Final Report Summary - PRECISE (Chemical-µPropulsion for an Efficient and Accurate Control of Satellites for Space Exploration)
Executive Summary:
PRECISE (chemical µPRopulsion for an Efficient and accurate Control of Satellites for Space Exploration) focuses on the development, manufacturing and testing of a MEMS-based (Micro Electro Mechanical Systems) monopropellant µCPS (micro Chemical Propulsion System) for highly accurate attitude control of satellites. The availability of such propulsion systems forms the basis for defining new mission concepts such as formation flying and rendezvous manoeuvres. These concepts require propulsion systems for precise attitude and orbit control manoeuvrability.
PRECISE combines European capabilities and know-how from universities, research organisations and experienced space companies for the research and development of a μCPS for future market demands. μCPS has been identified by the European Space Agency ESA to fill the gap between state-of-the-art electrical and chemical propulsion due to its compactness, low power requirements and low system weight. Due to these reasons the MEMS-based µCPS is considered as one of the key technologies for future satellite missions.
All partners in the consortium possess a sound experience in the topics they are called for and they can all look back on various successful projects in their company’s history. Two universities are involved, the University of Twente and the Surrey Space Center of the University of Surrey. In addition, two large research organisations are participating, the National Center for Scientific Research (CNRS) and the German Aerospace Center (DLR) who is coordinating the project. Finally, three industrial partners namely EADS Astrium Space Transportation, NanoSpace and NPO Mashinostroyenia are completing the consortium.
The term chemical micro propulsion is used for propulsion systems with thrust levels in the order of microNewtons (μN) up to several milliNewtons (mN) and generating thrust primarily by means of chemical energy of the propellant itself. A revolutionary feature is the highly compact, lightweight and modular architecture. The primary objective is the development of a µCPS prototype with all its single components, to be fired with hydrazine under space vacuum conditions.
Several aspects need to be considered for a consistent development of the µCPS:
- Definition of requirements and specifications, comprising S/C demands
- Research of critical propulsion aspects
- Development of crucial components
- Elaboration of the test facility infrastructure
- Development of diagnostic tools
- Further development of numerical flow simulation tools and comparison
- Manufacturing, assembly, integration and testing of the µCPS
Project Context and Objectives:
The project PRECISE consists of six major work packages. The project management is realised in WP1, the technical coordination in WP2 and the dissemination activities of the project in WP3. The main technical and research work will be conducted in WP4 and WP5. Some components developed in WP4 will be tested in WP6.
The general task of WP4 µCPS Technology Research is to coordinate the technical disciplines necessary to design and manufacture the MEMS components for a µCPS propulsion module and the required test infrastructure. The key activities of the present work package can be regarded as the preliminary research phase of PRECISE.
WP 4.1: The objective of "µCPS Requirements and Specifications" is the elaboration and harmonisation of the µCPS requirements for the two spacecraft Solar Sail and InspectorSat which finally shall result in a common requirement specification for the propulsion module. The high-level subsystem performance value estimations based on high-level mission requirements is determined. In addition, the the approximate μCPS design values with given mission requirements, such as the total S/C mass, the pointing knowledge and control, the orbital knowledge and control is determined. Basic requirements for the μCPS System are defined based on thrust level, total impulse, mass, lifetime etc., operation mode (SSF, PMF, duration) and power consumption. The requirements of solar sail orientation and attitude control will directly be merged and considered in the development of the μCPS to account for real application demands.
WP 4.2: The objective of "Propulsion Research and Development" is to support the thruster design, to elaborate the µCatalyst layout and select efficient functional layers for deposition on the chip to optimise the decomposition of hydrazine. In this context, the following topics are addressed for the analysis of micro fluidics:
- State of the art of super-hydrophobic (to reduce pressure drop) and super-hydrophilic (to increase catalytic decomposition) surfaces made of silicon or silicon carbide;
- State of the art of these surfaces in the field of microfluidic and possible ways to transform the surfaces;
- Design of the micro thruster and micro chemical propulsion systems.
- Establishment of mathematical and numerical models and performance prediction calculations.
For the development of the crucial micro catalyst the following tasks are performed:
- State of the art of silicon- and silicon carbide- supported catalysts.
- Design of micro catalysts required for decomposition processes inside micro volumes.
- Manufacture and characterization of micro-catalysts; different catalytic surface coating techniques will be assessed for surface made of silicon or silicon carbide. It is intended to investigate silicon carbide porous coatings formed by chemical interaction via solutions and CVD.
WP 4.3: The objective of "Components Development" is to develop crucial components of the thruster, such as the µValve, the µHeater and a µBubble actuator.
Within the proposed concept of the MEMS-based micro thruster the technology of micro valves becomes one of the key components providing reliable operation in terms of actuation possibility, propellant compatibility, sealing, required fast response characteristics, constant injection quantities, and a high number of activations. The objective of the µValve development is the design, manufacturing and evaluation of a propellant micro valve according to the set requirements like operational conditions, dimensions, electrical and mechanical interfaces within the μCPS under development. Initially, it will be proposed to modify the available micro valve concept. A new micro valve concept should be elaborated as the next development step in case no required improvement of the available hardware could be achieved with adequate modification measures. Within the first step available concepts and candidate components for micro valves should be assessed based on the requirements and the valve definition sheet which will be specified. Subsequently the required modifications will be elaborated for available prototypes and the manufacturing flow chart will be worked out. Finally, the µValve P/T will be manufactured and tested.
The implementation of a μHeater to heat the μChamber is required to improve the performance during the start phases of a micro thruster at low operational temperatures. The primary goal within this task is the adaptation of an already available μHeater concept which is based on the operational requirements and which considers mechanical and electrical interfaces of the micro thruster. A definition sheet will be specified for the realized design. Based on this sheet and according to space standards the μHeater will be manufactured and characterized within the test program. An alternative solution to be investigated is the integration of an external heater on the μChamber. This will reduce the complexity of the interfaces and hence might reduce the technical risks involved.
In addition, the use of μBubble Actuators is investigated for realization of a reliable, fast responding propellant injection principle for the delivery of ultra-small propellant amounts as required for a MEMS based μCPS. A feasibility study will be performed and a potentially suitable configuration will be considered. On this basis, a definition sheet will be specified and a first test device will be designed.
WP 4.4: The objective of "Test Facility Diagnostics" is to develop test facility diagnostic tools such that the thruster module and its components can be tested and the performance evaluated. A mass flow sensor is needed for reliable measurement of propellant consumption under real time conditions. For this purpose a Micro Coriolis Flow sensor will be designed and fabricated. An important advantage of flow sensors based on the Coriolis principle is that the true mass flow is measured, independent of the composition and properties of the fluid, and independent of pressure, flow profile and temperature of the fluid. A Coriolis Flow Sensor contains a vibrating tube segment in which a mass flow is subject to Coriolis forces. These Coriolis forces are directly proportional to the mass flow. The main challenge for a micro Coriolis flow sensor is to detect the extremely small Coriolis movements. Two cleanroom fabrication runs are planned in order to allow for a design and process flow optimization before fabrication of the final devices. The existing fabrication technology at UT based on silicon nitride channels embedded in the silicon substrate will be used as a basis. This should result in a sensor that is suitable for later integration into the micro propulsion system.
Besides the mass flow sensor work in, research will be performed on calorimetric flow sensors. The operation of a calorimetric sensor is based on transferring some heat to the fluid which is carried away by convection. The amount of heat that is carried away is a measure for the mass flow. Calorimetric flow sensors can be much more sensitive than Coriolis type flow sensors with a resolution down to the picoliter/minute level, which is the reason to realize these sensors in parallel. Of course, since calorimetric sensors rely on heat transport, the sensitivity is significantly influenced by fluid properties like specific heat and thermal conductance and the sensors require extensive calibration.
For plume analysis it is not possible to use a standard type flow sensor since the flow cannot be confined in a tube. Therefore, a Pitot tube sensor will be designed and realized for this purpose. The same basic fabrication technology will be used, although an adaptation will be necessary to allow for an open tube end. Various readout mechanisms (e.g. capacitive, thermal, piezoresitive) will be considered to detect the small pressure variations.
WP 4.5: The objectives of "Test Equipment, Facility and Modelling" are the definition of test facility requirements and the elaboration of the thrust measurement balance to be able to test the microthruster developed during this project and characterize its functional range. The thrust measurement balance will be used in DLR's vacuum chamber test facility. A trade off study will be performed, comparing different concepts for the thrust measurement rig. Advantages and disadvantages will be compared to find the most suitable approach to match the properties of the thruster and fulfil all given requirements. Based on this analysis a design will be selected, built and used during the final hot firing tests.
In addition, a CFD roadmap for the numerical flow simulations of µCPS will be derived which defines the path for the required extensions, modules and chemical models for the simulation of micro fluidic behaviour in µThrusters and µCatalyst chambers. The requirements, boundary conditions, initial conditions and dimensions for the CFD computations are set. With a comprehensive literature review and with a close cooperation and knowledge exchange with other work packages, the occurring flow phenomena in micro thrusters will be identified and analysed. With this knowledge, applicable numerical models will be reviewed and evaluated and a future model implementation strategy will be derived for a prospective implementation of these suitable models in the TAU code.
Since the TAU code is based on the compressible Navier-Stokes equations and can be used with structured as well as unstructured grids, the methods to be considered will be efficiently useable with all these functionalities. Focus will also be on the methods suitability with regard to the targeted coupling of the TAU code with the DSMC method to be conducted in WP5.2. The DSMC method is needed for the numerical plume expansion computations as the Navier-Stokes equations are not valid in domains of high rarefaction.
The key activities of WP5 - µCPS MAIT and CFD can be regarded as the realization phase of the μCPS within PRECISE dealing with:
- WP5.1 manufacturing, assembly and test of the μCPS prototype
- WP5.2 numerical simulation of the μCPS and enhancement of the numerical models
- WP5.3 quality assurance aspects
- WP5.4 final run-up of the test facility and hot firing tests within the vacuum chamber
-
Thereby the technological results elaborated within WP4 will be consequently implemented under compliance with the specified technical requirements. Additionally, activities dedicated to manufacturing, integration as well as verification of the μCPS prototype will be ensured applying elaborated QA processes and standards.
The objective of WP5.1 is to establish the manufacturing process for the final design of the μCPS. Critical integration processes will be investigated through feasibility tests. Subsequently, subsystems will be manufactured considering the technologies and hardware worked out within the previous research and development tasks, also with focus on the identified manufacturing methodology. Finally the subsystems and the delivered prototype of the micro thruster module will be assembled to the μCPS prototype.
In WP5.2 the numerical modelling and simulations will be performed. Based on the simulations the nozzle can be optimised and designed such that the performance of the µCPS is improved. These optimisation will be directly fed into the µCPS manufacturing.
The a priori CFD strategy has been outlined in WP4.5.2 CFD roadmap. The initial idea has been to extend CFD capabilities in the DLR TAU code in parallel to the development of the µThruster, so that, in the end of the project, experimental results can be used to validate CFD models. Objective of the chemical modelling within the PRECISE project consisted of enhancing the DLR TAU code by adding the capability of conducting finite rate chemistry of hydrazine decomposition. TAU enhancement encompasses all development activities of the DLR TAU code beyond chemical modelling of hydrazine which are necessary to accurately treat the investigated flows.
Quality assurance are also addressed within PRECISE. At present no certified methodology for the MEMS based propulsion system exists. However for further development and establishment of the investigated technology up to the highest technology readiness level reliable quality monitoring tools are obviously required. The first aim of the activities within this work package is the continuation of already running elaboration processes for the required QA methodology. On this basis the development, the manufacturing and the test activities within the project should be monitored. According to space standards the core competences and activities like the establishment of a QA plan, review test planning, manufacturing and operating procedures, non conformance and change treatment, participation on test readiness reviews and review of test results and synthesis will be carried out.
Finally, the hot firings of the developed µCPS will be performed. For verification of the µCPS within the demonstration test the available test facility (STG-MT) will be used.Within the scope of this work package the test infrastructure will be serviced and operated regarding the performance of all relevant development tests of the micro components, the micro thruster and the hydrazine demonstration firings with the μCPS prototype. For various firing modes the thrust will be measured with the thrust measurement balance to be developed and provided by WP4.5. After these first thrust measurements the comparison between expected (‘designed’) and measured data will allow an assessment on the development status of the thruster.
