RAFT PRE-SEPARATION CHARGING SAFETY ANALYSIS 9 Feb 2006 -------------------------------------------------------------------------------- The RAFT picosat separation switches are in the main battery-to-load circuit but they do not disconnect the solar panels from the system bus. In normal configuration, the alternate battery pack is wired to trickle charge from the bus. This configuration poses a pre-separation charging hazard to address. HAZARD #1: The maximum possible Solar Power current in full sun on three solar panels, is 290 mA of which 60 mA is a constant load of the receiver. This leaves 230 mA available for inadvertant charging. This is about 0.2C or about twice the trickle charge rate of the 1100 mAH NiCd battery system. However, while the picosats are in the launcher potential charging currents are drastically reduced. CONDITIONS: The Picosats are only handled and procesed indoors. Exposure to full sun during processing would be extremely rare and is easily avoided. After being integrated with the SSPL Launcher, they are totally enclosed and can only be exposed to full sun via light leaks in the deployer top and sides. The top has a 1? finger hole in it and one side has an access panel with 120 ¬? holes in it. This represents about 7 square inches of light leakage. MITIGATION: 1) The sun cannot be orthogonal to both the side and the top at the same time. This reduces the possible exposure from 7 down to about 6 sq in. 2) The solar panels are series strings of 18 cells each and the current from a solar panel is limited by its weakest cells. In this case, more than 90% of all cells remain in the dark, so the spotty light can not possibly produce rated current. 3) The holes in the access panels fall equally half on one picosat and half on the other. This reduces the effective area to only 3 sq in per picosat. CHARGING EFFECT: The potential charging current then is only 3 sq in compared to the on-orbit value of about 50 sq in (the Cosine(Theta) equivalent of corner illumination of three 25 sq in panels. This is 3/50 times 230 mA or about 15 mA. Since the nominal 10 hour trickle charge capacity for these cells is 110 mA and the safe "float" rating is about 1/3rd of that (35 mA), then this 15 mA can be maintained indefinitely with no damage, nor overheating of the cells. This 15 mA would produce no more than about 0.12 Watts of waste heat. This 0.12 Watts of waste heat is insignificant relative to the same heat gain the launcher top would receive from the Sun. Assuming a conservative absorptivity of 0.3, the solar heat gain of the launcher lid under the same sun conditions would be about 5 Watts. Thus the heat dissipated in the batteries due to inadvertant charging is less than 3% of the heat transfer and is insignificant. HAZARD #2: Battery B2 is trickle charged from the main bus during all non- transmitting conditions (96% of the time). It has one cell less than the main battery. To prevent abnormal circulating battery-to-battery currents in the case of a failed (shorted) cell in B2, a dual fault tolerant design is used consisting of the following features: 1) The voltage drop of the missing cell is made up by the sum of the voltage drops across a diode (0.7) and transistor Q1 (0.7). So under normal conditions there is no energy transfer from B1 to B2. When the solar bus (or B1) is above 7.2 volts nominal (1.2v per cell) charge current will flow into B2 as set by the forward bias of Q1 which is set to approximately 110 mA (the safe continuous trickle charge rate for B2). 2) Under the condition of a failed cell in B2, there are multiple current limiting protection mechanisms: A. The current is limited by Q1 to 110 mA (0.1C trickle charge rate) B. The current is limited by PTC fuse to 250 mA (0.23C rate) C. The solar panel system is current whole orbit limited to 120 mA avg 3. Peak power dissipation under these overcharge conditions of 0.23C capacity is 2 Watts, but the B2 battery pack is thermally conductive to the overalls paceframe which is already under a nominal solar flux dissipation condition of 22 Watts to the blackness of space. The additional 2 Watts of excess dissipation under fault conditions could only raise the temperature of the spacecraft about 2% (or about 6 degrees C). Even if the thermal resistance from B2 to the spaceframe is poor, say 50% then any temperature rise in the B2 cells due to a shorted cell is insignificant to the normal temperature fluctuations (+/- 10C) expected per orbit. The nominal spacecraft temperature is between 0 to +20C CONCLUSIONS: The charge current resulting form solar energy leakage onto the solar arrays while in the launcher and in the Payload Bay is about 1/3rd of the safe, indefinite float rating of the cells. Further this 15 mA would produce no more than about 0.12 Watts of waste heat which is insignificant to the solar heat gain the launcher top would receive from the Sun. Bob Bruninga US Naval Academy Satellite Lab 410-293-6417