Pulsed Plasma Thruster High vacuum Chamber Facility Report

Transcription

1 Pulsed Plasma Thruster High vacuum Chamber Facility Report 1

2 Faculty Mentor Dr. Daniel White Project Lead Aditya Khuller System Leads Joseph Mayer (Structures / Vacuum) Tyler Lanes (Command System / CAD) Brian Amaral (Structures) Patrick Kennedy (CAD) Rahul Verma (Vacuum) Team Members Dominic Bonelli (Command System) Omar Alavi (CAD, Structures) Surya Sarvajith (ANSYS sims) 2

3 Table of Contents Introduction 4 Project Summary 5 Deliverables and Objectives 7 Project Timeline ( ) 8 Project Expenses 9 Project Timeline ( ) 10 Project Components 11 CAD 12 Electrical Characterization 16 Command Sequencing 19 High vacuum Chamber Facility 21 Thermal Modeling 25 Machining and Fabrication 27 Strategic Partnerships 29 Impact 30 Future Planning/Outreach 32 3

4 Introduction Within the electric propulsion community, there is an increasing interest in the development of small electric thrusters to provide primary propulsion and attitude control, especially in a configuration that is suitable for integration within a cubesat platform. This has led our team to begin research and development of a Pulsed Plasma Thruster (PPT). PPTs are high, specific-impulse, low power electric thrusters. These thrusters are ideal for applications in small spacecraft, such as CubeSats, for use in attitude control, station keeping, and low thrust maneuvers. Additionally, PPTs can be used as a spacecraft's primary propulsion system for certain missions. Pulsed plasma thrusters are the first electric propulsion (EP) technology to fly in space and have been studied and developed since the Russian launch of the Zond-2 satellite in Replacing the standard momentum wheels and torque rods with a PPT system to perform attitude control maneuvers reduced the ACS (Attitude Control System( mass by 50 to 75% with no increase in required power over comparable wheel-based systems. PPTs accelerate small quantities of ablated fluorocarbon propellant such as PTFE, more commonly known as Teflon, to generate very small impulse bits from (100 to 1000 µns) at a high specific impulse (-1000s) (Myers et al, 1995). PPT attitude control systems provide many mission benefits through system simplicity, high specific impulse, and increased operational and mission lifespans. PPTs exploit the natural properties of plasma, via the Lorentz force, to produce thrust at high velocities accompanied by low fuel consumption. These systems also have the ability to operate at higher vacuum pressures comparable to those experienced at altitudes of geostationary orbits making them ideal for use on satellites.. With the advent of the versatility and utility of CubeSats to enable multiple, low-cost, high-risk science missions, a need for reliable, long-lasting propulsion systems has never been more prudent. 4

5 Project Summary The Sun Devil Satellite Laboratory (SDSL) Pulsed Plasma Thruster (PPT) team is currently performing research and development of novel PPT designs as a ultra-compact, efficient and low-power propulsion solution for CubeSat platforms. We have successfully tested a coaxial, benchtop PPT (that was fired without the use of a spark plug, a major life limiting factor) for use in testing our electrical systems, and to explore possible increases in efficiency by utilizing nontraditional designs and the removal of life limiting factors. Figure 1: Our revolutionary 8 thruster, 0.5U CubeSat PPT design. We have begun the design and development of an integrated 0.5U PPT unit for station keeping and ACS (attitude control) for small spacecraft in Low Earth Orbit (LEO). This unit consists of a revolutionary, world-first 8 thruster CubeSat PPT system, allowing a wider range of maneuvers including but not limited to: pitch, roll, and yaw. A compact, energy efficient high voltage power processing unit (PPU) is being designed along with a ready-to-use controls system software package for ground station personnel to implement maneuver and firing commands over I2C commands from the CubeSat on-board computer. 5

