Rocket Engine Feasibility Study for the JAXA Future Transportation Reference System

Transcription

1 Trans. JSASS Aerospace Tech. Japan Vol. 14, No. ists30, pp. Po_1_1-Po_1_23, 2016 Rocket Engine Feasibility Study for the JAXA Future Transportation Reference System By Asuka IIJIMA, 1) Daisuke NAKATA, 1) Masatoshi SUGIOKA, 1) Nobuhiro TANATSUGU, 1) Kazuyuki HIGASHINO, 1) Shinji ISHIMOTO 2) and Nobuyuki AZUMA 3) 1) Muroran Institute of Technology, Hokkaido, Japan 2) Japan Aerospace Exploration Agency, Chofu, Japan 3) Japan Aerospace Exploration Agency, Tsukuba, Japan (Received August th, 201) Bioethanol is a suitable candidate as a reusable rocket engine fuel because it is carbon neutral and environmentally friendly. Based on the JAXA future transportation reference system, the feasibility of a gas generator cycle fueled by bioethanol was examined in this paper. The result is considered to be a general design guideline of an ethanol-fueled rocket engine. A combustion pressure of MPa, an expansion ratio of 2, a turbine efficiency of 0.6 and a pump efficiency of 0. enabled the engine to satisfy the required specifications with an Isp efficiency of In addition, a flow experiment of bioethanol under high temperature and high pressure was conducted. Resulting from the EPMA analysis, the presence of sulfur and coking was recognized on the copper-alloy tube. Key Words: Reusable Launch System, Bioethanol, Regenerative Cooling, Coking, Sulfur Nomenclature A c : cross-sectional area of the chamber A e : nozzle exit area A t : throat area BL : bleed ratio D c : diameter of the combustion chamber D e : diameter of the nozzle exit D t : diameter of the throat I sp : specific impulse L e : axial length of the diverging part L t : axial length of the converging part L* : characteristic length of the chamber P a : ambient pressure P c : chamber pressure P wall : pressure at a separation limit V t : volume of the converging part 1. Introduction As the number of rocket launches will increase in the near future, their environmental compatibility becomes much more important. Disposable rockets currently in use are discarded at each time of launch, emitting carbon dioxide not only in the firing phase but also in the fuel production process. For example, hydrogen forms a great deal of carbon dioxide at the time of production by steam reforming of hydrocarbon. As an environmentally compatible rocket fuel, carbon-neutral bioethanol (BE) is considered to be a suitable candidate. Bioethanol is a non-toxic, non-cryogenic propellant and does not produce any soot in fuel-rich combustion. However, ethanol is poorer in its I sp performance compared to the other major hydrocarbon fuels such as methane or kerosene. Therefore, a careful feasibility study is needed for the given mission. In this paper, a feasibility study on an ethanol-fueled booster engine for a reusable small satellite launcher 1) is conducted. The launcher consists of a reusable booster and a single use upper-stage and has the ability to launch a 00 kg satellite to the low-earth orbit of 00 km. The discussion here is confined to the booster engine and the total system design is presented in another paper 1). From a practical viewpoint, sulfur attack in a regenerating cooling path would be a serious problem when bioethanol is used as a rocket fuel. The experimental results of a fuel flow test are also shown in this paper. After a -second ethanol flowing test in a 2-mm copper tube, the effect of a sulfur attack was examined by an electron probe microanalyzer. 2. Required Performance Requirement specifications resulted from the total system design of the small satellite launcher 1) as shown in Table 1. Only the case of two engines is discussed here. A Gas Generator (GG) cycle is assumed in this study as a feeding system even though it is not included in the requirement. As a practical advantage, the GG cycle enables an easy change to turbine power, making straightforward development possible. The feasibility of a staged-combustion (SC) cycle is not mentioned here, but one can easily estimate the theoretical performance of a SC cycle engine from the results of GG cycle analysis described in this paper. Copyright 2016 by the Japan Society for Aeronautical and Space Sciences and ISTS. All rights reserved. Po_1_1