Based on the provided test requirements in the test plan, the test procedures will be elaborated considering such aspects like handling of facility and specimen, operational and firing sequences, measurement lists. In accordance to the test plan, the setup of the test infrastructure will be set and verified with performing e.g. calibrations of the diagnostic tools. Finally the status will be determined and used for the test release which is usually decided during the test readiness review. Based on the achieved release the test campaign will be performed.
Project Results:
The main technical and research work was conducted in WP4 and WP5. The therein developed S&T results will be explained in the following. The figures to which the text refers are included in the attached PDF-document PRECISE-figures_final-report.pdf.
WP4 - µCPS Technology Research (AST)
The general task WP 4 coordinated the technical disciplines necessary to design and manufacture the MEMS components for a µCPS propulsion module and the required test infrastructure.
WP 4.1 µCPS Requirements and Specifications (AST)
The objective of "µCPS Requirements and Specifications" was the elaboration and harmonisation of the µCPS requirements for the two spacecrafts Solar Sail and InspectorSat which finally led to a common requirement specification.
Motivation to use a MEMS based µPropulsion System
Satellite propulsion is required in order to reach a satellites final orbit, to maintain this orbit and to perform orbit changes according to a specified mission.
The required propellant mass mP for a required maneuver (or a sum of maneuvers) is a function of the mission required velocity increment delta-v, the satellite mass mS and the performance (ISP) of the propulsion system:
ISP is a key performance factor that significantly influences the required propellant mass for a given manoeuvre. Changing from state-of-the-art cold gas systems to a hot as system increases the ISP by a factor of 2-3 allowing a significant propellant mass reduction and even manoeuvres that cannot be realized with cold gas propulsion.
Commercial Off the shelf (COTS) hot gas reaction systems are available only in bigger sizes and higher thrust levels that do not allow the implementation into a nanosatellite and to perform the required manoeuvres with the thrust and IBIT level needed.
Mission concept
The PRECISE project aims to provide a hot gas propulsion system for two satellites in the Nanosatellite class with a typical weight between 1 and 10kg (see Figure 1).
Type Mini Micro Nano Pico Femto
Weight 100 – 500 kg 10 – 100 kg 1 – 10 kg 0.1 – 1 kg < 0.1 kg
Type Myriade class SSTL X50 class 3U cubesat 1U cubesat / µcube experimental
Propulsion (typical) Classical propulsion system Cold / warm gas Cold / warm gas Cold gas Without propulsion
Figure 1: Classification of small Satellites
- The first InspectorSat mission is a concept designed by the Surrey Space Centre (SSC) to demonstrate the flight and utility of a Visual Inspection Payload (VIP).
- The second satellite designed by NPO Mashinostroyenia will provide a solar sail spacecraft (SSSC) as a co-orbital observation target for the VIP.
The mission needs for the 2 spacecraft were analysed in detail [D4.406] and the requirements for the following main PRECISE µCPS were defined [D4.414]:
- Thrust level: 10mN
- Total propellant throughput: 100g (qual factor 1,5 included)
Generic µPropulsion System Layout
Figure 2 shows a schematic view of a chemical propulsion flight system and the hardware which is to be developed within the actual PRECISE project:
Figure 2: Propulsion System components and Hardware to be developed within the PRECISE project
PRECISE Thruster Chip in stacked Design
In order to develop and test each component separately a stacked design was chosen where each component has a functional layer.
Figure 3 shows the different layers which are integrated on the µCPS test unit, the hot firing test rig and the test cell:
Figure 3: Stacked layer design of functional components
WP 4.2 Propulsion Research (LCO)
The objective of "Propulsion Research" was to support the thruster design. It was divided into two sub work-packages Micro Fluidics and Micro Catalyst.
µFluidics
Figure 4 shows the various flow domains in the thruster as described in [D4.404].
Liquid hydrazine enters a vaporization chamber (1), where it is preheated and transitions into the gaseous phase. An injector (2) channels the gaseous hydrazine onto the catalyst bed (3), where it is catalytically decomposed. The hot mixture of reactants then expands from the stagnation chamber (plenum) (4) through the nozzle throat (5) and the nozzle (6), before it leaves the thruster duct at the nozzle exit (7) and expands into vacuum (8).
Figure 4: Sketch of thruster flow domains
In [D4.411] the liquid flow parameters in the µcatalyst were calculated and different catalyst bed configurations assessed. Using iridium as an active catalyst phase material the required Average Expected Geometric Surface Area (AEGSA) can be calculated based on the hydrazine flow rate and the available catalyst volume.
Figure 5: µchannel and µpillar catalyst configuration
From the results obtained taking the flow of liquid hydrazine into account, it is expected that the μChannels + Alumina configuration reaches the proposed AEGS if the μCat bed can be marginally lengthened (and/or if the alumina layer thickness is increased) see Figure 5. The catalyst inlays were manufactured accordingly.
µCatalyst
With these inputs two different supports have been fabricated, a flat silicon support with and without a thermal oxide SiO2 and metallic alloy supports with micro-channels patterning. An alumina washcoat has been developed on some of these supports by a sol-gel process prior to iridium active phase deposition. The iridium active phase has been then deposited on these supports, as well as on supports which were not washcoated with alumina, either by wet impregnation (CD) magnetron sputtering (PD).
The obtained catalysts present a good iridium adhesion onto the different supports (mechanical stability). The prototypes have been characterized by AFM and SEM to determine the iridium nanoclusters dimensions and dispersion. Finally, droplets of pure hydrazine have been injected on the different prototypes at room temperature and pressure to highlight the catalytic activity.
The catalytic activity was tested at room temperature and at an elevated temperature of 50°C with the following results:
- No noticeable hydrazine decomposition process for the different µcatalyst prototypes at room temperature
- No noticeable hydrazine decomposition process for the different NicroFer prototypes at a temperature of 50 °C (setpoint)
- Low hydrazine decomposition process for the different Silicon patterned prototypes at a temperature of 50 °C (setpoint)
- Highest hydrazine decomposition process for the 20 nm iridium layer Silicon patterned prototypes at a temperature of 50 °C (setpoint) . This was confirmed with better decomposition tested at a temperature of 80 °C.
- No noticeable difference observed for the Silicon patterned prototypes with different straight channel dimensions (at fixed iridium thickness).
WP 4.3 Propulsion Component Development (NSP)
The objective of "Propulsion Component Development" was to develop components that shall be used for diagnostics in the thruster module or at the test bench.
µValve
After the assessment of different micro valve concepts a Phase Change Material (PCM) actuated µvalves with heritage from a cold gas project was selected.
The valve concept is based on a Phase Change Material (PCM) actuation principle (see Figure 6). An enclosed cavity is filled with PCM which increases in volume when it goes from solid to liquid phase. The phase change is achieved by resistive heating and the melting point can be chosen to fit the thermal requirements of the system. The flow is modulated proportionally by adjusting the applied power to heaters in a PCM-cavity, which actuates on the valve seat due to the expansion.
Figure 6: µValve operational principle
The valves are fabricated using four fusion bonded silicon chips and each valve stack has four integrated PCM actuated normally-closed valves. Valve interfaces are designed for bonding towards the nozzle package and electrically to the pod assembly connections.
The valve seat package is two fusion bonded square chips with the purpose to interface the nozzle package and also act as a normally-closed valve seat. The valve seat chips also include via holes and channels to lead gas from the valves to the nozzles. The inlet and outlets on the valve seat package is shown in Figure 7.
µHeater
The µcatalyst test results showed that the catalyst has to be preheated in order to have acceptable decomposition rates at the given surface area. The following figure shows a sample heater that was manufactured by NSP.
Figure 7: W-coated graphite µheater
µBubble Actuator
The use of μBubble Actuators has been investigated for realization of a reliable, fast responding propellant injection principle for the delivery of ultra-small propellant amounts as required for a MEMS based μCPS.
First, a list of requirements was made and an extensive literature search was performed to obtain a good overview of existing devices that could be suitable for propellant injection in a MEMS based μCPS. Unfortunately, the high demands of a relatively large maximum flow rate of 6 mg/s combined with pressures of up to 10 bar are extremely difficult to fulfil as explained in [D4.413]. Using state-of-the-art MEMS technology we estimated that a pressure of 0.5 bar can be reached. For larger pressures, stronger check valves and multiple pumps in cascade would be needed. Explosive bubble actuators as developed earlier can deliver 0.15 mW power at best, requiring 40 cascaded actuators. Alternatively, a feed pressure could be used; however in that case simple check valves will not suffice. Due to the feed pressure both valves would open and a constant flow through the system would result. When using a feed pressure the most straightforward system will probably be a fast, controlled open/close valve that can open during a precisely controlled time period.
WP 4.4 Test Facility Diagnostics (UT)
The objective of "Test Facility Diagnostics" was to develop the Coriolis flow sensor and the prandtl tube.
Coriolis Flow Sensor
The Coriolis type flow sensor uses Coriolis forces to measure directly a mass flow. This is realized by a vibrating tube, see following figure which shows a schematic drawing of a Coriolis sensor based on a rectangular tube shape. The tube is actuated in a torsional mode indicated by ωam, using the Lorentz force on an electrical ac-current ia flowing through a metal track on top of the tube and a static external magnetic field B imposed by two permanent magnets. A mass flow Φm inside the tube results in a Coriolis force Fc. The Coriolis force is capacitively detected by its induced out of plane vibration mode with an amplitude proportional to the mass flow (see Figure 8).
Figure 8: Coriolis Flow sensor – measurement principle and built hardware
In the frame of the PRECISE project Micro Coriolis mass flow sensors for measurement of hydrazine flow were successfully realized and three sensors were built.
Prandtl (Pitot) Tube
A miniaturised Prandtl-tube sensor incorporating a 6 mm long 40 μm diameter microchannel with integrated pressure sensors has been realised.
The sensor has been designed for the characterisation of the rarefied plume flow from the μCPS. The 4.5 × 2.5 mm sensor chip incorporates 400 mbar full-scale capacitive pressure sensors.
The capacitive pressure transducer is created by merging several microchannels into a rectangular pressure membrane, with outward-facing comb-fingers hinging on the microchannel sidewall. Additionally, thermistors can measure the temperature profile on the Prandtl tube and an integrated Pirani sensor can optionally measure vacuum pressures below 10 mbar. An electronics board and stainless steel probe, from which only the Pitot tube protrudes has been realised. Initial measurements of dynamic pressure on the Prandtl tube have been obtained by flowing air from a metal tube with 0.7 mm diameter, demonstrating the feasibility of the miniaturised Prandtl -tube sensor.
Figure 9: Miniaturized Prandtl tube sensor
Calorimetric Flow Sensor
A miniaturised calorimetric flow sensor was integrated on the same chip with the coriolis mass flow sensors. The calorimetric sensors are especially useful at the lower end of the measurement range where the Coriolis flow sensor is less accurate.
Figure 10: Miniaturized caloric flow sensor
WP 4.5 Test Facility Setup and Modelling (DLR)
The objectives of “Test Facility Setup and Modelling” was to hot fire test the µthruster and to perform numerical calculations that support the thruster design.
The thrust range of common used hydrazine thrusters on the market ends in the area around 0.1-1N. The thrust measurement principle used for those thrusters at Airbus DS is not suitable for forces in the mili-Newton range. For the thrust measurement of the µpropulsion thruster, a thrust balance for experimental testing has been developed and manufactured in the scope of the PRECISE project.
Requirements Definition
Requirements, specially tailored to the expected performance of the microthruster, have been defined and reported in D4.403. The Thrust Balance needs to be able to withstand all operating media, especially hydrazine, and it needs to be able to measure thrust in the range between 0 and 10 mN. A detailed description and explanation of all requirements is given in D4.403.
Trade-off Study
Several measurement principles have been investigated and presented in [D4.403]. As a result of this study, a concept using a thrust table, mounted on bending elements, has been selected. The whole concept has been presented in [D4.412]. The measurement principle is shown in Figure 11.
