VC04 Platform structure, deployment and control
Depth of coverage
At its target depth of L3. What the depth levels mean.
What this layer covers
Primary structure, deployables, attitude control, station keeping, orbit maintenance, propellant
Where it ends
Ends where the payload begins; the aperture structure is shared with VC05
Deep area
This layer is studied in more depth than most. It belongs to the deep area In-orbit assembly and mass.
Platform structure and control: where it stands
The platform is the satellite's frame, with the systems that point it and hold it in orbit. The 1978 US design, in its GaAlAs variant, planned 64 argon-ion thrusters to point the satellite, a thruster system never built at that scale (CLM-0332). The company Rockwell budgeted station-keeping, holding the satellite in its orbit position, at about 8 per cent of mass in argon over 30 years (CLM-1075). A 1980 study by NASA's Jet Propulsion Laboratory (JPL) modelled moving a 2,000 t payload between orbits with electric thrusters in 160 to 213 days, using projected thruster performance (CLM-0700). Stated first natural frequencies exist, but no source held here gives a computed first mode, the lowest natural vibration, or mode shapes for a 1 km antenna (CLM-0907).
Scope. A solar power satellite is a very large frame. Grumman's 1978 data for the Boeing Phase I design give the whole satellite as 10.7 by 5.3 km (CLM-0995). This section covers that frame: how it is built, how it is kept pointed and in position, and how its parts climb to geostationary orbit (the orbit in which a satellite stays above one point on the equator). The transmitting antenna itself is covered in From direct current to beam.
What is settled. None of the sources below describes a built structure of this size, and none computes how a kilometre-scale antenna vibrates (CLM-0907). Nearly all the evidence is design work from 1975 to 1981.
For pointing, the 1978 US design, in its gallium-aluminium-arsenide (GaAlAs) solar-cell variant, planned 64 argon-ion thrusters in eight clusters (CLM-0332). Their specific impulse (a measure of propellant economy) was 13,000 s. No argon-ion system of this scale has been built or flown. Its thruster life figures, 5,000 hours of grid life and a 5-year mean time between failures, are assumptions (CLM-0332).
Rockwell's April 1980 budget for its own design put station-keeping (holding the satellite in its orbital position) at a velocity change of 339.0 m/s per year (CLM-1075). Over 30 years that is 7.98 per cent of the spacecraft's mass, as argon propellant. Sunlight pressure alone accounts for 6.65 per cent. The thrusters would draw 34 MW on average (CLM-1075).
For the climb to geostationary orbit, a 1980 paper from NASA's Jet Propulsion Laboratory (JPL) modelled a 2,000 t payload moved by electric thrusters on 150 MW of solar power (CLM-0700). Trip times were 160 to 213 days, depending on the thruster. The thruster performance was projected, not demonstrated. Magnetoplasmadynamic (MPD) thrusters, a high-power electric type and the option needing fewest thrusters (23), have still not flown at 6 MW (CLM-0700).
A 1981 NASA summary sizes the electric ferry to carry 4,000 t on about 515 t of propellant per round trip (CLM-1026). Dividing the propellant by the cargo gives about 0.13 kg of propellant for each kg delivered. Its 120 cm thruster exceeded any fully qualified thruster of the time, the largest of which was 30 cm. A chemical stage, in a workshop figure for a Boeing concept, would carry 400 t on 830 t of propellant and need 280 flights per satellite (CLM-1026).
The climb passes through the Van Allen belts (zones of trapped radiation around Earth), which damage solar cells. A peer-reviewed 2019 model gives a power loss of about 1 to 3 per cent in quiet conditions over a 200-day climb with 150 micrometre cover glass (the protective glass over each cell). With 100 micrometre glass the loss can reach 15 per cent within the first 100 days. The figures are for communications satellites, not space solar power structures (CLM-0676). A 2025 Chinese design uses its transport vehicle as a shield and replaces that vehicle's arrays between trips. It states no dose or shield mass (CLM-0688). A 1980 Boeing study found that without annealing (heating cells to repair radiation damage) the ferry's arrays would last no more than 3 trips (CLM-1032).
What is uncertain. The main question is how much the frame will bend, and whether beam control can cope. A 1975 study of a 1 km antenna found that assembly tolerance was the largest source of bending, about 1.5 m at the edge in the worst case. That error is fixed and does not move, so beam control would correct for it rather than track it over time (CLM-0954, CLM-0956). Heat bending could be kept under 1 arc-minute (a sixtieth of a degree) if the top and bottom of the truss stayed within 4 K of each other. The 4 K is a requirement the study set, not an achieved result (CLM-0954). Rockwell's 1981 requirements allowed a peak out-of-plane deflection of 12 to 48 cm for its antenna (CLM-0997). They also set a minimum first natural frequency (the lowest rate at which the structure sways freely) of 0.0016 Hz, so that standard control methods could hold its pointing to 0.05 degree (CLM-0995). These are requirements, not computed values.
