The Fourth Asian Space Conference 2008 Taipei, Taiwan, October 1-3, 2008
ESEMS – AMEI BCU
Tse-Liang Yeh1, Shyh-Biau Jiang1, Huey-Ching Yeh2, Jen-Yann Liu2, Ko-Ichiro Oyama2, Ying-Hao Hsu1, Li-Yet Liu1, Chung-Jen Chou1, Jia-Wei Jiang1, Ji-Yi Peng1,
1Institute of Opto-Mechatronics Engineering, National Central University (NCU-OME)
2Institute of Space Science, National Central University (NCU-ISS)
nspotr-asc_tly97831.doc A1 97831 rev tly 912 917
The Fourth Asian Space Conference 2008 Taipei, Taiwan, October 1-3, 2008
1. ABSTRACT
An Aurora Magnetic-field and Electron energy distribution measurement Instrumentation - AMEI package, nick-named - the Block of Central University BCU was developed as part of the ESEMS Project. to investigate space weather signs for earth-quake prediction by the National Central University Scientific Payload Team.
Originally, the instrumentation included the Magnet-Resistance Magnetometer MRM, an Electron Temperature Probe ETP, and an Electron Energy Spectrometer. Regretfully, EES was delayed due to perceived contract limitation on the weight. The orbit will be sun synchronous, dusk to dawn at 11 o’clock local time. Instrumentation is operated at 150ms base sampling period, the one standard deviation accuracy achieved by MRM is 60 nT, those of 33pA in electron current and 6.8mV in reference potential by ETP, and 2.22 degree celsius by the six temperature sensors located around the case of the package.
The structure of data acquisition, command and control, and the qualification process of payload engineering for the satellite integration are described.
The functionality and survivability tests have been carried out. The flight unit was delivered to the Tatyana Team for integration with the MicroSat for launch in the third quarter of this year.
2. INTRODUCTION - Mission Purpose
As a launch opportunity into a sun synchronous dusk to dawn orbit was available at the same time the earthquake precursor in the ionosphere anomaly was discovered [Liu 2004, Devi 2004, Pulinets 2004, Oyama 2008], an instrumentation package focusing on the measurements of Aurora Magnetic field and Electron energy distribution (AMEI – lady in Taiwanese) was naturally proposed by the National Central University team. The proposal included a high speed Magneto-Resistance Magnetometer (MRM [Liu 2005]), an Electron Temperature Probe (ETP) [Oyama 1999], and a channel energy selector based Electron Energy Spectrometer (EES [Hardy 1984]) for nadir direction. However, the implementation, due to the perceived weight budget allowance on ESEMS – Tatyana II, the EES was sacrificed from the implemented Block of Central University (BCU). Six temperature sensors (TEMP) were installed on the walls of BCU instead to learn the experiences on the thermo calibration of MRM.
The BCU structure, which integrates the payload instruments, and the process to develop BCU for the qualification of the micro satellite will be discussed.
3. The Structure of BCU
Since MRM is fast responding, its three axes output can be sampled simultaneously at high speed. The three-dimensional coordinate calibration matrix can be adapted online [Liu 2005]. And, there is great potential to develop online extraction of the wave propagation behavior from the high-speed magnetic field measurements to research for the precursor sign of earthquake. These potential requirements on computational power warrant the utilization of a digital signal processor as the core to operate the instruments and to process their data [e.g. Mabry 1993]. The functional structure of AMEI-BCU is shown in Figure 1.
Figure 1 – Constitution of AMEI-BCU.
In addition to payload instruments, electric power supply converters, an interface connection to the Command and Data Handler C&DH (also known as Block of Information BI) for in flight controls and data downlink, an AVR boot loader, and its interface connection to an Engineering Computer for firmware upload and debugging are also included. At power on or hardware reset, the DSP would download its firmware codes from the nonvolatile EEPROM and transfer program execution to the codes.