2 Trans. JSASS Aerospace Tech. Japan Vol. 14, No. ists30 (2016) Oxidizer Fuel 2-engines Table 1. Required performance. 1) Liquid oxygen Ethanol Vacuum thrust > 8 kn Sea level thrust > 06 kn Figure 2 shows the expansion ratio versus vacuum thrust and specific impulse. Dashed lines are the minimum required specific impulse and vacuum thrust. Since the exit diameter is fixed in this study, a higher expansion ratio means a smaller throat diameter, resulting in a smaller thrust. 3-engines Vacuum I sp Vacuum thrust Sea level thrust > 23 kn > 41 kn > 31 s Throttling 100% 0% Number of reusability Nozzle exit diameter Engine mass (sum of 2 or 3 engines) > 100 times < 1.6 m < kg 3. Performance Analysis Fig. 1. Expansion ratio versus the nozzle exit static pressure Engine main combustion chamber design A parametric study of the main combustion chamber was conducted by NASA CEA2, and the results are shown below Main combustion chamber pressure Since high combustion chamber pressure causes technical difficultly during the development process, a lower combustion pressure is desirable. On the other hand, it is better to maintain a pressure above the critical pressure of ethanol (6.14 MPaA) so as not to cause the unstable phase changes of the fuel through the injectors. Therefore the combustion pressure was set to MPaA Mixture ratio (O/F) The optimum mixture ratio (O/F) of a LOX/ethanol engine was examined by NASA CEA2. From the results, the optimum O/F was determined to be 1.6 in the case of frozen flow and 2.0 in the case of equilibrium flow. Since the actual mixing characteristics are usually in the middle of both cases, this study adopted an O/F of Contraction ratio (Ac/At) The contraction ratio is fixed at three Expansion ratio (Ae/At) When determining the expansion ratio, it is necessary to account for flow separation limit. Flow separation limit is estimated by Schilling s empirical formula 2). Fig. 2. Expansion ratio versus vacuum thrust and specific impulse Combustion chamber configuration Figure 3 shows the nozzle shape of the main combustion chamber. The converging angle of the combustion chamber is fixed at 20, and the diverging angle is fixed at 1. L t and L e are determined so as to satisfy the contraction and expansion ratios. The characteristic length L* of the combustion chamber was set to 1 m equivalent to the value of the RP-1 referring to NASA SP-12 3). L c is determined from L* by using Eq. (2). From the above results, the main parameters of the combustion chamber are defined as shown in Table 2. (2) (1) According to NASA SP ), having a 20% margin for the estimated separation limit is recommended. Figure 1 shows the nozzle exit pressure calculated by NASA CEA2. The red dashed line shows the allowable lower limit of the static pressure determined by Eq. (1) accounting for 20 % of the margin. From this, the expansion ratio was set to 2. We used maximum allowed nozzle exit diameter (1.6 m as shown in Table 1) in this analysis, so the throat diameter is uniquely decided according to the expansion ratio. As the expansion ratio increases, the throat diameter decreases. Fig. 3. Nozzle shape of the main combustion chamber. Po_1_18

3 A. IIJIMA et al.: Rocket Engine Feasibility Study for the JAXA Future Transportation Reference System Table 2. Determined parameters of the main combustion chamber. Combustion chamber pressure Pc MPa Fuel temperature at the injector Tinj_fu 390 K Oxidizer temperature at the injector Tinj_ox 90 K Mixture ratio O/F 1.8 Main combustion chamber temperature Tc 34 K Characteristic exhaust velocity C* 131 m Vacuum thrust coefficient Cf_vac Sea level thrust coefficient Cf_sea 1.98 Vacuum specific impulse Isp_vac s Sea level specific impulse Isp_sea 31.4 s Nozzle exit diameter De 1.6 m Expansion ratio ε 2 Throat diameter Dt 0.32 m Vacuum thrust