|
Size: 361
Comment:
|
Size: 7571
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 4: | Line 4: |
| Line 9: | Line 10: |
| This could be the show stopper. Mass shielding is out of the question, | Silicon dioxide develops a positive charge when irradiated. An ionizing particle passes through, and generates hole-electron pairs. The electrons are highly mobile, and diffuse or drift out, while the holes get trapped, and leave a positive charge. Hafnium oxide develops a negative charge, trapping electrons. A stack of both shows promise as a rad-hard gate oxide, withstanding 10Mrad from a Cobalt 60 source with minimal shifts. I wonder if that is tuned for Co60? Perhaps a wider spectrum of radiation energies, as would be found in the Van Allen belt, would preferentially charge either the !HfO or the !SiO2, leaving a residual imbalance? In any case, it does demonstrate how modern gate oxides may be much more rad hard than older technologies. === Tantalum shielding === Tantalum capacitors over the tops and bottoms of chips can act as both shielding and bypass. Tantalum has a density of 16.7 gm/cm^3^ at 300K, so a 30 micron thick layer would add 0.05 grams/cm^2^ of shielding. According to [[SMAD]] figure 8-18, that might reduce the 1-15keV dose by a factor of 3-5, and perhaps a factor of 2 for electrons (figure 8-19). [ More accurate information needed] . To accommodate the extra thickness in the launch stack, the processors and memory and other chips might sit in interdigitated positions on every other server-sat. MORE LATER {{{#!wiki caution '''Volunteer Opportunity: Question ''' - does a small tantalum cap short out when an ionizing particle passes through it? That would actually be good, because it could remove power from the segment of a chip that is being hit, possibly removing drift fields from the gate oxide and helping immobilize charges. }}} === Space Junk and Debris === Server sky will be deployed in orbits higher than most [[http://en.wikipedia.org/wiki/Space_junk|space junk]]. The version 0.1 design deploys server-sats in m288 orbits, with semimajor axes of 12781 kilometers and average altitudes of 6411 kilometers. The vast majority of space debris is in lower orbits - it requires high launch velocities to even reach those altitudes. Most of the relevant debris will be associated with geosynchronous transfer orbits. Upper stages will be in inclined orbits from higher latitude launch sites (28 degrees [[http://en.wikipedia.org/wiki/Kennedy_Space_Center|KSC]], 46 degrees [[http://en.wikipedia.org/wiki/Baikonur_Cosmodrome|Baikonur]], 20 degrees [[http://en.wikipedia.org/wiki/Wenchang_Satellite_Launch_Center|Wenchang]]), and 5 degrees ([[http://en.wikipedia.org/wiki/Guiana_Space_Centre|Kourou]]). [[http://en.wikipedia.org/wiki/Molniya_orbit|Molniya]] orbits are lower, bit are less common. Before they decay, neither pass through the equatorial plane at m288 altitudes. Even for debris with apogees that cross the equatorial plane at m288, the area of that plane crossing is vast, 15 million square kilometers for a server sky region 200km across. A given debris particle intercepts on the order of square meters or square centimeters per pass, and crosses through only once every two hours or so. If its orbit decays in 5 years, it will make less than a thousand passes before its apogee falls below m288. For example, a 10m2 booster stage, passing through 1000 times, the chance that it will hit any particular server sat (without avoidance) is on the order of 100 parts per billion. Booster stages are easy to track by radar, so they will almost surely be avoided by server sat maneuvering. Smaller debris will be more numerous, and have higher surface-to-mass ratios, so those particles will decay faster. While any given debris particle may have a very small total intercept area through the m288 zone, there may be tens of thousands, and they will be harder to track, so this probably represents a larger problem. Note that very small particles will punch a hole through the server-sat, but probably won't shatter it. This may disable some portion of the server-sat (probably a segment of solar cell), but with a little redundancy the server-sat will still be maneuverable and capable of reduced activity. {{{#!wiki caution '''Volunteer Opportunities''' Compute the actual flux of particles that intercept the server-sky orbit. Compute the actual orbital elements of spent booster stages in decaying geosynchronous transfer orbits from various launch centers. }}} Some small fraction of this orbiting debris may have