Figure 11: Thrust measurement rig functional principle
The thruster is mounted on a thrust table, including all controlling electronics, required data acquisition systems and fuel supply. This thrust table is connected to a Base Plate by four flexible bending elements that are also used for power supply. If the thruster is fired, the Thrust Table will be displaced, proportional to the thrust force. The displacement is quite small, but it could be measured using very accurate sensors.
The chosen setup gives easy accessibility to test specimen and instrumentation. It is a simple and robust concept. The displacement of the Thrust Table could be adapted to match the aimed thrust level by exchanging the bending elements. Thereby the stiffness will be adapted, and the best suited thrust to displacement ratio could be chosen. This is not only important for different thrust levels, but also for the firing mode. For steady state firings, a lower stiffness with higher displacement could be chosen, to get a better accuracy. For Pulse Mode Firings, it is better to use thicker bending elements with higher stiffness, to prohibit interference between the Pulse Mode Firing frequency and the Eigenfrequency of the system. In the required measurement range, the thermal expansion of the material has to be taken into account, even by temperature changes around only one degree Celsius. This has been taken into account by selecting appropriate material with a low thermal expansion coefficient.
Design Phase
The concept trade of phase was followed by the detailed design phase. During this phase, the thrust measurement concept was further developed leading to the final design as shown in following figure. All equipment was either designed or selected from off the shelf products. Market research has been performed to find suitable instrumentation, calculations have been made to size the components and analyse the functionality, and the material has been selected based on the given constraints and requirements. The thrust measurement rig as designed consists of the elements as shown in Figure 12.
Figure 12: Thrust measurement rig
Manufacturing, Design and Assembly
After the design was fixed, all equipment has been procured and the components were manufactured. Additionally, some improvements have been implemented due to first test results and evaluation of the available hardware.
Implementation and run up
Final calibration and run up of the thrust balance was performed at DLR. As the thrust measurement rig is equipped with a voice coil calibration system, it was first required to calibrate the voice coil itself before this system could then be used to calibrate the thrust balance. Theoretically, there is a clearly defined force-current relationship of the voice coil. However, it turned out that the metallic parts of the thrust measurement rig are affecting the electromagnetic field of the voice coil and hence the force used for calibration. Also, the electromagnetic field of the voice coil also interfered with the electric scale that was used to measure the force. Most of the time initially planned for the thrust measurement rig run up was required to calibrate the calibration mechanism itself, until a final solution has been found.
Therefore less data than planned is available on the thrust measurement rig functionality. For steady state firing, the force to elongation relationship has been calibrated, which was then later used for the testing of the micropropulsion thruster. The thrust balance and its instrumentation worked therefore quite well, however external influences as e.g. induced by the vacuum pumps reduced the resolution due to vibrations. Nevertheless, the thrust measurement balance has registered and measured thrust during the hot firing tests and therefore proven its functionality.
Summary and Outlook
A propulsion system based on liquid monopropellants represent a significant improvement compared to currently available cold / warm gas systems.
The actual PRECISE project proved that a chemical µpropulsio system based on MEMS manufactured components is feasible. All necessary components successfully proved their functionality.
Based on actual requirements the following development work has to be done in the next step:
- Use of a non-toxic propellant instead of hydrazine: due to low decomposition temperatures a monopropellant system based on H2O2 seems to be a promising candidate. Material compatibility and H2O2 self-decomposition rate is a key point for this propellant. Synergies to currently running activities (ESA funding) to be used
- Development of a propellant tank in the 100ml range and additional components like valves, filters, service valves. Choice of tank configuration (bladder tank or propellant feeding via surface tension tank)
- Adaptation and verification of analytical tools in order to precisely predict the flow and heat transfer in µchannels
- Integration of a maximum number of components in order to use the advantage of a MEMS based system
- Optimization of catalyst and heater configuration in order to allow an optimum decomposition of propellant
WP5 Introduction
The key activities of work package 5 can be regarded as the realization phase of the μCPS within PRECISE dealing with:
• manufacturing, assembly and test of the μCPS prototype
• realization and run-up of the test facility
• numerical simulation of the μCPS and enhancement of the numerical models
• validation of numerical models through comparison with experimentally obtained data
Thereby the technological results elaborated within WP4 were consequently implemented under compliance with the specified technical requirements. Additionally, activities dedicated to manufacturing, integration as well as verification of the μCPS prototype was ensured by applying elaborated QA processes and standards. For verification of the micro chemical propulsion system within the demonstration test the available test facility (STG-MT) was used after implementation of the equipment provided within WP4.4 and WP4.5. In addition to numerical performance calculations, the flow solver TAU and the DSMC method were enhanced through the implementation of physical models and methods elaborated within Task 4.3.2 and Task 4.5.2.
NSP has been the work package leader and responsible for the coordination of the activities within WP5. Within WP5.1 NSP designed, manufactured, and delivered several MEMS-based microthruster parts including the prototype of the μCPS and monitored the activities related to QA. AST provided the required system engineering containing interfaces of micro thruster, specification of test requirements, supervision of the test setup and test performance as well as analysis of the test results. AST also provided one part of the micro thruster prototype for the integration within the μCPS prototype, and the major part of the integration unit. DLR was responsible for the run-up and operation of the STG-MT test facility in WP5.4 and performed preliminary analysis of the test results. The numerical simulations and the enhancement of the flow solvers were also covered by DLR in WP5.2.
WP5.1
The complete micro chemical propulsion system (µCPS) can be presented in different ways. One way, which also shows the modular thinking with exchangeable layers within the integration unit is shown in Figure 13.
Figure 13 Schematic of the µCPS presented to illustrate the modular approach with exchangeable layers in the integration unit stack.
However, WP5 mainly concerns the integration unit i.e. the “rocket engine” of the system.
After the conceptual modular approach was established the work continued with requirement and design specification. A definition of the MEMS micro Chemical Propulsion System (μCPS) hardware and also the fabrication sequence in terms of a Manufacturing Flow Chart was agreed. The related subsystems, which together with the μCPS comprise the PRECISE demonstrator system was also defined in a complete product tree. Part of that tree showing the parts of the integration unit are illustrated in Figure 14.
Figure 14 Product tree of the manufactured, integrated, and tested µCPS.
An exploded view of the integration unit design is shown in Figure 15 and some of the corresponding parts during assembly in Figure 16.
Figure 15 Exploded view of the integration unit design.
Figure 16. Integration unit parts during assembly prior to cold testing followed by shipping to DLR for hot firing tests.
During the course of this development project a huge number of different designs, and different material combinations, were produced and evaluated for every component in Level 4 in the product tree.
As an example µInserts were made in both metal and mono-crystalline silicon bulk material. Only the silicon version was made in twelve designs with two different etch depths, deposited with a matrix of several catalyst layers in different thicknesses using deposition techniques. IN addition to this the µInserts were manufactured in four different sizes. Three versions of the µInserts together with matching µChamber chips are illustrated in Figure 17.
Figure 17. a) µChamer chips, µInserts, and µDiagnostic chips spread out in three pairs
b) Close-up of µInserts with 25µm wide walls and 100 µm wide and 200 µm deep trenches.
A good example of the interdisciplinary exchange within the project is the design of the micro nozzle. The foreground from WP5.2 was implemented in an updated design of the micro nozzle hardware in WP5.1. See Figure 18.
Figure 18. µChamber chip (15x26 mm) selected as baseline for integration unit#2. for the µCPS #2. Photograph (left) and CAD-drawing (right).
Prior to hot firing tests with real hazardous propellant. Leakage and flow tests were performed with gas (aka cold testing). A validating flow test result is shown in Figure 19. Figure 20 shows the integration unit before the hot firing setup including harness.
Figure 19. Flow test results from the two completely assembled integration units.
Figure 20. Integration unit including harness prior to hot firing tests.
WP5.2 Numerical Modeling
First, an assessment of CFD contributions had to be carried out based on the physical phenomena break-down identified in the CFD roadmap.
Multiphase decomposition of hydrazine is a problem with substantial practical experience in the PRECISE Team (AST, LCO). We thus decided that CFD can contribute most effectively for the outcome of PRECISE by studying nozzle flow.
WP 5.2.1: Numerical Simulation of the Micro Thruster
A systematic study of geometrical and thermochemical model properties showed that a flow field prediction is (unfortunately) only feasible using a specieswise temperature dependent modeling in a 3D field. 2D simplifications, such as planar or axisymmetric greatly overpredict performance and were thus found to be inadequate. Furthermore, micro scale planar nozzle flow deviates significantly from their well known macroscopic, axisymmetric cousins; established methods and procedures thus do not apply here: It was found that the dominant feature degrading microscale nozzle performance is boundary layer build-up, threatening to choke the flow in the divergent part of the nozzle. Figure 21 illustrates this.
Figure 21: Thrust along the axis is found to stagnate around X=0.0005m and to degrade substantially after X=0.0015m.
This result suggests to use a shortened nozzle design to enhance performance while reducing size. Figure 23 shows both nozzles in comparison. The result of this optimization is an increase of predicted thrust from 7.1mN to 9.2 mN (+29.6%) and of specific impulse from 144.8 s to 187.6 s (+29.6%).
Figure 22: Comparison of original and truncated nozzle flow field. The red plane is the Mach 1 isosurface.
It was found that, contrary to conventional nozzles, a short conical nozzle with a wide exit angle is advantageous at the scales investigated here. Based on our findings, a new nozzle design has been suggested.
This geometry was provided to Nanospace and implemented in the actually manufactured hardware.
WP 5.2.2 Chemical Modeling
The infrastructure for chemically reacting flows is available in TAU and applicable for combustion modelling or the simulation of dissociation effects as occurring in atmospheric re-entry.
However, the catalysis of hydrazine demands new models for the reaction kinetics and was obtained in close collaboration with WP4.2. A two-step Hydrazine decomposition chemical kinetics mechanism has been integrated and shown to describe decomposition of N2H4 into N2, H2, and NH3.
WP 5.2.3. Enhancement of the flow solver with applicable models
The modelling depth of nozzle flow study presented in Enschede revealed that full 3D, viscous, high fidelity fluid model computations are (unfortunately) necessary to capture the characteristics of high aspect ratio rectangular micro nozzles, such as boundary layer dominance and effect of variable ratio of specific heats on expansion.
A theoretical assessment showed the high expected dependency of nozzle flow on gas composition and heat capacities. A CEA computation showed that, in addition to N2H4, a mixture of N2, H2, and NH3 can be expected as expanding gas. Detailed representations of these species have been implemented into TAU.
As multispecies 3D computations are numerically costly, an efficient way to create grids and start solutions is sought. Thus, a cartesian grid generator has been developed. This greatly enhanced geometry parameter studies: previously, for even minor changes in geometry, a new CAD model had to be created, from which in turn a new grid has to be set up.
Additionally, this required the computation to be started from scratch.
Now, any numerical description of the planar nozzle can be directly used to produce a new grid onto which a previous solution is mapped as a start solution, see Fig. 23.
Figure 23: Interpolation of initial flow field in grid generator.
To account for liquid flow inside feed lines and the catalyst bed, a new real gas model has been implemented. This Euler-Euler compressible multiphase model allows both vaporization and expansion through a convergent-divergent nozzle to supersonic velocities within the same code. So far it is a promising development surpassing the well established “Volume of Fluid” model, which does not allow for a widely varying density. The inherent vaporization model captures the constant temperature plateau during phase change without the need to track vapor-liquid interfaces or volumes. Systematic validation for different cases were performed, an example is shown in Figure 24. The model captures the vapour pressure, difference of onset and offset enthalpy match the latent heat. This study has been carried out for various pressures. Furthermore, the model restraints are tested. Fluid property library resolution has been found to have an effect on liquid fluid pressure oscillations which can be remedied by choosing an appropriate library resolution. Library size has been found to have negligible effect on computational cost.
Figure 24: Exemplary plot of vaporization model validation. Temperature plateau during phase change is successfully captured.
To improve convergence of the low Mach number flow of a liquid, first steps towards implementing a preconditionig method have been undertaken.
A literature study yielded a model applicable to the real gas flow regarded herein. The required differentials of the equation of state have been calculated and implemented into the real gas library model.
Validation has been carried out compared to literature and numerical values. The preconditioning matrix has been implemented.
WP5.3 Quality Assurance
At present no certified methodology for the MEMS based propulsion system exists. However for further development and establishment of the investigated technology up to the highest technology readiness level reliable quality monitoring tools are obviously required.