F_vac 108. kn Sea level thrust F_sea kn Total mass flow rate m_tot 32.3 kg/s Fuel mass flow rate m_fu kg/s Oxidizer mass flow rate m_ox kg/s 3.2. Regenerative cooling Overview of regenerative cooling The ability of ethanol regenerative cooling is lower than that of hydrogen. Thermal decomposition and coking could occur in ethanol at more than 600 K. As the tank temperature is assumed to 300 K, the allowable temperature rise is limited ( T = = 300 K). In this study, the feasibility of the regenerative cooling by ethanol is discussed. In addition, the number of reusability is also estimated considering thermal stress and thermal fatigue. Although the pressure at the inlet of the regenerative cooling channel is one of the variables, we fixed this to 10. MPa, corresponding to 1. P c Shape of the regenerative cooling channel Figure 4 shows a model of a regenerative cooling system. Channel wall design which uses a width b, depth h and radial wall thickness t w is assumed. Table 3 shows the design parameters of the regenerative cooling channel. This study uses b = 2 mm with h/b = 6.0 at the throat section. At a different location, b and h are proportional to the local diameter of the nozzle. In the current manufacturing technology, the aspect ratio (h/b) of 10 1 is possible. It is technically difficult to further reduce the thickness t w of the channel wall (< 0. mm). Table 3. The configuration of the regenerative cooling channel. Cooling channel width b 2.0 mm Cooling channel depth H 12.0 mm Cooling channel aspect ratio h/b 6.0 Wall thickness between the channels t s 1.2 mm Cooling wall radial thickness t w 0. mm Fig. 4. A Model for regenerative cooling system Result of the calculation The hot-gas heat transfer coefficient was calculated using the Bartz equation 3) and the coolant-side heat transfer coefficient was calculated by Dittus Boelter equation. To determine the pipe friction coefficient necessary for obtaining the pressure loss, the Swamee Jain equation is used. The inside and outside wall temperature difference is within an acceptable range. The hot gas-side throat wall temperature has exceeded the maximum allowable temperature, i.e., 80 K, for the use of copper. As a countermeasure, a thermal barrier coating could be implemented. When assuming the use of the 10-μm-thick thermal barrier coating of which the thermal conductivity is 3 W/mK, this would result in a coating surface temperature of 1236 K with a temperature of 836 K at the interface of the thermal barrier coating and copper wall and 686 K at the coolant-side wall of copper. By adding a thermal barrier coating, the operational parameters would fit within the acceptable temperature range for copper. Table 4. Pressure and temperature calculation results. Regenerative cooling channel inlet pressure 10. MPa Regenerative cooling channel outlet pressure Regenerative cooling channel inlet temperature Regenerative cooling channel outlet temperature Combustion gas side throat wall temperature Coolant side throat wall temperature Inside and outside wall temperature difference 8.4 MPa 300 K 31.3 K 93 K 4 K 181 K Number of reusability NASA-TM X-36 4) was referred in order to estimate the number of reusability. With the combustion gas-side wall temperature of 800 K, the difference between the coolant and hot side wall temperatures of 200 K, the engine failure would occur after 1000 times the thermal cyclic fatigue. If a safety Po_1_19

4 Trans. JSASS Aerospace Tech. Japan Vol. 14, No. ists30 (2016) factor of four is applied, the engine could be reused up to 20 times. As for the engine designed in this study, the hot-gas side temperature was 836 K and the difference between the coolant and hot side wall temperatures was 10 K. Both are close to the examples of NASA-TM X-36 and we conclude that the design satisfies the 100-times reusability with a safety factor of four GG cycle design analysis Overview The GG cycle is assumed as shown in Fig.. Although one-turbine configuration is also a possible candidate ), two independent turbine systems were used