perigees low enough to suffer appreciable drag in the extreme upper atmosphere. This will have the effect of lowering their apogees down into the server-sky orbit, to potentially collide with server sats. A problem with such decaying orbits is that their orbital elements become somewhat unpredictable - the apogee location will depend on the orientation and drag of the debris as it passes through the atmosphere, so it will be difficult to predict more than half an orbit in advance. However, there will still be perhaps an hour of tracking time between the most recent perigee and the passage through m288, and at 1 micron per second squared acceleration (5 percent of maximum), that is enough time for a server sat to move 2 or 3 meters out of the way of a precisely predicted orbital track. The key point here is precise prediction - to reduce the chances of a collision to zero, the radar systems will need to be very good. The loss of one server sat through a collision is not a big problem in itself - there are plenty more where that came from. The biggest risk is the debris resulting from the collision, which will tend to be in the same orbits as the rest of the server-sats, and may pose a long-term collision hazard for those. The effect of solar wind on these uncontrolled bits of secondary collision debris is unknown. In the long run, it will be better to remove space debris before it collides with the server sats. The server-sats are travelling faster than the debris they are likely to encounter, so any collisions between them and debris at apogee will most likely increase the orbital energy, raise the perigee, and lengthen the decay time of the debris. To remove debris (lower it's perigee into thick atmosphere) it must lose apogee velocity, either by colliding with material in a suborbital ballistic, material with an opposite inclination, or with material in a retrograde orbit. The important thing is to make sure that the small particulate debris from such collisions is captured. This is an unsolved problem. === Surface charging === Most of the area of a server-sat will be solar cells, with low voltages across the surface. If the LCD thrusters have glass covers with Indium oxide conductors on both sides, the outer conductors may be grounded relative to the rest of the server-sat. As all surfaces will be conductive, and at voltages less than one or two volts, there is no opportunity for arcing across surfaces, which sometimes happens to geosynchronous satellites in the solar wind ( see [[SMAD]] page 212-214 ). |
| Line 20: | Line 63: |
| Line 22: | Line 66: |
[[http://www.isde.vanderbilt.edu/content/muri_2008/dixit_muri2008.pdf]] Sriram Dixit et. al. at Vanderbilt University. Recent work on HfO/SiO2 stacked gates and radiation resistance. {{ attachment:800px-Ap8-omni-0.100MeV.png }} |
The Space Environment
The inner and outer van Allen belts
MORE LATER
Ionizing radiation and semiconductor damage
Silicon dioxide develops a positive charge when irradiated. An ionizing particle passes through, and generates hole-electron pairs. The electrons are highly mobile, and diffuse or drift out, while the holes get trapped, and leave a positive charge. Hafnium oxide develops a negative charge, trapping electrons. A stack of both shows promise as a rad-hard gate oxide, withstanding 10Mrad from a Cobalt 60 source with minimal shifts. I wonder if that is tuned for Co60? Perhaps a wider spectrum of radiation energies, as would be found in the Van Allen belt, would preferentially charge either the !HfO or the SiO2, leaving a residual imbalance? In any case, it does demonstrate how modern gate oxides may be much more rad hard than older technologies.
Tantalum shielding
Tantalum capacitors over the tops and bottoms of chips can act as both shielding and bypass. Tantalum has a density of 16.7 gm/cm3 at 300K, so a 30 micron thick layer would add 0.05 grams/cm2 of shielding. According to SMAD figure 8-18, that might reduce the 1-15keV dose by a factor of 3-5, and perhaps a factor of 2 for electrons (figure 8-19). [ More accurate information needed] . To accommodate the extra thickness in the launch stack, the processors and memory and other chips might sit in interdigitated positions on every other server-sat.
MORE LATER
Volunteer Opportunity: Question - does a small tantalum cap short out when an ionizing particle passes through it? That would actually be good, because it could remove power from the segment of a chip that is being hit, possibly removing drift fields from the gate oxide and helping immobilize charges.