6 Figure 2: A successful benchtop PPT test fire (without a spark plug / igniter). Additionally our project contains the design and construction of a high vacuum chamber facility rated at 10-8 Torr. This system is vital to this work for environmental testing and characterization but is also an asset to the SDSL club as a whole. This vacuum system will allow concurrent along with future projects to be tested in low and high vacuum environments. We currently have access to multiple roughing pumps rated to 10 3 torr, along with a diffusion and turbo pump rated to approximately torr. This system also includes a mass spectrometer that will allow for the testing of any particulates released within the chamber during a test. 6

7 Deliverables and Objectives 8 independent thrusters contained in an integrated 0.5U package Innovative thruster design having 8 thrusters for complete, multi-axis control to include, but not limited to pitch,roll and yaw maneuvers. Command and control software package Development of a command and control software package for use by ground station personnel to initiate maneuvers and fire actions. Efficient electrode configuration and geometry Research into the size, shape, and placement of the electrodes along with the use of novel materials and electrical systems to increase system efficiency. Viability of applied external magnetic field Conduct tests to find the viability of using an applied external magnetic field to direct and accelerate plasma. The strength and source of the magnetic field will be our main focus. Power processing unit design An innovative and cost effective circuit design incorporating an efficient and high speed switching system to ensure low power losses and high energy output. Investigation into high efficiency capacitors allowing greater power densities while simultaneously shrinking system size. High vacuum chamber facility Concurrent to the design of the PPT, the design and development of a high vacuum chamber facility. This will allow for testing of not only the PPT but also additional SDSL projects (NASA-funded Phoenix CubeSat mission,etc.). 7

8 Project Timeline ( ) Fall 2017 Semester (August - December) o Characterization, testing of power processing unit to include: PPU Discharge curve Capacitor function / energy density test Test multiple switch packages and configurations o Connect power processing unit to coaxial test thruster and perform test of different switching circuits and packages o Production of completed 8 thruster prototype PCB o Completion and testing of high vacuum chamber facility o Completion of thermal, stress modeling in ANSYS o Begin plasma physics modeling Spring 2018 Semester (January - May) o o o o Produce complete prototype of 8 thruster PPT system Complete ground station command and control software Test system functionality in high vacuum chamber facility Test fire prototype 0.5U thruster unit 8

9 Project Expenses Items Purchased Line Item # Line Item Name Cost 1 M2.5 Pan Head Screws $ Washers $ Nylon / Teflon / Ceramic Standoffs $ High Temperature Adhesive $ Nylon Threaded Rod $ Wing Nuts $ Miniature Snap-Action Switch $ Ultragrade 19 Rotary Vane Pump Oil $ High Voltage Oscilloscope Leads $ Custom PCBs Electrical Components (Capacitors, Switches,etc.) Aluminum for Machining Total Cost $ Projected Costs Line Item # Line Item Name Approx. Cost 1 Space-rated Compact, High Energy Density Capacitors $250 2 Vacuum Systems Equipment $300 3 PPT Housing Material (Ultem/Torlon) $400 4 Custom PPU PCBs $40 5 Langmuir Probe $400 6 PPT Structural Materials $150 9

10 7 PPT Propellant (Teflon) $150 8 Machining, Shop Work $100 Total Cost $

11 Project Components This multi-faceted project consists of various components: CAD Electrical Characterization Command Sequencing Vacuum System Thermal, Mechanical Modeling 11

12 CAD Figure 3: Current PPT Assembly full (left), PPT Assembly with transparent housing (right) The CAD is developed in SolidWorks and components are organized in folders with folder as well as individual labels for ease of modification as seen in Figure 2-2. A single thruster is modeled with the housings and components are mirrored as the orientation allows. Some are simply redrawn as needed individually. All modifications are made in assembly by editing individual components or adjusting mates to ensure parts line up and fit. Figure 4: Feature Tree of PPT Assembly 12