for LOX and ethanol in order to have design flexibility. Table shows the initial tank conditions. Tank pressure is set to 0.3 MPa considering NPSH 3, 6, ). The pressure loss coefficients in each propellant feed line are displayed in Table 6. The definition of the pressure loss coefficient k is shown in Eq. (3). Fig.. Gas generator cycle conceptual diagram. Table. Tank conditions. LOX tank pressure Ptank_ox 0.3 MPa Fuel tank pressure Ptank_fu 0.3 MPa (3) GG combustion pressure Table shows the given specifications of GG. The GG combustion chamber pressure is set to the same value as the main combustion chamber pressure. The GG combustion chamber is uncooled, so the chamber wall temperature must be less than 80 K considering a sufficient number of reusability. Fuel-rich side O/F was 0.0 to satisfy this combustion chamber temperature. Table. Parameters of the gas generator. Combustion chamber pressure Combustion chamber temperature MPa 833 K O/F 0.0 Contraction ratio Bleed ratio Bleed ratio is defined as the mass flow rate of the GG bleed gas to the total mass flow rate. In our analysis, the minimum bleed ratio to satisfy the power balance between turbines and pumps is surveyed by a simple iteration process Turbine and pump efficiencies The required bleed ratio was calculated varying turbine and pump efficiency from 0.2 to 0.. The same efficiency was assumed to LOX and fuel turbo pump. In the actual LOX/RP-1 engines, a pump efficiency is generally higher than 0., while a turbine efficiency is generally higher than Calculation results The resulting bleed ratio and system I sp are summarized in Tables 8 and 9, where the I sp efficiency is 1.0. One can easily estimate the required I sp efficiency by this table. For example, when the pump and turbine efficiency is 0.6, I sp efficiency must be above 0.94 (31/330.3). LOX tank temperature Ttank_ox 90 K Fuel tank temperature Ttank_fu 300 K BL[%] Table 8. Bleed ratio calculation results. Pump efficiency Table 6. Pressure loss conditions. Line Symbol k Baseline pressure LOX pump outlet Main combustion chamber k_ox 0.2 Pc Fuel pump outlet Main combustion chamber k_fu 0. Pc LOX pump outlet GG combustion chamber k_gg_ox 0.2 Pgg Fuel pump outlet GG combustion chamber k_gg_fu 0. Pgg GG combustion chamber Fuel turbine inlet k_α 0.1 Pgg Fuel turbine outlet LOX turbine inlet k_β 0.1 Pturb_fu_e LOX turbine outlet Outside air k_γ 2.0 P_out Turbine efficiency Turbine efficiency Table 9. Isp efficiency calculation results. Pump efficiency Isp[s] * Chamber Isp = s Po_1_20

5 A. IIJIMA et al.: Rocket Engine Feasibility Study for the JAXA Future Transportation Reference System Detail analysis In the previous section, the GG cycle, which enables a straightforward development process, was found to be feasible for the required mission specification with a combustion pressure of MPa, a turbine pump efficiency of 0.6, and an I sp efficiency of For further validation, an analysis is conducted using Rocket Propulsion Analysis ver2.2 (RPA) a commercial software. The input parameters are listed in Table 10 and the results are summarized in Table 11. In the RPA analysis, the bleed ratio was estimated to be much larger. This is because this software assumes the considerable inducer pressure loss and the pump exit pressure. As a result, the vacuum system I sp (312.6 s) does not satisfy the specification requirements (31 s). The estimated I sp efficiency was 0.9, which nearly corresponds to the desired value in the previous section. Table 10. Combustion pressure Input parameters for RPA analysis. MPa Expansion ratio 2 Contraction ratio 3 Throat diameter Nozzle shape 0.32 m 1 Conical Table 11. Comparison of calculations and RPA 2.2 analysis. Analytical (Sec. 3.3.) RPA Vacuum chamber I sp s s Bleed ratio Isp efficiency 0.94 (desired) 0.9 (calculated) Vacuum system I sp 31.1 s s Sea level chamber I sp s 21.6 s Sea level system I sp 29. s s Vacuum system thrust 1028 kn 1041 kn Sea level system thrust 96 kn 82 kn Vacuum thrust coefficient Sea level thrust coefficient Fuel turbine power 2664 kw 2893 