Space Junk and Debris
Server sky will be deployed in orbits higher than most space junk. The version 0.1 design deploys server-sats in m288 orbits, with semimajor axes of 12781 kilometers and average altitudes of 6411 kilometers. The vast majority of space debris is in lower orbits - it requires high launch velocities to even reach those altitudes.
Most of the relevant debris will be associated with geosynchronous transfer orbits. Upper stages will be in inclined orbits from higher latitude launch sites (28 degrees KSC, 46 degrees Baikonur, 20 degrees Wenchang), and 5 degrees (Kourou). Molniya orbits are lower, bit are less common. Before they decay, neither pass through the equatorial plane at m288 altitudes.
Even for debris with apogees that cross the equatorial plane at m288, the area of that plane crossing is vast, 15 million square kilometers for a server sky region 200km across. A given debris particle intercepts on the order of square meters or square centimeters per pass, and crosses through only once every two hours or so. If its orbit decays in 5 years, it will make less than a thousand passes before its apogee falls below m288. For example, a 10m2 booster stage, passing through 1000 times, the chance that it will hit any particular server sat (without avoidance) is on the order of 100 parts per billion. Booster stages are easy to track by radar, so they will almost surely be avoided by server sat maneuvering. Smaller debris will be more numerous, and have higher surface-to-mass ratios, so those particles will decay faster. While any given debris particle may have a very small total intercept area through the m288 zone, there may be tens of thousands, and they will be harder to track, so this probably represents a larger problem.
Note that very small particles will punch a hole through the server-sat, but probably won't shatter it. This may disable some portion of the server-sat (probably a segment of solar cell), but with a little redundancy the server-sat will still be maneuverable and capable of reduced activity.
Volunteer Opportunities
Compute the actual flux of particles that intercept the server-sky orbit.
Compute the actual orbital elements of spent booster stages in decaying geosynchronous transfer orbits from various launch centers.
Some small fraction of this orbiting debris may have perigees low enough to suffer appreciable drag in the extreme upper atmosphere. This will have the effect of lowering their apogees down into the server-sky orbit, to potentially collide with server sats. A problem with such decaying orbits is that their orbital elements become somewhat unpredictable - the apogee location will depend on the orientation and drag of the debris as it passes through the atmosphere, so it will be difficult to predict more than half an orbit in advance. However, there will still be perhaps an hour of tracking time between the most recent perigee and the passage through m288, and at 1 micron per second squared acceleration (5 percent of maximum), that is enough time for a server sat to move 2 or 3 meters out of the way of a precisely predicted orbital track. The key point here is precise prediction - to reduce the chances of a collision to zero, the radar systems will need to be very good.
The loss of one server sat through a collision is not a big problem in itself - there are plenty more where that came from. The biggest risk is the debris resulting from the collision, which will tend to be in the same orbits as the rest of the server-sats, and may pose a long-term collision hazard for those. The effect of solar wind on these uncontrolled bits of secondary collision debris is unknown.
In the long run, it will be better to remove space debris before it collides with the server sats. The server-sats are travelling faster than the debris they are likely to encounter, so any collisions between them and debris at apogee will most likely increase the orbital energy, raise the perigee, and lengthen the decay time of the debris. To remove debris (lower it's perigee into thick atmosphere) it must lose apogee velocity, either by colliding with material in a suborbital ballistic, material with an opposite inclination, or with material in a retrograde orbit. The important thing is to make sure that the small particulate debris from such collisions is captured. This is an unsolved problem.
Surface charging
Most of the area of a server-sat will be solar cells, with low voltages across the surface. If the LCD thrusters have glass covers with Indium oxide conductors on both sides, the outer conductors may be grounded relative to the rest of the server-sat. As all surfaces will be conductive, and at voltages less than one or two volts, there is no opportunity for arcing across surfaces, which sometimes happens to geosynchronous satellites in the solar wind ( see SMAD page 212-214 ).
MORE LATER
Ionizing radiation, charge upsets and latchup
MORE LATER
Drag
MORE LATER
References
http://www.isde.vanderbilt.edu/content/muri_2008/dixit_muri2008.pdf Sriram Dixit et. al. at Vanderbilt University. Recent work on HfO/SiO2 stacked gates and radiation resistance.