13 The mechanical design of the PPT model uses 4 pairs of thrusters with those above offset perpendicular to those below as seen in Figure 2-1. The design is nearing its first prototype to fabricate. A flange extends from a side of the casings above the PCB board which line up with casings below so a small screw can travel through a top flange, the PCB board, and a bottom thruster flange. Since there are two on each side connected to both sets of thrusters above and below, no rotation can be expected from the thruster assemblies after each flange screw is assembled. As seen in Figure 2-3, the casing includes pins to hold electrodes, a stop for keeping the Teflon in place, and a flare to aid in thrust produced during firing. A top is separate from the casing to allow ease of assembly as the electrodes have holes for pins from the top and bottom casing parts to line up in. A bracket holds the top of the casing onto the thruster with a screw and a slotted pin prevents movement along the top of the casing. Some material is added to the thickness of the inner wall of the casing to give some extra clearance between the end of the screw attaching the bracket and the side of the inner electrode. A tungsten tip is used to replace a spark plug and an insulator is to be fitted between the sparker and inner electrode it travels through. It extends out from the casing to aid in attaching to the circuit which is more visible in Figure

14 Figure 5: Single thruster assembly with outer electrode and casing top transparent Tracks which line up the thrusters and hold the Teflon in place have stops behind each casing to spread any stresses it may cause on the thruster assembly during assembly or testing. There is also a stop in the casing connected to the upper stop of the Teflon to prevent damage of the electrodes by a compressive force against the pins. Refer to Figure 2-3 for the ends of tracks. A spring holding the Teflon in place as it decreases length it held up by an assembly pictured in Figure 2-4. Both stoppers have a cylinder shape offset higher than center and travel about 45% the length of the compressed spring to keep the spring in an ideal position. For ease of assembly, one has a rod that travels into another which line it up before being placed in the tracks. Figure 6: View of spring assembly with outer track transparent The next steps at this stage will be to scale up the components of v (pictured) and 3D print to refine the attachments and any potentially misaligned components. After a new version is made based upon these findings, then drawings for fabrication can be made, reviewed, and used for 14

15 fabrication for an initial prototype. Electrical components will be fit onto the board as final components are chosen. Further modifications will follow after the prototype is tested. Figure 7: (Above) An engineering drawing of the benchtop PPT thruster. (Below) A drawing of the current primary casing for the CubeSat PPT thruster 15

16 Electrical Characterization A high voltage PPU (Power Processing Unit) has been characterized, using LTSpice simulations and lab testing.the figure below shows the modelling and simulation that has been done to determine electrical characteristics such as power dissipation, dv/dt, etc. The flowchart of the PPT shows how the 3-5V voltage from the spacecraft bus is taken and boosted to 1kV via an EMCO HV booster. The 1kV signal is the input to both the flyback circuit and the discharge circuit in parallel. The flyback circuit receives commands from the OBC (On-Board Computer) via i2c protocol based on user input. Although the OBC determines which thruster to fire, the flyback circuit and discharge circuit exhibit behavior that will be consistent no matter which thruster is chosen. Figure 8: Block/Stage Diagram of CubeSat Power Processing Unit (PPU) The flyback circuit steps up the 1kV DC signal to roughly 20 kv DC using a 1:12 ratio transformer with a high voltage transistor for switching. The transistor requires a pulsed square wave with 40kHz frequency, 50% duty cycle. The flyback circuit will step up to a high enough voltage that the spark plug ignites and can initiate the discharge. The discharge circuit takes the 1kV input stored in the capacitor bank and discharges it through the PPT electrodes, with the assistance of the igniter spark from the flyback circuit. 16

17 Figure 9: Flyback, Igniter circuit simulation (LTSpice IV) Currently, prototyping and testing is being done to determine which switch will most effectively minimize power dissipation. A complete power processing circuit has been constructed and the electrical and software team members are attempting to integrate the command structure described in the following section. Figure 10: Flyback voltage transient results Future tasks for the electrical team will be converting the prototyping being done with Arduinos and breadboards, into actual PCBs with surface mount components. Idealized designs can be seen in the CAD modelling section. 17

18 Figure 11: LTSpice schematic for benchtop PPT Figure 12: Voltage across PPT (blue) and current through benchtop PPT (green) in LTSpice simulation 18