kw Oxidizer turbine power 2486 kw 29 kw Propellant total flow 32.6 kg/s 32.1 kg/s GG drive gas flow. kg/s 14.4 kg/s Engine weight kg Thrust/weight ratio - 92 Oxidizer Fuel Oxidizer tank temperature Fuel tank temperature Oxidizer tank pressure LOX ethanol 90 K 300 K 0.3 MPa Sea level thrust obtained from RPA was much smaller than that of our original analysis. This was due to the difference of the assumption on the flow separation limit. At any rate, both results satisfied the specification requirement shown in Table 1 (> 06 kn). Fuel tank pressure 0.3 MPa Inlet velocity 4 m/s Cycle GG GG combustion pressure MPa GG combustion temperature 833 K Pump arrangement F-O Series Pump efficiency 0.6 Turbine efficiency 0.6 Fuel turbine expansion ratio 4.20 Oxidizer turbine expansion ratio 4.44 Fuel turbine rotation speed rpm Oxidizer turbine speed rpm Injector pressure loss 0. MPa Valve pressure loss 0.3 MPa Regenerative cooling pressure loss 2.1 MPa Regenerative cooling temperature rise 90 K Two-layer flow consideration On Ionization consideration On Flow separation On Throttling 0% 100% Table 12. Case A and B calculation results. Nominal Case A Case B Chamber ratio thrust, s Bleed ratio Isp efficiency Vacuum system specific impulse, s Sea level on the chamber ratio thrust, s Sea level on the system ratio thrust, s Vacuum system thrust, kn Sea level on the system thrust, kn Vacuum thrust coefficient Sea level on the thrust coefficient Fuel turbine work, kw Oxidizer turbine work, kw Propellant total flow, kg/s GG flow, kg/s Engine weight, kg Thrust-weight ratio We performed a wide-range parametric survey in the previous section and had more precise results by RPA. Since Po_1_21

6 Trans. JSASS Aerospace Tech. Japan Vol. 14, No. ists30 (2016) 4.2. Experimental conditions An impact of sulfur corrosion or coking is affected by the following parameters8): Inner wall surface temperature of the cooling channel Shear force by BE BE sulfur concentration (for sulfur corrosion only) In this study, we focused on the temperature effect and the experimental conditions shown in Table 13 were used at the BE flow test. The experimental pressure is MPa, which is above the critical pressure of ethanol (6.14 MPa). The total flow time is s, corresponding to the required burn time at one launch. Due to a limitation of the tank capacity, s was achieved by multiple tests. the vacuum system Isp obtained by RPA has not satisfied the mission requirements, some of the input parameters should be adjusted. The nominal case and adjusted cases (Case A and B) are shown in Table 12. The nominal case parameters are the same as those in Table 11. As for Case A, the nozzle shape is changed to a bell nozzle from the original conical nozzle assumption. In addition, the pump efficiency was increased to 0. in Case B. The vacuum system Isp of Case B satisfied the mission requirements. As for the engine weights, the RPA empirical database resulted in a 8 kg/engine, which satisfies the mission requirement. Additional performance improvements would also be possible by increasing the initial tank pressure (ex. 0.4 MPa) or the combustion chamber pressure (ex MPa). However the engine weight is increased in both cases and serious trade-off study is necessary. Table 13. Experimental conditions of BE flow test. No. 4. Sulfur Corrosion and Coking within the Regenerative Cooling Channels Flow Pressure time Mass Specimen Temperature flow inlet of the rate* temperature* specimen heating 4.1. Experimental apparatus The corrosion of cooling channels by sulfur is a serious concern when considering the use of BE. Furthermore, the deposition of carbon components due to the thermal decomposition is to be studied. In this study, BE was flown in the copper-alloy tube under high temperature and pressure conditions, simulating the regenerative cooling channels of rocket engines in order to understand the impact of sulfur corrosion and coking. Figure 6 illustrates the experimental apparatus. The BE fuel is flown through the tube specimens, which simulate the regenerative