The first aim of the activities was the continuation of already running elaboration processes for the required QA methodology.
The development strategy for MEMS fabrication at NanoSpace is adopted according to industrial MEMS experiences, i.e. to identify critical steps and try to manufacture feasible structures and components in silicon and iterate towards functional systems. The keyword is process compatibility! On this basis the development, the manufacturing and the test activities within the project was monitored. Most focus has been on the MEMS manufacturing, where most of the non-conformancies was reported and taken care of. In short the MEMS manufacturing process sequence is as follows: a) Write a Process Identification Document (PID). Make a feasible process sequence using standard processes and materials. b): Make the mask layouts in a CAD-programme (e.g. AutoCAD).c) Call for a PID Meeting. During the PID meeting the process sequence will be checked. When the process sequence is checked and approved by the Chief Engineer, a product name is defined and a batch number is allocated. The PID shall be saved as “PID N### Title” with the specific batch number followed by the title. The PID contains the intended process. A Process Flow Chart is generated and saved the same way. This finishes the meeting and the batch is released for manufacturing. c) Send the sorted mask files to the mask manufacturer. d) Start the batch.
When a batch is finished, the PID is saved as LOG
Delivery: A delivery is documented with a Delivery Note.
If a non-conformance has occured, a non-conformance report (NCR) has to be written.
The QA process can be schematically illustrated as in Figure 25.
Figure 25. Block diagram of the used QA process for MEMS fabrication.
WP5.4 Hot Firing test of the μCPS
The scope of this work package covers both the development of the test infrastructure as well as the actual hydrazine demonstration firings using the μCPS prototype.
The test set-up, comprising the subsystem μCPS (Figure 26), hydrazine tank and tubing (Figure 27), thrust balance (Figure 28), DAQ, vacuum chamber including pumps. The micro thruster was installed in a vacuum chamber on the DLR thrust balance, developed in WP4.5 in order to verify hot fire thrust.
Figure 26. μCPS
Figure 27. Tank and tubing as mounted on the thrust plate of the measurement rig
Figure 28. System under test in vacuum chamber.
The evaluation system was equipped with a large number of sensors, in order to measure primarily temperature and pressure. In addition to tough vacuum conditions, the toxic and harsh propellant (hydrazine) complicates calibration etc.
The final hot firing test was planned according to the matrix in Figure 29. The sequence is shown in Figure 30.
Figure 29. Test matrix for the hot firing.
Figure 30. Hot firing sequence.
There was a clear visible thermal reaction, but steady state was not reached and hence no proper thrust measurements possible. The shorter hot firing pulses the more reaction but also lower thrust pulses out of scope of the thrust balance range/resolution, see Figure 30.
WP6
The μComponents developed in WP4.3 have been tested in WP6. These are the μValve and the μHeater realized by NanoSpace and the Bubble actuator test structures realized by the University of Twente. Delivery and testing of the μValve and μHeater components was delayed slightly due to the move of NanoSpace to new facilities and the need for setting up the test infrastructure. After finalized hook-up and start-up of the new test facility, preparations for testing were done and then all experimental work including documentation was done in the months February and March 2013. Due to time constraints several engineers worked in parallel to successfully demonstrate the valve functionality using liquid in combination with MEMS valves and also heater experiments were carried out on test chips with custom sized cavities to simulate catalyst beds inserts. The objective of the performed development testing was to support the design feasibility and to assist in the evolution of a design prior to WP5. Such development tests are often used to validate new design concepts and the application of new concepts and techniques into a new configuration, in our case the use of µValves in combination with liquid and the evaporation of liquid using µHeaters. Successful testing of these components was an important step towards the realization of the complete µCPS in WP5.
The Bubble actuator test structures realized by the University of Twente were also delayed. The fabrication of these structures in WP4.3 turned out to be extremely difficult and most wafers were lost in the fabrication process. Some test structures did survive and these were evaluated in WP6.2. Most importantly, it could be concluded that more successful fabrication of devices should be possible with some adjustments to both the fabrication process and the design. Research towards an integrated Bubble actuator which started with PRECISE will continue at the University of Twente. A new fabrication run has started in July 2014 and new devices are expected in September.
Potential Impact:
PRECISE fosters the cross-border cooperation’s and partnerships between renowned European organisations and leading research centres across Europe and beyond. PRECISE unites experienced and advanced companies for chemical micro propulsion and merges their expertise to develop a unique, efficient and reliable µCPS. This approach improves the industrial supply chain and reduce future product development lead times. The geographical spread of beneficiaries includes France, Germany, the Netherlands, Russia, Sweden and the United Kingdom.
The MEMS-based monopropellant µCPS is an emerging technology which is currently under development and is considered amongst satellite experts as one of the key technologies for future satellite missions due to the very low thrust levels and compactness the system provides. The analysis of the current unmanned commercial and scientific spacecraft, satellites and missions of the last decades, as well as the foresight into the next decades shows a clear change of the design philosophy; an increasing trend towards compact systems is observable. Thus, the propulsion system has to become smaller, more accurate and more efficient to fulfil the requirements for exact and highly dynamic positioning of distributed small satellite systems. The system to be developed within PRECISE fulfils these requirements and is the next technology step improving available MEMS cold gas propulsion concepts.
The investigated formation flying concept of the solar sail together with the inspection satellite and the inspection unit for target observation are seen as an important stepping stone towards totally autonomous systems. SRY and NPO see the development of the small MEMS thruster as providing the last missing piece in enabling the technologies to mount this mission.
µCPS has been identified by ESA to fill the gap between state-of-the-art electrical and chemical propulsion due to its outstanding system parameters like the low power consumption and its low system weight. The technology status, mission needs and market perspectives are summarized in the ESA µCPS roadmap.
Already initiated cooperation projects between AST, SRY and NPO on this field are directly related to the targeted development of MEMS based hydrazine monopropellant system within the µCPS roadmap. Results of the mission elaboration as well as considerations on the spacecraft design will be implemented and continued within PRECISE.
German Aerospace Center - DLR
The cooperative activities within PRECISE were essential to increase the knowledge and the understanding of DLR regarding the flow inside µThrusters. The optimisation of the nozzle design could be derived from these outcomes and realised with the help of the consortium, in particular Nanospace who manufactured the thruster prototypes on short notice.
The testing infrastructure had to be setup in a novel way to be able to realise the final hot firings of the µCPS. The experience of Astrium was of great help for the testing layout. Furthermore, novel sensor devices could be tested and included in the testing chain, such as the prototypes from UT. In addition, for the analysis of the final results the experience of LCO was very important for the interpretation of the sensor data.
All these results were presented on worldwide conferences, taking place mainly in the United States of America and in Europe, national and European workshops and published in conference proceedings and Journal papers.
LACCO
a) Professor Kappenstein delivered a course for students, colleagues and PRECISE beneficiaries at Conference Hotel Drienerburght in Enschede, the Netherlands, on the 8th February 2013: " Course - Kinetics, Catalysis and Propulsion. Fundamental basis, Objectives - Current challenges". This course was divided into two sessions: Session 1 – Catalysis and Catalysts (duration: 1 h), Session 2 – Catalysis for Propulsion (duration: 1 h). The presentation materials which were used during this short course (117 pages) is too large to include it in this report but can be obtained by contacting directly Prof. Kappenstein from LCO.
b) Efforts in dissemination were also undertaken by Professor Charles Kappenstein (LCO). His dissemination activities in this period include:
- Open lecture “From the laboratory to space, an exciting adventure: kinetics, catalysis and propulsion.” with a presentation of the PRECISE Project:
Le Studium conferences, 31st March 2014, Orléans, France, (50 participants, lecture given in French)
NanoSpace
This project has been a great opportunity to increase NanoSpace know-how on chemical propulsion. NanoSpace has established a good working relation and had knowledge exchange with relevant expertise within the consortium.
The work performed is well in-line with the micropropulsion road map NanoSpace follows as seen in Figure 32.
Figure 32. Nanospace µ-propulsion roadmap
NanoSpace ambition is to continue the development of the microthruster module, the MEMS-based part, and make it a product and hopefully sell it to a propulsion system supplier as Astrium.
NanoSpace is a supplier of MEMS-based products for space. It has already flight demonstrated a highly miniaturised MEMS-based cold gas system. Currently, hot gas system (resistojets) are beginning to mature and sold as engineering models. The natural next step in the product line development is to develop a chemical microthruster, which will both increase the specific impulse and the thrust level (See figure SS in section above).
NanoSpace ambition is to continue the work on the thruster module, which are MEMS-based, eventually in collaboration with a catalyst supplier (from the consortium). This could be done in a three-to-five year perspective.
If so, NanoSpace intend to try to protect the method of fabrication of some crucial parts by applying for patent as usual. However, it might need another similar sized development project to reach TRL-5 or TRL-6 before a productificaiton can be made solely by NanoSpace. Otherwise, this needs to be done in close collaboration with a prime contractor.
Surrey Space Center
The research into the potential investigation in space of solar sail properties has directly influenced other research within SSC related to solar sailing. There is a plan for the inspection of a Cubesail spacecraft from another cubesat as part of the ESA QB50 mission which builds directly from the work in PRECISE.
The research has been disseminated internally within SSC and SSTL through seminars given on the development of sail inspection and visual inspection payloads. We also have advertised the work carried out on the testbed facility through the Astrodynamics website:
http://www.surrey.ac.uk/ssc/research/astrodynamics/research/index.htm
The research has also featured on SSC Facebook and Twitter accounts to advertise to the wider public.
University of Twente / MESA+ Institute for nanotechnology
PRECISE has allowed the successful realization of a micro Coriolis flow sensor with a 20 times larger flow range than earlier sensors and a micro Pitot tube sensor which is the first sensor in its kind. Although the research is far from completed and several issues still need to be solved, important progress has been made. Operation of the Coriolis mass flow sensor was demonstrated successfully and this will open up new application areas for this sensor, also outside the field of space systems. The results were presented at several conferences and workshops: MFHS 2014 (Freiburg, Germany, October 2014), Sensor 2013 (Nurnberg, Germany, May 2013), 8th ESA Round Table on Micro and Nano Technologies for Space Applications (Noordwijk, The Netherlands, October 2012), MFHS 2012 (Enschede, The Netherlands, October 2012). The sensor is also used in two master courses at the University of Twente: “Measurement systems for mechatronics” and “MEMS design”. A micro Pitot tube sensor did not yet exist and the sensor realized within PRECISE was presented at the IEEE MEMS conference (San Francisco, USA, January 2014), which is the flagship annual event of the MEMS community. The sensor contains a capacitive pressure sensor that can also be integrated on other microfluidic chips based on the same technology. It is already included in several new flow sensor designs and contributes to a new research theme towards microfluidic multi-parameter sensors, i.e. sensors that measure not only a single quantity like flow or pressure but also medium parameters like viscosity, density and thermal conductivity.
In addition, national activities in the field of µCPS are ongoing and strongly fostered in France, Germany, the Netherlands, Sweden and the United Kingdom:
• LCO is developing monolith-based catalysts within an ESA contract "Preparation, Characterization and Evaluation at the Lab Level of Catalysts for Low Thrust Bipropellant Thruster". Further, they are a partner in the FP7 project GRASP.
• NSP’s miniaturized cold gas propulsion system is demonstrated on PRISMA – a European mission to demonstrate autonomous formation flying and rendezvous. In addition, NSP’s parent company SSC has so far developed in total six spacecraft. Ongoing activities are Small GEO, Proba-3 and SMART OLEV
• UT is partner in the NanoNextNL programme, funded by the Dutch government, which aims at increasing and spreading knowledge in the field of Micro Systems MST and MEMS. Special attention will be given to problems that obstruct the further application of MST in industry. In this context, the Micro Coriolis flow sensor and the Pitot tube for plume measurements are first spin-out products of the MESA+ institute / TST group to be used in space systems. Thus, PRECISE successfully opened a new field of applications and opportunities for the institute
• DLR and AST are coordinating their national activities on the thruster and propulsion sector with the national MoU “Propulsion 2020”. The MoU and the therein included activities were initiated in 2006, also involving PhD students
List of Websites:
www.mcps-precise.com
Dr. Markus Gauer (markus.gauer@dlr.de)
PRECISE (chemical µPRopulsion for an Efficient and accurate Control of Satellites for Space Exploration) focuses on the development, manufacturing and testing of a MEMS-based (Micro Electro Mechanical Systems) monopropellant µCPS (micro Chemical Propulsion System) for highly accurate attitude control of satellites. The availability of such propulsion systems forms the basis for defining new mission concepts such as formation flying and rendezvous manoeuvres. These concepts require propulsion systems for precise attitude and orbit control manoeuvrability.