When AVR is plugged in, it can arbitrate the access of EEPROM so that it can upload new firmware codes through its RS422 interface with an Engineering PC and then write into EEPROM before resetting the DSP to execute the new firmware. The receiving end of RS422 is high impedance differential ended, therefore, it can be conveniently tabbed by the RX+,- of a RS422-USB converter to an Engineering PC. This way, the engineering PC sees the data in both directions of the communication between BCU and BI making the debugging of the information integration very easy.
Figure 2 –EIA-422 communication interface.
Since the data through put of MRM may exceed that of the micro satellite data downlink, onboard Flash memory data buffer is provided so that data sampling can be continued over a full scan of the earth surface without interruption nor data loss. In accordance, interfacing commands are designed to allow the data sampling be switched on and off, buffered data be uploaded by BI for down link to ground, and also flash memory buffer reset to leave room for logging data on a new event. A "get ready for Power-Off" command is also necessary to save flash memory indices into EEPROM so that data buffered in the flash can still be accessed after electric power turn off and turn on. Calendar information is passed from BI to BCU every time when data sampling process is switched on for robust update. Calendar time tag is attached to every sampling data sequence and system data sequence for identification and event correlation.
The commands and data are exchanged between BCU and BI through the RS422 port of DSP. Two engineering commands, Sensor_Report and Sensor_Reset, are also designed for sensor checking and calibration while DSP is connected to an engineering PC instead of the BI.
Figure 3 – Timing Phases of Data Sampling in BCU
Instruments are operated in timed phases. In addition to avoiding driving signal switching before data measurements to minimize mutual interferences, allowance of enough signal settling time before measurements is also critical to the quality of measured data. Therefore, the data processing DSP needs to operate in two parallel loops: the timer-interrupt driven loop and communication driven command and response loop [Jiang 2005]. The plan to schedule a proper operation of instruments for good measurements is programmed in the timer loop, while the alteration of operation parameters and modes as well data transfer out of BCU is executed in the command and response loop as shown in Figure 3.
The detailed reasoning of BCU architecture design is presented in a companion paper in the student competition session [Hsu 2008].
Figure 4 – Command Processing
The effects of control commands are passed to the timer loop through global flags and parameters, while measurement results are passed to the command and response loop through the data buffer, communication packet bugger, and their access indices set in the timer loop. Setting a flag to switch the sensor data output from data buffer to communication buffer in the timer driven loop, we switched the operation from the normal flight mode to Sensor Report debugging and calibration mode while keeping the timing schedule in fidelity. The systematic uniformity in programming structure brings efficiency and reliability to the project.
4. THE PROCESS OF SATELLITE PAYLOAD DEVELOPMENT
In a scientific payload project, there are scientists to define a make sense collection of instruments whose data would bear enough information to review discriminative insights of the domain to be explored. The scientists also need to specify the nominal magnitude, the dynamic range, the resolution, the variation, the frequency response, and accuracy of the measuring devices so that the data bear enough discriminative power. The scientists capable of instrumentation are also responsible for the development and the calibration of the instruments.
The basic responsibility of the engineering team is system integration: coordinate, control, operate, and taking measurements with the instruments. The engineers need to coordinate the resources of time, space, and power to avoid interferences. The engineers also need to facilitate operational environmental conditions. Many a times, good engineers also need to customize the instrumentation packages to meet budget, weight, dimensions, power, environmental, and data throughput constraints.
The engineers also need to interface with the satellite team to reserve resources and accommodations in the satellite development process, as well as to get qualified for integration into the satellite system. The payload engineers must know both the concerns and development steps of the micro satellite team.
The micro satellite team would request a resources and installation requirements upfront to get their design work started. If a payload package is still under development at the initiation of a micro satellite project, foreseeable margin in the budget and foreseeable concerns must be listed in the document. Otherwise, painful and even handicapping sacrifices would be inflicted. Requirements of installation, namely orientation, field of view, attachment, environmental conditions, vulnerabilities, accommodations, requirement of data volume, electrical and information interface connections should be formally provided to the micro satellite design team.
Resource budget needs to be updated to actual measurements of electric power consumption properties, mechanical properties, and electro-magnetic interferences and susceptance. Electric power consumption characteristics include power on current surge, power on minimum current limit, operational voltage range, loading current fluctuation, and power off voltage surge. Mechanical properties include mass, center of gravity, coordinate system, moment of inertia.