19 Command Sequencing The PPT s firing commands are controlled by an on-board computer (OBC) and a separate firing controller (uc). The OBC acts as the master in an I2C relation. It decodes incoming serial commands from a source such as a user s computer and transmits the command via I2C protocol to the slave uc. The uc reads in the transmitted data and activates the corresponding thrusters. The OBC actively waits in a loop for serial commands from the user and only transmits the data when a character has been received. When this happens, the PPT will decode the command, set up the firing sequence, and then shut down the uc while the thruster is firing. The intent behind shutting down the uc is to prevent it from being short-circuited by the electromagnetic pulse which occurs during the discharge. After firing, the uc reboots and the code loops. To operate the PPT from a remote computer, the operator will enter a single character which corresponds to a command. As shown in the tables below, the single thruster commands are accessed through an ASCII character which corresponds to the byte that tells the uc which thruster to fire. For the multi-thruster (maneuver) commands, values 0 through 5 are used. The command system is being prototyped using Arduino. An Arduino Mega 2560 acts as the OBC while an Arduino Uno acts as the uc. To simulate a command, a character is sent through the serial monitor on a remote computer. The Arduino Uno is wired to eight LEDs which represent the eight individual thrusters. If the command is a single thruster fire, only one LED will light up, and if the command is a maneuver, two LEDs will light up. Figure 13: Prototype of command control system using Arduino 19

20 Figure 14: Table of single thruster fire commands and their corresponding values: Single Thruster Fire IF VALUE (DEC) VALUE (HEX) VALUE (ASCII) COMMAND Top Front Left A Top Front Right B Top Rear Right C Top Rear Left D Bottom Left Front E Bottom Right Front F Bottom Right Rear G Bottom Left Rear Figure 15: Table of multi-thruster fire commands and their corresponding values: Multi-Thruster Fire (Maneuver) IF VALUE (DEC) VALUE (HEX) COMMAND THRUSTERS FIRE Negative Pitch Thrusters 3+4 Top Rear Positive Pitch Thrusters 1+2 Top Front Negative Roll Thrusters 7+6 Bottom Right Positive Roll Thrusters 5+8 Bottom Left Negative Yaw Thrusters 6+8 Bottom Rear Right + Bottom Front Left Positive Yaw Thrusters 5+7 Bottom Rear Left + Bottom Front Right 20

21 High -Vacuum Chamber System Figure 16: CAD model of vacuum chamber, thruster ledge and feedthrough vs actual assembly. The current vacuum chamber system which was designed and developed in-house, consists of two aluminium enclosure plates, an acrylic cylindrical wall, and a vacuum gauge assembly. We have access to a number of vacuum pumps including two rotary vane roughing pumps, a high vacuum diffusion pump, and a high vacuum turbo pump. The two stage rotary vane pump, a positive displacement type, can provide a pressure of 10-3 mbar. The pump consists of an inlet & exhaust valve, spring, stator, rotor, blade and an oil reservoir. When air enters the inlet valve, an eccentrically mounted rotor traps, compresses and transfers it to the exhaust valve. The valve is spring loaded and allows the air to discharge when atmospheric pressure is exceeded. 21

22 Figure 17: Rotary vane pump principle of operation The oil diffusion pump, also known as a fluid entrainment pump, is quite robust due to it not having any moving parts and it being able to function over a pressure range of approximately 10 2 to Torr. The assembly consists of an inlet and outlet, oil vapor nozzles, cooling lines, heater, and a baffle. The pump heats the oil and uses the vapor to capture air molecules. The cooling lines reduce the temperature of the oil vapors when they contact the wall and air molecules are released. The combination of gravity and the downward direction of the vapor moves the air molecules toward the bottom of the pump. By continually forcing the air molecules downward a high pressure area is formed at the bottom of the pump. Figure 18: DIffusion pump principle of operation 22

23 Turbo pumps, also known as turbomolecular pumps work on the principle of kinetic energy transfer. When air enters through the inlet port, angled blades rotating at a high rate of speed transfer kinetic energy to the air molecules and propel them downwards at high speeds. An ultimate pressure of about Torr can be achieved. Figure 19: Turbomolecular pump principle of operation 23