This site's own calculation (medium confidence, because the bending comes from scaling laws rather than a structural model) takes a thermal bending range of 2.5 to 141 mm rms (root mean square, a typical size) for a 1 km, 40 m deep composite antenna (CLM-0956). The antenna aims by sampling a pilot signal, a beacon sent up from the ground. In that range, detectors that sample the pilot every 10 to 18 m would keep the beam within its error budget. At 18 m that holds up to about 400 mm rms of bending, and at 10 m up to about 1.3 m. The calculation uses the bending shapes of a simple unsupported beam, because no source held here gives stiffness data for CASSIOPeiA, a current helical design (CLM-0956).
The gaps are in the dynamics. No source held here gives a computed first mode (the lowest natural vibration) or mode shapes for a 1 km antenna structure. The wider literature on large deployable structures is likely to hold relevant analyses (CLM-0907). None of the calculations above models motion forced by gravity gradient, control or sudden temperature change at eclipse (CLM-0956). A 1978 estimate of heat bending in a 5.4 km module, whose text and plots disagree, is covered in Getting rid of the waste heat (CLM-0998). A January 1980 structures workshop said that optimism about the structures "would be based more on wishes and past success than on hard evidence" (CLM-1075).
At end of life, international guidelines call for a geostationary satellite to be raised by 235 km plus a term for the push of sunlight. In these records, arithmetic on that rule shows that the sunlight term is the larger for a structure lighter than about 8.5 kg per square metre of sunlit area, at the highest reflectivity the rule allows. No source held here gives that mass per square metre for any design, so the result is not a finding about one (CLM-0751).
What is missing. The computed modes and forced-motion model above. A structural-dynamics study of the CASSIOPeiA design, or a measured deployment test, could fill much of that gap.
Read next. Phase coherence across a flexible array. Threats to a GEO solar power satellite. From direct current to beam. Getting rid of the waste heat. Launch and the climb to orbit. Running it, wearing out, and retiring it.
Records
59 of the 73 records tagged to this layer are published. The rest are not yet published. A record is published only after a second reader has checked it and it has passed every publication check. What the gate checks.
| Record | In plain language | Tier | Confidence |
|---|---|---|---|
| CLM-0235 | A 2020 UK study rated the UK behind other countries in many technologies for a solar power satellite. For three key ones, large mirrors, space solar panels and power beaming, it named the UK's lack of manufacturing experience or capacity as the main barrier. It rated closing these three gaps by 2050 as medium to very high difficulty, depending on the technology. | T4 | HIGH |
| CLM-0331 | Rockwell International's 1978 design study estimated the satellite would need about 93 tonnes of thruster fuel resupplied every year, a very small fraction of its roughly 30,000-tonne mass. | T3 | MED |
| CLM-0332 | The 1978 design's steering system used 64 ion thrusters running on argon gas, grouped into eight clusters, to keep the satellite pointed correctly. | T3 | MED |
| CLM-0333 | In one 1978 design, structure and mechanisms made up about 18% of the satellite's dry mass; a 2010 paper gives about 20% as a general rule for the main structure of aerospace systems. The two figures measure different things, so they are not two estimates of one share. | T5 | MED |
| CLM-0338 | The 1978 design assumed each steering thruster would fail on average once every five years, and that its grids would wear out after about 5,000 hours (roughly seven months) of running. It carried 64 thrusters so that spares could cover failures between yearly maintenance visits. | T3 | MED |
| CLM-0441 | A normal geostationary satellite needs a small, regular fuel-based nudge every year to stay in place. One 2019 design proposes cutting most of this fuel need for a solar power satellite by placing it in an orbit tilted by about 8 degrees that resists being pushed off station, and keeps the satellite pointing the right way passively using its symmetric shape and sunlight pressure. | T2 | MED |
| CLM-0539 | A UK consultancy, working for the European Space Agency and using the same design as its reference point, actually published a full weight breakdown -- something the original 2019 paper never did. It comes out lower than the headline claim, at about 0.70 kilowatts per kilogram rather than 1, even though it leaves out several categories of weight altogether. It also rates one of the satellite's power-conversion stages as barely more than an idea on paper, and its input notes say the lightest case for the radio equipment is an expert's judgement, not something anyone has measured or built. | T4 | MED |
| CLM-0672 | A 2020 paper by the builders of NASA's NEXT-C ion engine and NASA engineers gives the engine's ground-test performance; NASA's 2024 space-solar-power cost report names this engine but gives no performance figures for it. This is a real, tested engine, not a hypothetical one. | T4 | HIGH |
| CLM-0673 | Electric thrusters get roughly seven to fourteen times more push per kilogram of fuel than ordinary rocket fuel does, based on real, published engine performance figures. | T4 | MED |
| CLM-0676 | A peer-reviewed study calculates that solar panels lose a real, measurable amount of power, up to about 15% in the worst case tested, just from spending months passing through Earth's radiation belts during a slow electric-propulsion trip to high orbit. Thinner protective glass on the cells makes this worse; starting from a higher point in the transfer orbit makes it better. | T2 | HIGH |
