Application of Solid State Recorders to Spacecraft

Abstract

The Solid State Recorder (SSR) on the NASDA’s Mission Demonstration test Satellite-1 (MDS-1 or “Tsubasa”) which was launched in February 2002 has finished one-year operation, and post-operational phase is on going now. The collected data of commercial devices (64Mbit DRAM) as a memory on SSR affected by space radiation are reported. Here, 1) ratio of Single Event Upsets (SEU), 2) distribution of Total Ionizing Doze (TID) effect on body frame, is clarified. However a significant difference of 3) trend of consumed current on memory, 4) correlation of refresh interval and SEU ratio are not recognized, and 5) dynamic memory relocation of physical damaged memory chips are not executed yet. Thus we would like to examine describe the results of 1) and 2) mainly in this paper. INTRODUCTION The objectives of MDS-1 (Fig.1) mission are to verify the function of commercial parts in orbit, to verify minimization technology for components, and to measure space environment data. MDS-1 was launched February 4, 2002 into the geostationary transfer orbit aboard H-IIA launch vehicle No.2 from NASDA/Tanegashima space center. The SSR on MDS-1 is experimental module that aims to demonstrate its mechanism, performance and durability of commercial semiconductors with high density Chip Sized Package (CSP) technique as recording devices. We have analyzed flight data of SSR and summarized the following; (1) Trend of SEU ratio and validity of shield effect by the body flame (2) Distribution of TID effect inside of the satellite are clarified. On the other hand, (3) Trend of consumed current on memory devices (4) Correlation between DRAM refresh interval and SEU ratio are not recognized apparently and, (5) Degradation and/or failure of memory devices (6) Memory dynamic relocation caused by above (5) automatically are not executed yet. We have finished routine operation of SSR on February 2003 and here report above-mentioned (1) and (2) mainly. Besides we would like to evaluate total system capability of SSR, for instance, consideration with trend of TID effect with long period, and severe temperature cycle in space environment. Fig.1 MDS-1 (“Tsubasa”) MISSION OF MDS-1/SSR Mission Purpose It is very important to use on-board data recorder on non-geostationary satellite because of an existence of communication black-out period. NASDA have studied and developed solid state recorder the reason 54th International Astronautical Congress of the International Astronautical Federation, the International Academy of Astronautics, and the International Institute of Space Law 29 September 3 October 2003, Bremen, Germany IAC-03-U.2.a.01 Copyright © 2003 by the International Astronautical Federation. All rights reserved. 2 that advantage of no-mechanical/rotation parts, low power consumption and high access speed compared with conventional magnetic tape recorder or hard-disk. So solid state recorder utilizes plenty of semiconductor devices which hardware parts itself and stored data should be protected from physical damage and software bit errors. Experiment of SSR’s main mission purpose is: (1) To acquire the status of hard/software error in space environment (2) To confirm the function or performance of Error Detection and Correction (EDAC) mechanism (3) To evaluate capability of stack memory module with high-density commercial semiconductor devices Finally, total SSR system availability as for a flight component is demonstrated and evaluated thorough experiment. SSR Overview SSR was designed to simulate data recorder as a flight component. Also 64Mbits DRAM stack memory module as a recording media is same as real bus module. Beyond that some functions listed below are added to accomplish this experiment. (1) Recording data is not from other mission instruments but from SSR’s own signal generator (known data) (2) Comparison function between known data and stored data is added. This result is collected on-board directly without other error sources e.g., communication link noise (3) Bit errors in all memory domain are counted by using EDAC mechanism (4) Some parameters are available as below; Two types of EDAC code EDAC enable/disable DRAM refresh interval SEU patrol period Generated data patterns (fixed, pseudo-random) Major specifications of SSR flight model are described in Table-1 and overview is shown in Fig.2 SSR Experiment There are two types of observing aspect to evaluate a condition of SSR while this experiment typically; one is measurement of SEU, the other is that of TID effect. Regarding SEU, we count the number of SEU and compare with value of simulated data. So SSR uses 768 chips of 64Mbits DRAM, as a lot of sample chips lead high data reliabilities, in other words, practical flight data. Details are introduced in later section, but shortly summarized; the speed of recovering data is faster than that of bit error accumulation, so that all data can be corrected. As for TID effect, 14 Dosimeters (DOS) which are also developed by NASDA are set up in SSR, accumulated dose (from 1 to 10 Gy) and temperature (from -129 to 133 degrees Celsius) can be measured through experiment. In addition, consumption current is measured by SSR to detect Single Event Latch-up (SEL) too. Put simply there is no symbolic data on current because value of TID is also few. Details are introduced the following section. Table-1 SSR Major Specifications Data Volume 32Gbit (4GB) for mission data 16Gbit (2GB) for check bits Memory 64Mbit DRAM stack memory module EDAC Reed-Solomon or Hamming code Mass 12.1kg Dimension 247 x 460 x 314mm Power Consumption 22W Fig.2 SSR flight model