PRECISE combines European capabilities and know-how from universities, research organisations and experienced space companies for the research and development of a μCPS for future market demands. μCPS has been identified by the European Space Agency ESA to fill the gap between state-of-the-art electrical and chemical propulsion due to its compactness, low power requirements and low system weight. Due to these reasons the MEMS-based µCPS is considered as one of the key technologies for future satellite missions.
All partners in the consortium possess a sound experience in the topics they are called for and they can all look back on various successful projects in their company’s history. Two universities are involved, the University of Twente and the Surrey Space Center of the University of Surrey. In addition, two large research organisations are participating, the National Center for Scientific Research (CNRS) and the German Aerospace Center (DLR) who is coordinating the project. Finally, three industrial partners namely EADS Astrium Space Transportation, NanoSpace and NPO Mashinostroyenia are completing the consortium.
The term chemical micro propulsion is used for propulsion systems with thrust levels in the order of microNewtons (μN) up to several milliNewtons (mN) and generating thrust primarily by means of chemical energy of the propellant itself. A revolutionary feature is the highly compact, lightweight and modular architecture. The primary objective is the development of a µCPS prototype with all its single components, to be fired with hydrazine under space vacuum conditions.
Several aspects need to be considered for a consistent development of the µCPS:
- Definition of requirements and specifications, comprising S/C demands
- Research of critical propulsion aspects
- Development of crucial components
- Elaboration of the test facility infrastructure
- Development of diagnostic tools
- Further development of numerical flow simulation tools and comparison
- Manufacturing, assembly, integration and testing of the µCPS
Project Context and Objectives:
The project PRECISE consists of six major work packages. The project management is realised in WP1, the technical coordination in WP2 and the dissemination activities of the project in WP3. The main technical and research work will be conducted in WP4 and WP5. Some components developed in WP4 will be tested in WP6.
The general task of WP4 µCPS Technology Research is to coordinate the technical disciplines necessary to design and manufacture the MEMS components for a µCPS propulsion module and the required test infrastructure. The key activities of the present work package can be regarded as the preliminary research phase of PRECISE.
WP 4.1: The objective of "µCPS Requirements and Specifications" is the elaboration and harmonisation of the µCPS requirements for the two spacecraft Solar Sail and InspectorSat which finally shall result in a common requirement specification for the propulsion module. The high-level subsystem performance value estimations based on high-level mission requirements is determined. In addition, the the approximate μCPS design values with given mission requirements, such as the total S/C mass, the pointing knowledge and control, the orbital knowledge and control is determined. Basic requirements for the μCPS System are defined based on thrust level, total impulse, mass, lifetime etc., operation mode (SSF, PMF, duration) and power consumption. The requirements of solar sail orientation and attitude control will directly be merged and considered in the development of the μCPS to account for real application demands.
WP 4.2: The objective of "Propulsion Research and Development" is to support the thruster design, to elaborate the µCatalyst layout and select efficient functional layers for deposition on the chip to optimise the decomposition of hydrazine. In this context, the following topics are addressed for the analysis of micro fluidics:
- State of the art of super-hydrophobic (to reduce pressure drop) and super-hydrophilic (to increase catalytic decomposition) surfaces made of silicon or silicon carbide;
- State of the art of these surfaces in the field of microfluidic and possible ways to transform the surfaces;
- Design of the micro thruster and micro chemical propulsion systems.
- Establishment of mathematical and numerical models and performance prediction calculations.
For the development of the crucial micro catalyst the following tasks are performed:
- State of the art of silicon- and silicon carbide- supported catalysts.
- Design of micro catalysts required for decomposition processes inside micro volumes.
- Manufacture and characterization of micro-catalysts; different catalytic surface coating techniques will be assessed for surface made of silicon or silicon carbide. It is intended to investigate silicon carbide porous coatings formed by chemical interaction via solutions and CVD.
WP 4.3: The objective of "Components Development" is to develop crucial components of the thruster, such as the µValve, the µHeater and a µBubble actuator.
Within the proposed concept of the MEMS-based micro thruster the technology of micro valves becomes one of the key components providing reliable operation in terms of actuation possibility, propellant compatibility, sealing, required fast response characteristics, constant injection quantities, and a high number of activations. The objective of the µValve development is the design, manufacturing and evaluation of a propellant micro valve according to the set requirements like operational conditions, dimensions, electrical and mechanical interfaces within the μCPS under development. Initially, it will be proposed to modify the available micro valve concept. A new micro valve concept should be elaborated as the next development step in case no required improvement of the available hardware could be achieved with adequate modification measures. Within the first step available concepts and candidate components for micro valves should be assessed based on the requirements and the valve definition sheet which will be specified. Subsequently the required modifications will be elaborated for available prototypes and the manufacturing flow chart will be worked out. Finally, the µValve P/T will be manufactured and tested.
The implementation of a μHeater to heat the μChamber is required to improve the performance during the start phases of a micro thruster at low operational temperatures. The primary goal within this task is the adaptation of an already available μHeater concept which is based on the operational requirements and which considers mechanical and electrical interfaces of the micro thruster. A definition sheet will be specified for the realized design. Based on this sheet and according to space standards the μHeater will be manufactured and characterized within the test program. An alternative solution to be investigated is the integration of an external heater on the μChamber. This will reduce the complexity of the interfaces and hence might reduce the technical risks involved.
In addition, the use of μBubble Actuators is investigated for realization of a reliable, fast responding propellant injection principle for the delivery of ultra-small propellant amounts as required for a MEMS based μCPS. A feasibility study will be performed and a potentially suitable configuration will be considered. On this basis, a definition sheet will be specified and a first test device will be designed.
WP 4.4: The objective of "Test Facility Diagnostics" is to develop test facility diagnostic tools such that the thruster module and its components can be tested and the performance evaluated. A mass flow sensor is needed for reliable measurement of propellant consumption under real time conditions. For this purpose a Micro Coriolis Flow sensor will be designed and fabricated. An important advantage of flow sensors based on the Coriolis principle is that the true mass flow is measured, independent of the composition and properties of the fluid, and independent of pressure, flow profile and temperature of the fluid. A Coriolis Flow Sensor contains a vibrating tube segment in which a mass flow is subject to Coriolis forces. These Coriolis forces are directly proportional to the mass flow. The main challenge for a micro Coriolis flow sensor is to detect the extremely small Coriolis movements. Two cleanroom fabrication runs are planned in order to allow for a design and process flow optimization before fabrication of the final devices. The existing fabrication technology at UT based on silicon nitride channels embedded in the silicon substrate will be used as a basis. This should result in a sensor that is suitable for later integration into the micro propulsion system.
Besides the mass flow sensor work in, research will be performed on calorimetric flow sensors. The operation of a calorimetric sensor is based on transferring some heat to the fluid which is carried away by convection. The amount of heat that is carried away is a measure for the mass flow. Calorimetric flow sensors can be much more sensitive than Coriolis type flow sensors with a resolution down to the picoliter/minute level, which is the reason to realize these sensors in parallel. Of course, since calorimetric sensors rely on heat transport, the sensitivity is significantly influenced by fluid properties like specific heat and thermal conductance and the sensors require extensive calibration.
For plume analysis it is not possible to use a standard type flow sensor since the flow cannot be confined in a tube. Therefore, a Pitot tube sensor will be designed and realized for this purpose. The same basic fabrication technology will be used, although an adaptation will be necessary to allow for an open tube end. Various readout mechanisms (e.g. capacitive, thermal, piezoresitive) will be considered to detect the small pressure variations.
WP 4.5: The objectives of "Test Equipment, Facility and Modelling" are the definition of test facility requirements and the elaboration of the thrust measurement balance to be able to test the microthruster developed during this project and characterize its functional range. The thrust measurement balance will be used in DLR's vacuum chamber test facility. A trade off study will be performed, comparing different concepts for the thrust measurement rig. Advantages and disadvantages will be compared to find the most suitable approach to match the properties of the thruster and fulfil all given requirements. Based on this analysis a design will be selected, built and used during the final hot firing tests.
In addition, a CFD roadmap for the numerical flow simulations of µCPS will be derived which defines the path for the required extensions, modules and chemical models for the simulation of micro fluidic behaviour in µThrusters and µCatalyst chambers. The requirements, boundary conditions, initial conditions and dimensions for the CFD computations are set. With a comprehensive literature review and with a close cooperation and knowledge exchange with other work packages, the occurring flow phenomena in micro thrusters will be identified and analysed. With this knowledge, applicable numerical models will be reviewed and evaluated and a future model implementation strategy will be derived for a prospective implementation of these suitable models in the TAU code.
Since the TAU code is based on the compressible Navier-Stokes equations and can be used with structured as well as unstructured grids, the methods to be considered will be efficiently useable with all these functionalities. Focus will also be on the methods suitability with regard to the targeted coupling of the TAU code with the DSMC method to be conducted in WP5.2. The DSMC method is needed for the numerical plume expansion computations as the Navier-Stokes equations are not valid in domains of high rarefaction.
The key activities of WP5 - µCPS MAIT and CFD can be regarded as the realization phase of the μCPS within PRECISE dealing with:
- WP5.1 manufacturing, assembly and test of the μCPS prototype
- WP5.2 numerical simulation of the μCPS and enhancement of the numerical models
- WP5.3 quality assurance aspects
- WP5.4 final run-up of the test facility and hot firing tests within the vacuum chamber
-
Thereby the technological results elaborated within WP4 will be consequently implemented under compliance with the specified technical requirements. Additionally, activities dedicated to manufacturing, integration as well as verification of the μCPS prototype will be ensured applying elaborated QA processes and standards.
The objective of WP5.1 is to establish the manufacturing process for the final design of the μCPS. Critical integration processes will be investigated through feasibility tests. Subsequently, subsystems will be manufactured considering the technologies and hardware worked out within the previous research and development tasks, also with focus on the identified manufacturing methodology. Finally the subsystems and the delivered prototype of the micro thruster module will be assembled to the μCPS prototype.
In WP5.2 the numerical modelling and simulations will be performed. Based on the simulations the nozzle can be optimised and designed such that the performance of the µCPS is improved. These optimisation will be directly fed into the µCPS manufacturing.
The a priori CFD strategy has been outlined in WP4.5.2 CFD roadmap. The initial idea has been to extend CFD capabilities in the DLR TAU code in parallel to the development of the µThruster, so that, in the end of the project, experimental results can be used to validate CFD models. Objective of the chemical modelling within the PRECISE project consisted of enhancing the DLR TAU code by adding the capability of conducting finite rate chemistry of hydrazine decomposition. TAU enhancement encompasses all development activities of the DLR TAU code beyond chemical modelling of hydrazine which are necessary to accurately treat the investigated flows.
Quality assurance are also addressed within PRECISE. At present no certified methodology for the MEMS based propulsion system exists. However for further development and establishment of the investigated technology up to the highest technology readiness level reliable quality monitoring tools are obviously required. The first aim of the activities within this work package is the continuation of already running elaboration processes for the required QA methodology. On this basis the development, the manufacturing and the test activities within the project should be monitored. According to space standards the core competences and activities like the establishment of a QA plan, review test planning, manufacturing and operating procedures, non conformance and change treatment, participation on test readiness reviews and review of test results and synthesis will be carried out.
Finally, the hot firings of the developed µCPS will be performed. For verification of the µCPS within the demonstration test the available test facility (STG-MT) will be used.Within the scope of this work package the test infrastructure will be serviced and operated regarding the performance of all relevant development tests of the micro components, the micro thruster and the hydrazine demonstration firings with the μCPS prototype. For various firing modes the thrust will be measured with the thrust measurement balance to be developed and provided by WP4.5. After these first thrust measurements the comparison between expected (‘designed’) and measured data will allow an assessment on the development status of the thruster.
Based on the provided test requirements in the test plan, the test procedures will be elaborated considering such aspects like handling of facility and specimen, operational and firing sequences, measurement lists. In accordance to the test plan, the setup of the test infrastructure will be set and verified with performing e.g. calibrations of the diagnostic tools. Finally the status will be determined and used for the test release which is usually decided during the test readiness review. Based on the achieved release the test campaign will be performed.
Project Results:
The main technical and research work was conducted in WP4 and WP5. The therein developed S&T results will be explained in the following. The figures to which the text refers are included in the attached PDF-document PRECISE-figures_final-report.pdf.
WP4 - µCPS Technology Research (AST)
The general task WP 4 coordinated the technical disciplines necessary to design and manufacture the MEMS components for a µCPS propulsion module and the required test infrastructure.
WP 4.1 µCPS Requirements and Specifications (AST)
The objective of "µCPS Requirements and Specifications" was the elaboration and harmonisation of the µCPS requirements for the two spacecrafts Solar Sail and InspectorSat which finally led to a common requirement specification.
Motivation to use a MEMS based µPropulsion System
Satellite propulsion is required in order to reach a satellites final orbit, to maintain this orbit and to perform orbit changes according to a specified mission.
The required propellant mass mP for a required maneuver (or a sum of maneuvers) is a function of the mission required velocity increment delta-v, the satellite mass mS and the performance (ISP) of the propulsion system:
ISP is a key performance factor that significantly influences the required propellant mass for a given manoeuvre. Changing from state-of-the-art cold gas systems to a hot as system increases the ISP by a factor of 2-3 allowing a significant propellant mass reduction and even manoeuvres that cannot be realized with cold gas propulsion.
Commercial Off the shelf (COTS) hot gas reaction systems are available only in bigger sizes and higher thrust levels that do not allow the implementation into a nanosatellite and to perform the required manoeuvres with the thrust and IBIT level needed.
Mission concept
The PRECISE project aims to provide a hot gas propulsion system for two satellites in the Nanosatellite class with a typical weight between 1 and 10kg (see Figure 1).
Type Mini Micro Nano Pico Femto
Weight 100 – 500 kg 10 – 100 kg 1 – 10 kg 0.1 – 1 kg < 0.1 kg
Type Myriade class SSTL X50 class 3U cubesat 1U cubesat / µcube experimental
Propulsion (typical) Classical propulsion system Cold / warm gas Cold / warm gas Cold gas Without propulsion
Figure 1: Classification of small Satellites
- The first InspectorSat mission is a concept designed by the Surrey Space Centre (SSC) to demonstrate the flight and utility of a Visual Inspection Payload (VIP).
- The second satellite designed by NPO Mashinostroyenia will provide a solar sail spacecraft (SSSC) as a co-orbital observation target for the VIP.
The mission needs for the 2 spacecraft were analysed in detail [D4.406] and the requirements for the following main PRECISE µCPS were defined [D4.414]:
- Thrust level: 10mN
- Total propellant throughput: 100g (qual factor 1,5 included)
Generic µPropulsion System Layout
Figure 2 shows a schematic view of a chemical propulsion flight system and the hardware which is to be developed within the actual PRECISE project:
Figure 2: Propulsion System components and Hardware to be developed within the PRECISE project
PRECISE Thruster Chip in stacked Design
In order to develop and test each component separately a stacked design was chosen where each component has a functional layer.
Figure 3 shows the different layers which are integrated on the µCPS test unit, the hot firing test rig and the test cell:
Figure 3: Stacked layer design of functional components
WP 4.2 Propulsion Research (LCO)
The objective of "Propulsion Research" was to support the thruster design. It was divided into two sub work-packages Micro Fluidics and Micro Catalyst.
µFluidics
Figure 4 shows the various flow domains in the thruster as described in [D4.404].
Liquid hydrazine enters a vaporization chamber (1), where it is preheated and transitions into the gaseous phase. An injector (2) channels the gaseous hydrazine onto the catalyst bed (3), where it is catalytically decomposed. The hot mixture of reactants then expands from the stagnation chamber (plenum) (4) through the nozzle throat (5) and the nozzle (6), before it leaves the thruster duct at the nozzle exit (7) and expands into vacuum (8).
Figure 4: Sketch of thruster flow domains
In [D4.411] the liquid flow parameters in the µcatalyst were calculated and different catalyst bed configurations assessed. Using iridium as an active catalyst phase material the required Average Expected Geometric Surface Area (AEGSA) can be calculated based on the hydrazine flow rate and the available catalyst volume.
Figure 5: µchannel and µpillar catalyst configuration
From the results obtained taking the flow of liquid hydrazine into account, it is expected that the μChannels + Alumina configuration reaches the proposed AEGS if the μCat bed can be marginally lengthened (and/or if the alumina layer thickness is increased) see Figure 5. The catalyst inlays were manufactured accordingly.
µCatalyst
With these inputs two different supports have been fabricated, a flat silicon support with and without a thermal oxide SiO2 and metallic alloy supports with micro-channels patterning. An alumina washcoat has been developed on some of these supports by a sol-gel process prior to iridium active phase deposition. The iridium active phase has been then deposited on these supports, as well as on supports which were not washcoated with alumina, either by wet impregnation (CD) magnetron sputtering (PD).
The obtained catalysts present a good iridium adhesion onto the different supports (mechanical stability). The prototypes have been characterized by AFM and SEM to determine the iridium nanoclusters dimensions and dispersion. Finally, droplets of pure hydrazine have been injected on the different prototypes at room temperature and pressure to highlight the catalytic activity.
The catalytic activity was tested at room temperature and at an elevated temperature of 50°C with the following results:
- No noticeable hydrazine decomposition process for the different µcatalyst prototypes at room temperature
- No noticeable hydrazine decomposition process for the different NicroFer prototypes at a temperature of 50 °C (setpoint)
- Low hydrazine decomposition process for the different Silicon patterned prototypes at a temperature of 50 °C (setpoint)
- Highest hydrazine decomposition process for the 20 nm iridium layer Silicon patterned prototypes at a temperature of 50 °C (setpoint) . This was confirmed with better decomposition tested at a temperature of 80 °C.
- No noticeable difference observed for the Silicon patterned prototypes with different straight channel dimensions (at fixed iridium thickness).
WP 4.3 Propulsion Component Development (NSP)
The objective of "Propulsion Component Development" was to develop components that shall be used for diagnostics in the thruster module or at the test bench.
µValve
After the assessment of different micro valve concepts a Phase Change Material (PCM) actuated µvalves with heritage from a cold gas project was selected.
The valve concept is based on a Phase Change Material (PCM) actuation principle (see Figure 6). An enclosed cavity is filled with PCM which increases in volume when it goes from solid to liquid phase. The phase change is achieved by resistive heating and the melting point can be chosen to fit the thermal requirements of the system. The flow is modulated proportionally by adjusting the applied power to heaters in a PCM-cavity, which actuates on the valve seat due to the expansion.
Figure 6: µValve operational principle
The valves are fabricated using four fusion bonded silicon chips and each valve stack has four integrated PCM actuated normally-closed valves. Valve interfaces are designed for bonding towards the nozzle package and electrically to the pod assembly connections.
The valve seat package is two fusion bonded square chips with the purpose to interface the nozzle package and also act as a normally-closed valve seat. The valve seat chips also include via holes and channels to lead gas from the valves to the nozzles. The inlet and outlets on the valve seat package is shown in Figure 7.
µHeater
The µcatalyst test results showed that the catalyst has to be preheated in order to have acceptable decomposition rates at the given surface area. The following figure shows a sample heater that was manufactured by NSP.
Figure 7: W-coated graphite µheater
µBubble Actuator
The use of μBubble Actuators has been investigated for realization of a reliable, fast responding propellant injection principle for the delivery of ultra-small propellant amounts as required for a MEMS based μCPS.
First, a list of requirements was made and an extensive literature search was performed to obtain a good overview of existing devices that could be suitable for propellant injection in a MEMS based μCPS. Unfortunately, the high demands of a relatively large maximum flow rate of 6 mg/s combined with pressures of up to 10 bar are extremely difficult to fulfil as explained in [D4.413]. Using state-of-the-art MEMS technology we estimated that a pressure of 0.5 bar can be reached. For larger pressures, stronger check valves and multiple pumps in cascade would be needed. Explosive bubble actuators as developed earlier can deliver 0.15 mW power at best, requiring 40 cascaded actuators. Alternatively, a feed pressure could be used; however in that case simple check valves will not suffice. Due to the feed pressure both valves would open and a constant flow through the system would result. When using a feed pressure the most straightforward system will probably be a fast, controlled open/close valve that can open during a precisely controlled time period.
WP 4.4 Test Facility Diagnostics (UT)
The objective of "Test Facility Diagnostics" was to develop the Coriolis flow sensor and the prandtl tube.
Coriolis Flow Sensor
The Coriolis type flow sensor uses Coriolis forces to measure directly a mass flow. This is realized by a vibrating tube, see following figure which shows a schematic drawing of a Coriolis sensor based on a rectangular tube shape. The tube is actuated in a torsional mode indicated by ωam, using the Lorentz force on an electrical ac-current ia flowing through a metal track on top of the tube and a static external magnetic field B imposed by two permanent magnets. A mass flow Φm inside the tube results in a Coriolis force Fc. The Coriolis force is capacitively detected by its induced out of plane vibration mode with an amplitude proportional to the mass flow (see Figure 8).
Figure 8: Coriolis Flow sensor – measurement principle and built hardware
In the frame of the PRECISE project Micro Coriolis mass flow sensors for measurement of hydrazine flow were successfully realized and three sensors were built.
Prandtl (Pitot) Tube
A miniaturised Prandtl-tube sensor incorporating a 6 mm long 40 μm diameter microchannel with integrated pressure sensors has been realised.
The sensor has been designed for the characterisation of the rarefied plume flow from the μCPS. The 4.5 × 2.5 mm sensor chip incorporates 400 mbar full-scale capacitive pressure sensors.
The capacitive pressure transducer is created by merging several microchannels into a rectangular pressure membrane, with outward-facing comb-fingers hinging on the microchannel sidewall. Additionally, thermistors can measure the temperature profile on the Prandtl tube and an integrated Pirani sensor can optionally measure vacuum pressures below 10 mbar. An electronics board and stainless steel probe, from which only the Pitot tube protrudes has been realised. Initial measurements of dynamic pressure on the Prandtl tube have been obtained by flowing air from a metal tube with 0.7 mm diameter, demonstrating the feasibility of the miniaturised Prandtl -tube sensor.
Figure 9: Miniaturized Prandtl tube sensor
Calorimetric Flow Sensor
A miniaturised calorimetric flow sensor was integrated on the same chip with the coriolis mass flow sensors. The calorimetric sensors are especially useful at the lower end of the measurement range where the Coriolis flow sensor is less accurate.
Figure 10: Miniaturized caloric flow sensor
WP 4.5 Test Facility Setup and Modelling (DLR)
The objectives of “Test Facility Setup and Modelling” was to hot fire test the µthruster and to perform numerical calculations that support the thruster design.
The thrust range of common used hydrazine thrusters on the market ends in the area around 0.1-1N. The thrust measurement principle used for those thrusters at Airbus DS is not suitable for forces in the mili-Newton range. For the thrust measurement of the µpropulsion thruster, a thrust balance for experimental testing has been developed and manufactured in the scope of the PRECISE project.
Requirements Definition
Requirements, specially tailored to the expected performance of the microthruster, have been defined and reported in D4.403. The Thrust Balance needs to be able to withstand all operating media, especially hydrazine, and it needs to be able to measure thrust in the range between 0 and 10 mN. A detailed description and explanation of all requirements is given in D4.403.
Trade-off Study
Several measurement principles have been investigated and presented in [D4.403]. As a result of this study, a concept using a thrust table, mounted on bending elements, has been selected. The whole concept has been presented in [D4.412]. The measurement principle is shown in Figure 11.
Figure 11: Thrust measurement rig functional principle
The thruster is mounted on a thrust table, including all controlling electronics, required data acquisition systems and fuel supply. This thrust table is connected to a Base Plate by four flexible bending elements that are also used for power supply. If the thruster is fired, the Thrust Table will be displaced, proportional to the thrust force. The displacement is quite small, but it could be measured using very accurate sensors.
The chosen setup gives easy accessibility to test specimen and instrumentation. It is a simple and robust concept. The displacement of the Thrust Table could be adapted to match the aimed thrust level by exchanging the bending elements. Thereby the stiffness will be adapted, and the best suited thrust to displacement ratio could be chosen. This is not only important for different thrust levels, but also for the firing mode. For steady state firings, a lower stiffness with higher displacement could be chosen, to get a better accuracy. For Pulse Mode Firings, it is better to use thicker bending elements with higher stiffness, to prohibit interference between the Pulse Mode Firing frequency and the Eigenfrequency of the system. In the required measurement range, the thermal expansion of the material has to be taken into account, even by temperature changes around only one degree Celsius. This has been taken into account by selecting appropriate material with a low thermal expansion coefficient.
Design Phase
The concept trade of phase was followed by the detailed design phase. During this phase, the thrust measurement concept was further developed leading to the final design as shown in following figure. All equipment was either designed or selected from off the shelf products. Market research has been performed to find suitable instrumentation, calculations have been made to size the components and analyse the functionality, and the material has been selected based on the given constraints and requirements. The thrust measurement rig as designed consists of the elements as shown in Figure 12.
Figure 12: Thrust measurement rig
Manufacturing, Design and Assembly
After the design was fixed, all equipment has been procured and the components were manufactured. Additionally, some improvements have been implemented due to first test results and evaluation of the available hardware.
Implementation and run up
Final calibration and run up of the thrust balance was performed at DLR. As the thrust measurement rig is equipped with a voice coil calibration system, it was first required to calibrate the voice coil itself before this system could then be used to calibrate the thrust balance. Theoretically, there is a clearly defined force-current relationship of the voice coil. However, it turned out that the metallic parts of the thrust measurement rig are affecting the electromagnetic field of the voice coil and hence the force used for calibration. Also, the electromagnetic field of the voice coil also interfered with the electric scale that was used to measure the force. Most of the time initially planned for the thrust measurement rig run up was required to calibrate the calibration mechanism itself, until a final solution has been found.
Therefore less data than planned is available on the thrust measurement rig functionality. For steady state firing, the force to elongation relationship has been calibrated, which was then later used for the testing of the micropropulsion thruster. The thrust balance and its instrumentation worked therefore quite well, however external influences as e.g. induced by the vacuum pumps reduced the resolution due to vibrations. Nevertheless, the thrust measurement balance has registered and measured thrust during the hot firing tests and therefore proven its functionality.
Summary and Outlook
A propulsion system based on liquid monopropellants represent a significant improvement compared to currently available cold / warm gas systems.
The actual PRECISE project proved that a chemical µpropulsio system based on MEMS manufactured components is feasible. All necessary components successfully proved their functionality.
Based on actual requirements the following development work has to be done in the next step:
- Use of a non-toxic propellant instead of hydrazine: due to low decomposition temperatures a monopropellant system based on H2O2 seems to be a promising candidate. Material compatibility and H2O2 self-decomposition rate is a key point for this propellant. Synergies to currently running activities (ESA funding) to be used
- Development of a propellant tank in the 100ml range and additional components like valves, filters, service valves. Choice of tank configuration (bladder tank or propellant feeding via surface tension tank)
- Adaptation and verification of analytical tools in order to precisely predict the flow and heat transfer in µchannels
- Integration of a maximum number of components in order to use the advantage of a MEMS based system
- Optimization of catalyst and heater configuration in order to allow an optimum decomposition of propellant
WP5 Introduction
The key activities of work package 5 can be regarded as the realization phase of the μCPS within PRECISE dealing with:
• manufacturing, assembly and test of the μCPS prototype
• realization and run-up of the test facility
• numerical simulation of the μCPS and enhancement of the numerical models
• validation of numerical models through comparison with experimentally obtained data
Thereby the technological results elaborated within WP4 were consequently implemented under compliance with the specified technical requirements. Additionally, activities dedicated to manufacturing, integration as well as verification of the μCPS prototype was ensured by applying elaborated QA processes and standards. For verification of the micro chemical propulsion system within the demonstration test the available test facility (STG-MT) was used after implementation of the equipment provided within WP4.4 and WP4.5. In addition to numerical performance calculations, the flow solver TAU and the DSMC method were enhanced through the implementation of physical models and methods elaborated within Task 4.3.2 and Task 4.5.2.
NSP has been the work package leader and responsible for the coordination of the activities within WP5. Within WP5.1 NSP designed, manufactured, and delivered several MEMS-based microthruster parts including the prototype of the μCPS and monitored the activities related to QA. AST provided the required system engineering containing interfaces of micro thruster, specification of test requirements, supervision of the test setup and test performance as well as analysis of the test results. AST also provided one part of the micro thruster prototype for the integration within the μCPS prototype, and the major part of the integration unit. DLR was responsible for the run-up and operation of the STG-MT test facility in WP5.4 and performed preliminary analysis of the test results. The numerical simulations and the enhancement of the flow solvers were also covered by DLR in WP5.2.
WP5.1
The complete micro chemical propulsion system (µCPS) can be presented in different ways. One way, which also shows the modular thinking with exchangeable layers within the integration unit is shown in Figure 13.
Figure 13 Schematic of the µCPS presented to illustrate the modular approach with exchangeable layers in the integration unit stack.
However, WP5 mainly concerns the integration unit i.e. the “rocket engine” of the system.
After the conceptual modular approach was established the work continued with requirement and design specification. A definition of the MEMS micro Chemical Propulsion System (μCPS) hardware and also the fabrication sequence in terms of a Manufacturing Flow Chart was agreed. The related subsystems, which together with the μCPS comprise the PRECISE demonstrator system was also defined in a complete product tree. Part of that tree showing the parts of the integration unit are illustrated in Figure 14.
Figure 14 Product tree of the manufactured, integrated, and tested µCPS.
An exploded view of the integration unit design is shown in Figure 15 and some of the corresponding parts during assembly in Figure 16.
Figure 15 Exploded view of the integration unit design.
Figure 16. Integration unit parts during assembly prior to cold testing followed by shipping to DLR for hot firing tests.
During the course of this development project a huge number of different designs, and different material combinations, were produced and evaluated for every component in Level 4 in the product tree.
As an example µInserts were made in both metal and mono-crystalline silicon bulk material. Only the silicon version was made in twelve designs with two different etch depths, deposited with a matrix of several catalyst layers in different thicknesses using deposition techniques. IN addition to this the µInserts were manufactured in four different sizes. Three versions of the µInserts together with matching µChamber chips are illustrated in Figure 17.
Figure 17. a) µChamer chips, µInserts, and µDiagnostic chips spread out in three pairs
b) Close-up of µInserts with 25µm wide walls and 100 µm wide and 200 µm deep trenches.
A good example of the interdisciplinary exchange within the project is the design of the micro nozzle. The foreground from WP5.2 was implemented in an updated design of the micro nozzle hardware in WP5.1. See Figure 18.
Figure 18. µChamber chip (15x26 mm) selected as baseline for integration unit#2. for the µCPS #2. Photograph (left) and CAD-drawing (right).
Prior to hot firing tests with real hazardous propellant. Leakage and flow tests were performed with gas (aka cold testing). A validating flow test result is shown in Figure 19. Figure 20 shows the integration unit before the hot firing setup including harness.
Figure 19. Flow test results from the two completely assembled integration units.
Figure 20. Integration unit including harness prior to hot firing tests.
WP5.2 Numerical Modeling
First, an assessment of CFD contributions had to be carried out based on the physical phenomena break-down identified in the CFD roadmap.
Multiphase decomposition of hydrazine is a problem with substantial practical experience in the PRECISE Team (AST, LCO). We thus decided that CFD can contribute most effectively for the outcome of PRECISE by studying nozzle flow.
WP 5.2.1: Numerical Simulation of the Micro Thruster
A systematic study of geometrical and thermochemical model properties showed that a flow field prediction is (unfortunately) only feasible using a specieswise temperature dependent modeling in a 3D field. 2D simplifications, such as planar or axisymmetric greatly overpredict performance and were thus found to be inadequate. Furthermore, micro scale planar nozzle flow deviates significantly from their well known macroscopic, axisymmetric cousins; established methods and procedures thus do not apply here: It was found that the dominant feature degrading microscale nozzle performance is boundary layer build-up, threatening to choke the flow in the divergent part of the nozzle. Figure 21 illustrates this.
Figure 21: Thrust along the axis is found to stagnate around X=0.0005m and to degrade substantially after X=0.0015m.
This result suggests to use a shortened nozzle design to enhance performance while reducing size. Figure 23 shows both nozzles in comparison. The result of this optimization is an increase of predicted thrust from 7.1mN to 9.2 mN (+29.6%) and of specific impulse from 144.8 s to 187.6 s (+29.6%).
Figure 22: Comparison of original and truncated nozzle flow field. The red plane is the Mach 1 isosurface.
It was found that, contrary to conventional nozzles, a short conical nozzle with a wide exit angle is advantageous at the scales investigated here. Based on our findings, a new nozzle design has been suggested.
This geometry was provided to Nanospace and implemented in the actually manufactured hardware.
WP 5.2.2 Chemical Modeling
The infrastructure for chemically reacting flows is available in TAU and applicable for combustion modelling or the simulation of dissociation effects as occurring in atmospheric re-entry.
However, the catalysis of hydrazine demands new models for the reaction kinetics and was obtained in close collaboration with WP4.2. A two-step Hydrazine decomposition chemical kinetics mechanism has been integrated and shown to describe decomposition of N2H4 into N2, H2, and NH3.
WP 5.2.3. Enhancement of the flow solver with applicable models
The modelling depth of nozzle flow study presented in Enschede revealed that full 3D, viscous, high fidelity fluid model computations are (unfortunately) necessary to capture the characteristics of high aspect ratio rectangular micro nozzles, such as boundary layer dominance and effect of variable ratio of specific heats on expansion.
A theoretical assessment showed the high expected dependency of nozzle flow on gas composition and heat capacities. A CEA computation showed that, in addition to N2H4, a mixture of N2, H2, and NH3 can be expected as expanding gas. Detailed representations of these species have been implemented into TAU.
As multispecies 3D computations are numerically costly, an efficient way to create grids and start solutions is sought. Thus, a cartesian grid generator has been developed. This greatly enhanced geometry parameter studies: previously, for even minor changes in geometry, a new CAD model had to be created, from which in turn a new grid has to be set up.
Additionally, this required the computation to be started from scratch.
Now, any numerical description of the planar nozzle can be directly used to produce a new grid onto which a previous solution is mapped as a start solution, see Fig. 23.
Figure 23: Interpolation of initial flow field in grid generator.
To account for liquid flow inside feed lines and the catalyst bed, a new real gas model has been implemented. This Euler-Euler compressible multiphase model allows both vaporization and expansion through a convergent-divergent nozzle to supersonic velocities within the same code. So far it is a promising development surpassing the well established “Volume of Fluid” model, which does not allow for a widely varying density. The inherent vaporization model captures the constant temperature plateau during phase change without the need to track vapor-liquid interfaces or volumes. Systematic validation for different cases were performed, an example is shown in Figure 24. The model captures the vapour pressure, difference of onset and offset enthalpy match the latent heat. This study has been carried out for various pressures. Furthermore, the model restraints are tested. Fluid property library resolution has been found to have an effect on liquid fluid pressure oscillations which can be remedied by choosing an appropriate library resolution. Library size has been found to have negligible effect on computational cost.
Figure 24: Exemplary plot of vaporization model validation. Temperature plateau during phase change is successfully captured.
To improve convergence of the low Mach number flow of a liquid, first steps towards implementing a preconditionig method have been undertaken.
A literature study yielded a model applicable to the real gas flow regarded herein. The required differentials of the equation of state have been calculated and implemented into the real gas library model.
Validation has been carried out compared to literature and numerical values. The preconditioning matrix has been implemented.
WP5.3 Quality Assurance
At present no certified methodology for the MEMS based propulsion system exists. However for further development and establishment of the investigated technology up to the highest technology readiness level reliable quality monitoring tools are obviously required.