| CLM-0682 | A 2013 design paper, written years before and independently of NASA's 2024 study, states almost the same speed-change requirement for reaching high orbit that NASA later published - two unconnected sources agreeing on the same physical number. | T5 | MED |
| CLM-0688 | The Chinese design shields the satellite parts from radiation during the climb to high orbit, using the tug itself as the shield. It also plans to replace the tug's own solar panels in low orbit, along with refuelling it, so that the tug can be reused. | T2 | HIGH |
| CLM-0700 | A 1980 JPL paper already worked out how to move a two-thousand-tonne payload from low orbit to high orbit using electric thrusters. It needed 150 megawatts of electrical power and took between 160 and 213 days depending on which thruster was used. The magnetoplasmadynamic option needed only 23 thrusters against 920 to 2,000 for the ion options. | T3 | HIGH |
| CLM-0701 | The 1980 study assumes each kilogram being moved has 75 watts of electrical power pushing it. One real electric satellite manages between about 2 and 9 watts per kilogram of its whole launch weight. The study's figure is 8 to 33 times that, but it counts only the cargo, so on a like-for-like count the gap would be smaller. That assumption, more than the thruster type, is why its travel times look so short for such a huge payload. | T0 | HIGH |
| CLM-0714 | The 2021 UK study by Frazer-Nash Consultancy proposed that the satellite fly itself to its final orbit, using its own solar panels to run its own electric thrusters. Chemical tugs would take the parts only partway, to about 5,900 km, just above the inner radiation belt, where the satellite would be built. The study said the plan needed further work and gave no figures for the self-powered trip. | T4 | HIGH |
| CLM-0751 | Working the international disposal formula through by hand shows that for a very large and very light satellite the extra height it must climb at retirement is dominated not by the fixed 235 km allowance but by the push of sunlight on it, and that for anything lighter than about 8.5 kilograms per square metre facing the sun that sunlight term is the larger of the two. | T0 | HIGH |
| CLM-0885 | A normal communications satellite spends about 50 metres per second of velocity change a year holding its slot. Working backwards from the 1978 design's own fuel budget, that design would spend about 360 a year against the push of sunlight on its very large area - seven times a normal satellite's entire budget - because it is vast and light. | T0 | HIGH |
| CLM-0907 | No source held here gives a computed model of how a kilometre-scale structure of this kind bends: its modes and their shapes. Two separate calculations here both stop at the same point for want of it. Until somebody models that structure, both answers stay as ratios rather than numbers. | T0 | MED |
| CLM-0954 | The 1975 designers worked out how much their kilometre-wide antenna would bend, precisely so they could decide how good the beam-steering correction had to be. The biggest bend came from building it slightly wrong in the first place - roughly three quarters of a metre to a metre and a half at the edge, depending on how the stated angle is read - and that does not move. Heat bending could be held to a few centimetres if the top and bottom of the truss were kept within four degrees of each other. | T3 | HIGH |
| CLM-0956 | If the antenna structure bends by the amounts the 1975 design allowed for - millimetres to a few centimetres from heating - then sensors every 10 to 18 metres, which is the size of the building blocks it was made from, are far more than enough to keep the beam within its error budget. They would only become too sparse if the bending reached tens of centimetres. | T0 | MED |
| CLM-0995 | The whole ten-kilometre satellite was designed to sway once every fourteen minutes, and the antenna was required to be at least as stiff as once every ten minutes so that ordinary control could keep it pointed to a twentieth of a degree. | T3 | HIGH |
| CLM-0996 | The beams the reference satellite would be built from are specified down to the wall thickness, and two contractors give the stiffness numbers. | T3 | HIGH |
| CLM-0997 | The designers said how flat the antenna had to stay: within about half a metre overall, with the surface elements out of true by no more than a twelfth of a degree, under sunlight, gravity and a 24 degree temperature difference across the frame. | T3 | HIGH |
| CLM-0998 | With the top of a five-kilometre section of the satellite two hundred degrees Fahrenheit hotter than the bottom, the estimate has its free end bending by about eight or about eighty metres during construction. The report's text says eight metres and its own chart says about eighty; the report does not say which is right. | T3 | MED |
| CLM-1026 | The electric ferry takes half a year to climb to geostationary orbit but burns about an eighth of a kilogram of argon for every kilogram it carries; a chemical stage would burn two kilograms of propellant per kilogram and need almost three hundred trips per satellite. That ratio, not speed, is why the programme chose electric. | T3 | HIGH |
| CLM-1032 | If the ferry's solar cells could not be repaired in flight it would wear out in three trips and the cost per kilogram would rise by about a third; thicker glass would fix that. A ten-minute delay restarting the engines every orbit cost more than the heat of being near Earth, and a chemical ferry would cost twice as much. | T3 | MED |
| CLM-1075 | