on the left-hand side the integrated ceramic nozzle on zle hardware units have been made applying the LPI the P4.2 and on the right-hand side during the vacuum process. Thanks to continuous optimization of the pro-

416 S. Schmidt et al. /Acta Astronautica 55 (2004) 409 – 420 Fig. 12. Production of the Aestus nozzle structure bymeans of the winding technique at the Friedrichshafen Production Centre. this tailor-made composite layup, the nozzle contour could be made so as to be near net shape as well as weight-optimized. Fig. 12 illustrates how the 3bres are laid up bymeans of the winding technique at the Friedrichshafen Production Centre. The loads occurring on launching the Ariane ne￾cessitate providing a sti;ening ring at the end of the nozzle. In contrast to the nozzle structure, the sti;en￾ing ring was made bymeans of the prepreg technique. Fig. 13 shows prepreg manufacture on the left-hand side, on the right-hand side the sti;ening ring is shown on the layup tool. The ring and nozzle were joined subsequently. Since development was launched, altogether 5 noz￾zle hardware units have been made applying the LPI process. Thanks to continuous optimization of the pro￾Fig. 13. Prepreg manufacture (on the left) and laminated sti;ening ring (on the right). duction and process techniques, production time was cut byapprox. 30%. The 3rst hot-3ring test took place in the year 2000. 3.3.2. Hot-4ring test Within the framework of the test programme, funded in-house, the structural integrityof and the thermal load on the C/SiC nozzle extension were ver￾i3ed in a sine-load vibration and vacuum hot-3ring test. The test was performed on the DLRs P4.2 test facilityin Lampoldshausen in the year 2000. The scheduled and realized test time amounted to 150 s at a combustion-chamber pressure of 11 bars and with a mixture ratio [O=F] of 2.05. Fig. 14 shows on the left-hand side the integrated ceramic nozzle on the P4.2 and on the right-hand side during the vacuum test

S. Schmidt et al./ Acta Astronautica 55(2004)409-420 Ta2W-%32 :日1:36.71 NIMONIC TEMP-SENXXR Fig. 14. Ceramic nozzle on test bench 4.2(left-hand side )and during the vacuum hot-firing test. CHAMIER NIMONI 3.3.3. Summary The use of nozzle extensions made of C/Sic for upper-stage engines was successfully demonstrated with respect to NO//L NIMONIC EXTENSION Mechanical loads during the transient start-up and shut-down phase Manner of functioning of the metal/ceramic inter face Manufacture of a tailor-made ceramic structure with adapted stiffening ring Design, manufacture and hot-firing test of full-scale nozzle components. Weight reduction(60% compared to metal nozzle) Fig. 15. 400 N engine(flight version) 3.4. 400n combustion chamber construction. Further advantages compris Combustion chambers for apogee-and attitude-control engines for satellites are currently made of refractory Simplification of the construction method by redt heavy metals such as rhenium, iridium and plat ing the individual components(single-piece con- inum. Due to the high stability to chemical attack struction), hence reduced test effort. and high service temperature of up to 1850 K, the Increase in the permissible wall temperatures ofcur- refractory metals are used as the material for com- rently 1900-2200 K (with suitable layer system), bustion chambers. Besides high material and man ence increase in specific impulse(performance) ufacturing costs as well as the substantial use of. Reduction of engine mass of 30-50% raw materials, heavy metals exhibit a high density, amounting to more than 21 g/cm" Fig 15 depicts the In order to study the use of fibre-composite ceramics current 400 N engine(flight version ). for small thrusters, in 1998 the first hot-firing tests The potential offered by CMCs as a structural ma- were carried out at sea level with different C/SiC terial for small thrusters lies among other things in the combustion chambers. The propellant compatibil learly lower manufacturing costs compared to metal ity(MMH/N2O4), diverse clamping concepts, and