cooling channels. The impact of sulfur corrosion and coking is assessed by cutting a specimen and analyzing its surface after the flow test, as shown in Fig.. The specimen is made of copper-alloy (SMC), is 10 mm in length and has an inner diameter of 2 mm. section* s MPa g/s K K *target value 4.3. Evaluation method After the flow test, specimens were cut and the inside surface was analyzed by an electron probe micro-analyzer (EPMA) in order to seek the existence of sulfur or carbon components. If peeling occurred due to a severe corrosion, some flakes would be caught in the filter attached downstream of the specimen Experimental result Figure 8 shows the EPMA results for Test 1. The existence of sulfur and carbon components was recognized. Similar results were obtained in the other cases (Test 2, 3, and 4). Fig. 6. Experimental apparatus of BE flow test. Fig. 8. EPMA results (test No.1). Trace of sulfur and carbon components are observed. Fig.. Cross sectional view of the specimen made of SMC. 6 Po_1_22

7 A. IIJIMA et al.: Rocket Engine Feasibility Study for the JAXA Future Transportation Reference System Quantitative comparison was difficult because just a trace was observed in each case. The downstream filters did not catch any flakes through the experiment. Summarizing these experimental facts, there would be no problem for one-time use, but a cumulative effect should be studied hereafter.. Conclusions Aerospace Laboratory Report, NAL TR-16, ) Azuma, N., Sato, M., Tadano, M., Masuoka, T., Moriya, S., Aoki, K., Kawashima, H., Okita, K., Tamura, T. and Niu, K.: Methane hot gas flow testing for evaluating the effect of sulfur corrosion and coking for LOX/methane engine (in Japanese), Preprints of the 2 th Meeting on Aeronautical and Space Science, Performance analysis Feasibility on an ethanol-fueled booster engine for a reusable small satellite launcher is described. As a result the use of the GG cycle with a LOX/ethanol propellant is feasible. This result could be used as a general design guideline of ethanol-fueled engine. For an engine with a combustion pressure of MPa, an expansion ratio of 2, a turbine efficiency of 0.6, and a pump efficiency of 0., the required specifications were satisfied with an I sp efficiency of Assuming a use of 10-μm-thick thermal barrier coating, this would result in a coating surface temperature of 1236 K with a temperature of 836 K at the interface of the thermal barrier coating and copper wall and 686 K at the coolant-side wall of copper. A 1000-fold reusability, before fatigue fracture at the channel walls was determined. Although, if a safety factor of four is considered, the reused is 20-fold..2. Sulfur corrosion and coking in generation cooling channels With respect to sulfur corrosion and coking, a trace of sulfur and the existence of carbon components were observed on the copper-alloy specimen after the -s ethanol flow test at high pressure ( MPa) and high temperature conditions. Surface peeling was not observed. More detailed study is needed to evaluate the cumulative effects of a sulfur attack and coking in the regenerative cooling channel. References 1) Ishimoto, S. and Okita, K.: Design Study and Technology Development for Future Reusable Space Vehicles, Proceedings of the 30 th International Symposium on Space Technology and Science, ) Liquid Rocket Engine Nozzles, NASA SP-8120, ) Design of Liquid Propellant Rocket Engines, NASA SP-12, ) Experimental Fatigue Life Investigation of Cylindrical Thrust Chambers, NASA-TM X-36, 19. ) Kawatsu, K., Negushi, H. and Yamanishi, N.: Conceptual study of LOX/ethanol regeneratively cooled rocket engine (in Japanese), JSASS , Preprints of the 4 th Meeting on Aeronautical and Space Science, ) Yamada, J., Kamijo, K. and Watanabe, M.: Research of Ekisan-liquid water rocket engine turbo pump system (in Japanese), National Aerospace Laboratory report, NAL TR-696, ) Kamijo, K., Shimura, T. and Hashimoto, R.: The suction performance of the rocket for the liquid oxygen and liquid hydrogen pump inducer (in Japanese), National Po_1_23