The survivability qualification is the next the satellite team is concerned with. Before full environmental test documents can be presented to earn the qualification, the satellite team pays no interest in the functionality nor the communication protocol of the payload. However, once the structure and major resource providing modules of the satellite system get designed and manufactured, the pressure of the time to integration crashes down on the payload team: all certified survivability test documents (the formulary, the log book) should be ready for review, otherwise, the time window to get on the flight passes by. Once the documents are reviewed, the physical package are examined, information interfacing with BI kicks in, and the operational manual and in flight operation plan (the cyclogram) should be filed and approved. The payload leaves the hands of the payload engineering in care of the satellite team. Tests in the system integration are to be conducted and reported by the satellite team according to the operation manual prepared by the payload team.
Therefore, the payload engineering should get aware of the environmental test requirements for each particular launch project so that no unqualified tests would be carried out wastefully. The test requirements cover the expected vibration, impact, temperature, moisture, vacuum pressure, heating, radiation during storage, transportation, launch, transition to orbit, and space mission. Quality of soldering, secured wiring and attachment, and apriori survivability screening of components are critical to survivability.
The team should also estimate the time table of the micro satellite team and schedule their tasks and responsibilities self autonomously. Otherwise, the human resources can be sunk in fulfilling paper work requests from the satellite project manager much ahead of payload development work. The functionality design, implementation, verification, calibration, and environmental tests are the solitary responsibility of the payload engineering warranting top priority. The design definition of the first successful prototype can be delivered to the satellite team cordially. If there is no active response from the satellite team, the payload team do need to device a way of integration with minimal BI efforts yet resulting in acceptable reliability and feasible data processing complexity.
The payload engineering also need to run through foreseeable scenarios of data reception to realize what kind of calibration data is necessary for data analyses. Scientists cannot image what an imperfect world can be, therefore, it rely on the engineering team to develop appropriate parser to organize received information stream into tokens with physical meaning for scientific analyses. The process of data processing must be executed to get debugged with deficiencies in calibration tests.
The engineering teams should be aware of the necessity of a flight unit and the availability of at least one engineering unit. The reasoning of how they should be tested and utilized individually should be familiarized to avoid dreadful mistakes.
If samples of necessary documents were made available to a novice payload team, the period of learning stage could have been much reduced. A tutorial workshop would reduce the threshold even further. Project management is vital to the quality of a project.
Software Development and Methodologies
Implementing the operation of payloads in functional phases with accurate real time sampling while keeping the flexibility of command and controls and adaptability to different payload functional characteristics, the software structural design by state machine and the software management by modularization based on the concept of software IC are applied in the process of software development [Jiang 2007].
Every year a generation of senior students must grow out of our "opto-mechatronics engineering core course series for the implementation of automatic controls" to carry on the legacy in payload engineering. Systematic uniformity in programming structure and style also makes the heritage of both hardware and software from one student generation to the next every year much easier.
5. Performance of Instruments
5.1 Calibration Test Results on MRM
The set and reset raw data in volt obtained by the sensing circuits alone three axes in the Helmhotz chamber are shown in Figure 5.a. The magnetic field applied is shown in Figure 5.b in the range of –50000 to +50000 nT in X, Y, Z directions respectively. Applying the transformation matrix to the raw data to get the correct magnetic field component values appropriate calibration matrix parameters were obtained by least square parameter estimation. The estimated X, Y, Z component measurements in volt in response to the applied magnetic field are shown in rows of Figure 5.c. The horizontal axis shows the magnetic flux density applied. The magnetic field was applied in X, Y, Z directions respectively in the three corresponding columns in Figure 5.c. The calibrated measurements can be obtained with the standard deviation of 58, 62, and 52 nT in the three directions as shown in Figure 5.d. With improved magnetic field compensation algorithm, standard deviation less than 10 nT at 30KHz sampling rate is already possible [Chou 2008]. The temperature drift in the measurement is shown in Figure 5.e.