The first aim of the activities was the continuation of already running elaboration processes for the required QA methodology.
The development strategy for MEMS fabrication at NanoSpace is adopted according to industrial MEMS experiences, i.e. to identify critical steps and try to manufacture feasible structures and components in silicon and iterate towards functional systems. The keyword is process compatibility! On this basis the development, the manufacturing and the test activities within the project was monitored. Most focus has been on the MEMS manufacturing, where most of the non-conformancies was reported and taken care of. In short the MEMS manufacturing process sequence is as follows: a) Write a Process Identification Document (PID). Make a feasible process sequence using standard processes and materials. b): Make the mask layouts in a CAD-programme (e.g. AutoCAD).c) Call for a PID Meeting. During the PID meeting the process sequence will be checked. When the process sequence is checked and approved by the Chief Engineer, a product name is defined and a batch number is allocated. The PID shall be saved as “PID N### Title” with the specific batch number followed by the title. The PID contains the intended process. A Process Flow Chart is generated and saved the same way. This finishes the meeting and the batch is released for manufacturing. c) Send the sorted mask files to the mask manufacturer. d) Start the batch.
When a batch is finished, the PID is saved as LOG
Delivery: A delivery is documented with a Delivery Note.
If a non-conformance has occured, a non-conformance report (NCR) has to be written.
The QA process can be schematically illustrated as in Figure 25.
Figure 25. Block diagram of the used QA process for MEMS fabrication.
WP5.4 Hot Firing test of the μCPS
The scope of this work package covers both the development of the test infrastructure as well as the actual hydrazine demonstration firings using the μCPS prototype.
The test set-up, comprising the subsystem μCPS (Figure 26), hydrazine tank and tubing (Figure 27), thrust balance (Figure 28), DAQ, vacuum chamber including pumps. The micro thruster was installed in a vacuum chamber on the DLR thrust balance, developed in WP4.5 in order to verify hot fire thrust.
Figure 26. μCPS
Figure 27. Tank and tubing as mounted on the thrust plate of the measurement rig
Figure 28. System under test in vacuum chamber.
The evaluation system was equipped with a large number of sensors, in order to measure primarily temperature and pressure. In addition to tough vacuum conditions, the toxic and harsh propellant (hydrazine) complicates calibration etc.
The final hot firing test was planned according to the matrix in Figure 29. The sequence is shown in Figure 30.
Figure 29. Test matrix for the hot firing.
Figure 30. Hot firing sequence.
There was a clear visible thermal reaction, but steady state was not reached and hence no proper thrust measurements possible. The shorter hot firing pulses the more reaction but also lower thrust pulses out of scope of the thrust balance range/resolution, see Figure 30.
WP6
The μComponents developed in WP4.3 have been tested in WP6. These are the μValve and the μHeater realized by NanoSpace and the Bubble actuator test structures realized by the University of Twente. Delivery and testing of the μValve and μHeater components was delayed slightly due to the move of NanoSpace to new facilities and the need for setting up the test infrastructure. After finalized hook-up and start-up of the new test facility, preparations for testing were done and then all experimental work including documentation was done in the months February and March 2013. Due to time constraints several engineers worked in parallel to successfully demonstrate the valve functionality using liquid in combination with MEMS valves and also heater experiments were carried out on test chips with custom sized cavities to simulate catalyst beds inserts. The objective of the performed development testing was to support the design feasibility and to assist in the evolution of a design prior to WP5. Such development tests are often used to validate new design concepts and the application of new concepts and techniques into a new configuration, in our case the use of µValves in combination with liquid and the evaporation of liquid using µHeaters. Successful testing of these components was an important step towards the realization of the complete µCPS in WP5.
The Bubble actuator test structures realized by the University of Twente were also delayed. The fabrication of these structures in WP4.3 turned out to be extremely difficult and most wafers were lost in the fabrication process. Some test structures did survive and these were evaluated in WP6.2. Most importantly, it could be concluded that more successful fabrication of devices should be possible with some adjustments to both the fabrication process and the design. Research towards an integrated Bubble actuator which started with PRECISE will continue at the University of Twente. A new fabrication run has started in July 2014 and new devices are expected in September.
Potential Impact:
PRECISE fosters the cross-border cooperation’s and partnerships between renowned European organisations and leading research centres across Europe and beyond. PRECISE unites experienced and advanced companies for chemical micro propulsion and merges their expertise to develop a unique, efficient and reliable µCPS. This approach improves the industrial supply chain and reduce future product development lead times. The geographical spread of beneficiaries includes France, Germany, the Netherlands, Russia, Sweden and the United Kingdom.
The MEMS-based monopropellant µCPS is an emerging technology which is currently under development and is considered amongst satellite experts as one of the key technologies for future satellite missions due to the very low thrust levels and compactness the system provides. The analysis of the current unmanned commercial and scientific spacecraft, satellites and missions of the last decades, as well as the foresight into the next decades shows a clear change of the design philosophy; an increasing trend towards compact systems is observable. Thus, the propulsion system has to become smaller, more accurate and more efficient to fulfil the requirements for exact and highly dynamic positioning of distributed small satellite systems. The system to be developed within PRECISE fulfils these requirements and is the next technology step improving available MEMS cold gas propulsion concepts.
The investigated formation flying concept of the solar sail together with the inspection satellite and the inspection unit for target observation are seen as an important stepping stone towards totally autonomous systems. SRY and NPO see the development of the small MEMS thruster as providing the last missing piece in enabling the technologies to mount this mission.
µCPS has been identified by ESA to fill the gap between state-of-the-art electrical and chemical propulsion due to its outstanding system parameters like the low power consumption and its low system weight. The technology status, mission needs and market perspectives are summarized in the ESA µCPS roadmap.
Already initiated cooperation projects between AST, SRY and NPO on this field are directly related to the targeted development of MEMS based hydrazine monopropellant system within the µCPS roadmap. Results of the mission elaboration as well as considerations on the spacecraft design will be implemented and continued within PRECISE.
German Aerospace Center - DLR
The cooperative activities within PRECISE were essential to increase the knowledge and the understanding of DLR regarding the flow inside µThrusters. The optimisation of the nozzle design could be derived from these outcomes and realised with the help of the consortium, in particular Nanospace who manufactured the thruster prototypes on short notice.
The testing infrastructure had to be setup in a novel way to be able to realise the final hot firings of the µCPS. The experience of Astrium was of great help for the testing layout. Furthermore, novel sensor devices could be tested and included in the testing chain, such as the prototypes from UT. In addition, for the analysis of the final results the experience of LCO was very important for the interpretation of the sensor data.
All these results were presented on worldwide conferences, taking place mainly in the United States of America and in Europe, national and European workshops and published in conference proceedings and Journal papers.
LACCO
a) Professor Kappenstein delivered a course for students, colleagues and PRECISE beneficiaries at Conference Hotel Drienerburght in Enschede, the Netherlands, on the 8th February 2013: " Course - Kinetics, Catalysis and Propulsion. Fundamental basis, Objectives - Current challenges". This course was divided into two sessions: Session 1 – Catalysis and Catalysts (duration: 1 h), Session 2 – Catalysis for Propulsion (duration: 1 h). The presentation materials which were used during this short course (117 pages) is too large to include it in this report but can be obtained by contacting directly Prof. Kappenstein from LCO.
b) Efforts in dissemination were also undertaken by Professor Charles Kappenstein (LCO). His dissemination activities in this period include:
- Open lecture “From the laboratory to space, an exciting adventure: kinetics, catalysis and propulsion.” with a presentation of the PRECISE Project:
Le Studium conferences, 31st March 2014, Orléans, France, (50 participants, lecture given in French)
NanoSpace
This project has been a great opportunity to increase NanoSpace know-how on chemical propulsion. NanoSpace has established a good working relation and had knowledge exchange with relevant expertise within the consortium.
The work performed is well in-line with the micropropulsion road map NanoSpace follows as seen in Figure 32.
Figure 32. Nanospace µ-propulsion roadmap
NanoSpace ambition is to continue the development of the microthruster module, the MEMS-based part, and make it a product and hopefully sell it to a propulsion system supplier as Astrium.
NanoSpace is a supplier of MEMS-based products for space. It has already flight demonstrated a highly miniaturised MEMS-based cold gas system. Currently, hot gas system (resistojets) are beginning to mature and sold as engineering models. The natural next step in the product line development is to develop a chemical microthruster, which will both increase the specific impulse and the thrust level (See figure SS in section above).
NanoSpace ambition is to continue the work on the thruster module, which are MEMS-based, eventually in collaboration with a catalyst supplier (from the consortium). This could be done in a three-to-five year perspective.
If so, NanoSpace intend to try to protect the method of fabrication of some crucial parts by applying for patent as usual. However, it might need another similar sized development project to reach TRL-5 or TRL-6 before a productificaiton can be made solely by NanoSpace. Otherwise, this needs to be done in close collaboration with a prime contractor.
Surrey Space Center
The research into the potential investigation in space of solar sail properties has directly influenced other research within SSC related to solar sailing. There is a plan for the inspection of a Cubesail spacecraft from another cubesat as part of the ESA QB50 mission which builds directly from the work in PRECISE.
The research has been disseminated internally within SSC and SSTL through seminars given on the development of sail inspection and visual inspection payloads. We also have advertised the work carried out on the testbed facility through the Astrodynamics website:
http://www.surrey.ac.uk/ssc/research/astrodynamics/research/index.htm
The research has also featured on SSC Facebook and Twitter accounts to advertise to the wider public.
University of Twente / MESA+ Institute for nanotechnology
PRECISE has allowed the successful realization of a micro Coriolis flow sensor with a 20 times larger flow range than earlier sensors and a micro Pitot tube sensor which is the first sensor in its kind. Although the research is far from completed and several issues still need to be solved, important progress has been made. Operation of the Coriolis mass flow sensor was demonstrated successfully and this will open up new application areas for this sensor, also outside the field of space systems. The results were presented at several conferences and workshops: MFHS 2014 (Freiburg, Germany, October 2014), Sensor 2013 (Nurnberg, Germany, May 2013), 8th ESA Round Table on Micro and Nano Technologies for Space Applications (Noordwijk, The Netherlands, October 2012), MFHS 2012 (Enschede, The Netherlands, October 2012). The sensor is also used in two master courses at the University of Twente: “Measurement systems for mechatronics” and “MEMS design”. A micro Pitot tube sensor did not yet exist and the sensor realized within PRECISE was presented at the IEEE MEMS conference (San Francisco, USA, January 2014), which is the flagship annual event of the MEMS community. The sensor contains a capacitive pressure sensor that can also be integrated on other microfluidic chips based on the same technology. It is already included in several new flow sensor designs and contributes to a new research theme towards microfluidic multi-parameter sensors, i.e. sensors that measure not only a single quantity like flow or pressure but also medium parameters like viscosity, density and thermal conductivity.
In addition, national activities in the field of µCPS are ongoing and strongly fostered in France, Germany, the Netherlands, Sweden and the United Kingdom:
• LCO is developing monolith-based catalysts within an ESA contract "Preparation, Characterization and Evaluation at the Lab Level of Catalysts for Low Thrust Bipropellant Thruster". Further, they are a partner in the FP7 project GRASP.
• NSP’s miniaturized cold gas propulsion system is demonstrated on PRISMA – a European mission to demonstrate autonomous formation flying and rendezvous. In addition, NSP’s parent company SSC has so far developed in total six spacecraft. Ongoing activities are Small GEO, Proba-3 and SMART OLEV
• UT is partner in the NanoNextNL programme, funded by the Dutch government, which aims at increasing and spreading knowledge in the field of Micro Systems MST and MEMS. Special attention will be given to problems that obstruct the further application of MST in industry. In this context, the Micro Coriolis flow sensor and the Pitot tube for plume measurements are first spin-out products of the MESA+ institute / TST group to be used in space systems. Thus, PRECISE successfully opened a new field of applications and opportunities for the institute
• DLR and AST are coordinating their national activities on the thruster and propulsion sector with the national MoU “Propulsion 2020”. The MoU and the therein included activities were initiated in 2006, also involving PhD students
List of Websites:
www.mcps-precise.com
Dr. Markus Gauer (markus.gauer@dlr.de)