24 Figure 20: A schematic diagram of our high vacuum chamber facility The above schematic shows the basic structure of our high vacuum chamber system. The primary or backing vacuum pump removes the bulk of the air and other gases from the vacuum chamber. Once the pressure in the chamber is reduced, removing additional molecules becomes exponentially harder. As a result, a secondary high vacuum pump is utilized to pull the vacuum level within the chamber to lower levels suitable for testing purposes. 24

25 Thermal Modeling The pulsed plasma thruster is going to experience various thermal loads such as thermal stress on the casing and the electrodes during Teflon ablation and gasdynamic expansion. The PPT also experiences heat accumulation due to the discharge of voltage from the capacitors. The heat from ablation of the teflon block contributes to the thrust requirements of PPT, but the heat from capacitor discharge is going to add stress to the components around it. Nevertheless, a heat flow has to be determined for the thruster to determine the heat sinks needed to remove the heat generated from the capacitors. For determining an approximate temperature distribution during teflon ablation a steady-state thermal analysis was performed on the thruster. The analysis was performed on ANSYS steady state thermal analysis module. The FEA solver solves the following heat conduction equation. The following materials were applied to the thruster model. Spark Plug - Tungsten Electrodes - Copper Propellant - Teflon Casing - Torlon 4203 The analysis was performed to obtain results on temperature distribution in atmospheric conditions. It was performed in atmospheric conditions to test the temperature distribution for a worst case scenario and due to the limited temperature inputs in the solver. The future tests would be performed using COMSOL to better emulate the actual conditions of the PPT working environment. The initial temperature was set to 295 K. Assumptions that were made was that the PPT was firing at a constant rate. The temperature for the surface of the propellant at which the teflon block starts to ablate is between 600 K to 850 K. For a lower ablation rate the surface temperature of the teflon block was chosen to be 673 K. The heat is transferred to the surface of the casing from the teflon black through convection. Since the analysis was performed for standard atmospheric conditions the film coefficient for convection was chosen as standard air simplified case. The results for temperature distribution from the teflon block onto the casing is provided in the figure below. 25

26 Figure 21: Temperature distribution for a single thruster Future analysis would include the transient thermal analysis on the thrusters and the capacitors, Vibrational and dynamic analysis on the structure. 26

27 Machining and Fabrication A majority or the Pulsed Plasma Thruster systems along with the high vacuum chamber components were machined and fabricated in-house at the Sun Devil Satellite Laboratory, with help from the Mechanical & Aerospace Engineering Machine Shop at ASU. With the help of the Mechanical & Aerospace Engineering Machine Shop at ASU we were able to cut a hole in the top of the vacuum chamber endplate, thus allowing the use of an electrical feedthrough that greatly increasing our testing abilities. Recently we redesigned the way in which systems are held within the chamber by machining and adding standoffs which will be welded onto the endplate creating a vacuum tight seal. The standoffs have been designed to accept nylon threaded rod which will replace carriage bolts previously used to hold up our test bed. A secondary 6061 Aluminum endplate was needed for the vacuum chamber since our original one has been modified with a connection to accept our diffusion pump. With the help of the Arizona State University Polytechnic campus machine shop we were able to cut a new endplate using a CNC Plasma cutting table. Figure 22: Turning Teflon down to the proper diameter on a lathe 27

28 Our original proof of concept PPT electrode was machined out of a C101 Copper round bar and the teflon fuel sample was turned down to the proper diameter and a through hole drilled in it with the use of a lathe. Our prototype of the PPT housing will be machined out of either Boron Nitride or Torlon using standard machining practices on a manual end mill or a CNC mill. Some of our parts will be off the shelf components such as stainless steel machine screws, nylon standoffs, and gaskets. With the resources available through Arizona State University and the knowledge of our structures team we will be able to accomplish these tasks and any others which may arise. 28

29 Strategic Partnerships With an increasing focus on having SmallSats accompanying science-class missions for Earth and interplanetary missions, active research in electric propulsion being conducted concurrently at ASU and a strategic partner can serve to catalyze a collaborative relationship. Figure 23: Dr. Dan White, professor at the School for Engineering of Matter, Transport & Energy at Arizona State University during benchtop PPT testing For example, individual component research, development and testing (Hall thruster cathode design improvement, for example) can be conducted at the Sun Devil Satellite Laboratory at ASU under the guidance of Dr. Dan White to help improve an overall high-level partner technology (a Hall thruster, for example). 29

30 Impact The PPT team at SDSL received no external funding (we receive around $1000 from the Fulton Schools of Engineering at ASU for our project). Despite that, we have engaged students within the club as well as members of the public (through outreach events like Fulton s DiscoverE Day and SESE s Night of the Open Door) in a number of different ways this past year. Figure 24: Aditya Khuller (Left) and Joseph Mayer (Right) inspect the electrical connections on the benchtop PPT This project in particular has helped garner a lot of interest amongst old and new club members, and has significantly boosted club member recruitment. We were able to do this by giving the students a number of projects that they could jump into and get some real world experience not found in the classroom. Skills including setting up and performing space and engineering experiments, learning to work together on teams and subsystem teams, using engineering methods to design and build prototypes (by using CAD software as well as machine shop tools) and go beyond what is taught in class by learning about vacuum systems, their operation as well as electromagnetic processes to increase thrust and efficiency. These skills are paramount to learn how to be an engineer in the real world, with 30

31 real problems and real risks, with budgetary and engineering constraints to achieve a tangible goal. Biweekly presentations on progress and lessons learned are encouraged to bolster confidence and improve presentation skills. By developing our own in-house vacuum chamber and testing facilities, SDSL can then allow At a smaller organizations (like other clubs, and even nearby high schools) a chance to get their hands onto our testing equipment for their projects under our supervision. 31

32 Future Planning / Outreach The Sun Devil Satellite Laboratory (SDSL) has put itself into a great position to expand its horizons in the next few years. On April 6, 2016, NASA announced that SDSL s Phoenix mission had been chosen to be funded and launched on the SLS Heavy Launch Rocket. SDSL is currently leading a team of undergraduate students, including members of the Pulsed Plasma Thruster project, at Arizona State University for this mission. Led by experienced faculty such as Dr. Judd Bowman and Dr. Philip Christensen, the Phoenix CubeSat is on schedule to be launched in The mission will fly a thermal microbolometer on a 3U CubeSat to measure the effects of the heat island phenomenon in different cities across the United States. Figure 25: Below (left to right): SDSL s NASA-funded Phoenix mission; SDSL s AIAA Cansat competition; SDSL s Pulsed Plasma Thruster research and development project The AIAA annual CanSat competition has given us (SDSL) a great base to build our organization on. It has provided SDSL with a yearly, multidisciplinary project that involves building, designing, flying and successfully recovering an integrated circuit module while recording flight data. Networking within Arizona State University, industry, and government entities is the key to future growth of SDSL. With the help of professors on campus we are communicating with JPL, NASA, and industry leaders through the ASU NewSpace initiative to receive a better understanding of what these organizations are looking for in student clubs such as SDSL. By better understanding what these organization s requirements are, we believe that SDSL will be able to develop a working relationship with them. Partnering with organizations such as Raytheon, we have been able to accomplish all that we have on our projects. 32

33 We are currently working on implementing a K-12 outreach program with nearby schools. We would like to integrate their Physics lessons, with an Application lesson about electric propulsion. SDSL would like to implement this program in the hopes of inspiring students into attending college and enrolling in STEM programs while there. We believe that by educating and sparking an interest in the younger generations we can ensure a bright future for space exploration and spacecraft technology. We have also considered implementing usage of the popular rocketry game Kerbal Space Program in order to give students a fun and exciting way of understanding the concepts behind launch vehicles and orbital dynamics. Figure 26: A demonstration of SDSL s gravity well at ASU Homecoming The PPT project has also created an informative guide to electric propulsion systems which are distributed during outreach events. Our project has also developed ties with The Society of Physics Students at ASU, who frequently do demonstrations of Lorentz-force applications allowing us to show real world demos of electromagnetic waves in space propulsion. 33