S. Schmidt et al. /Acta Astronautica 55 (2004) 409 – 420 417 Fig. 14. Ceramic nozzle on test bench 4.2 (left-hand side) and during the vacuum hot-3ring test. 3.3.3. Summary The use of nozzle extensions made of C/SiC for upper-stage engines was successfullydemonstrated with respect to: • Mechanical loads during the transient start-up and shut-down phase. • Manner of functioning of the metal/ceramic inter￾face. • Manufacture of a tailor-made ceramic structure with adapted sti;ening ring. • Design, manufacture and hot-3ring test of full-scale nozzle components. • Weight reduction (60% compared to metal nozzle). 3.4. 400 N combustion chamber Combustion chambers for apogee- and attitude-control engines for satellites are currentlymade of refractory heavymetals such as rhenium, iridium and plat￾inum. Due to the high stabilityto chemical attack and high service temperature of up to 1850 K, the refractorymetals are used as the material for com￾bustion chambers. Besides high material and man￾ufacturing costs as well as the substantial use of raw materials, heavymetals exhibit a high density, amounting to more than 21 g=cm 2 . Fig. 15 depicts the current 400 N engine (Oight version). The potential o;ered byCMCs as a structural ma￾terial for small thrusters lies among other things in the clearlylower manufacturing costs compared to metal Fig. 15. 400 N engine (Oight version). construction. Further advantages comprise: • Simpli3cation of the construction method byreduc￾ing the individual components (single-piece con￾struction), hence reduced test e;ort. • Increase in the permissible wall temperatures of cur￾rently 1900–2200 K (with suitable layer system), hence increase in speci3c impulse (performance). • Reduction of engine mass of 30–50%. In order to studythe use of 3bre-composite ceramics for small thrusters, in 1998 the 3rst hot-3ring tests were carried out at sea level with di;erent C/SiC combustion chambers. The propellant compatibil￾ity(MMH=N2O4), diverse clamping concepts, and

418 S. Schmidt et al./ Acta Astronautica 55(2004)409-420 Fig. 16. C/SiC combustion chamber on the P1.5 and during the Fig. 18. Modified C/SiC combustion chamber during the hot-firing as newly developed layer systems were carried out. The main objective of the material and component tests was to verify the long-term behaviour(>1 h) and to find out the maximum permissible component temperatures. The test time amounted to 5700 s at a combustion-chamber pressure of 1l bars Fig. 17. Modified C/SiC combustion chambers coated with different Fig. 18 shows the combustion chamber during the hot-firing test The hot-firing tests conducted since 1998 with investigating differing layer systems with regard to coated combustion chambers made of C/SiC yielded long-term deployment, comprised the main areas of important insights into the application potential of CMCs for small thrusters. The feasibility together at a combustion-chamber pressure of 10 bars. By with the positive effects on engine performance were varying the mixture ratio [o/F] in the range between verified 1.64 and 1.92, maximum wall temperatures of up to 1700C were determined. Fig. 16 illustrates the 3.5. Further developments C/SiC combustion chamber on test bench P1.5 in Lampoldshausen. In the course of developing cost-effective manu- In this year, further hot-firing tests with modified facturing methods for structural components stion chambers(Fig. 17)as well of fibre composites, in recent years various

418 S. Schmidt et al. /Acta Astronautica 55 (2004) 409 – 420 Fig. 16. C/SiC combustion chamber on the P1.5 and during the hot-3ring test. Fig. 17. Modi3ed C/SiC combustion chambers coated with di;erent layers. investigating di;ering layer systems with regard to long-term deployment, comprised the main areas of e;ort. The accumulated test time amounted to 3200 s at a combustion-chamber pressure of 10 bars. By varying the mixture ratio [O=F] in the range between 1.64 and 1.92, maximum wall temperatures of up to 1700◦C were determined. Fig. 16 illustrates the C/SiC combustion chamber on test bench P1.5 in Lampoldshausen. In this year, further hot-3ring tests with modi3ed and optimized combustion chambers (Fig. 17) as well Fig. 18. Modi3ed C/SiC combustion chamber during the hot-3ring test. as newly developed layer systems were carried out. The main objective of the material and component tests was to verifythe long-term behaviour (¿ 1 h) and to 3nd out the maximum permissible component temperatures. The test time amounted to 5700 s at a combustion-chamber pressure of 11 bars. Fig. 18 shows the combustion chamber during the hot-3ring test. The hot-3ring tests conducted since 1998 with coated combustion chambers made of C/SiC yielded important insights into the application potential of CMCs for small thrusters. The feasibilitytogether with the positive e;ects on engine performance were veri3ed. 3.5. Further developments In the course of developing cost-e;ective manu￾facturing methods for structural components made of 3bre composites, in recent years various textile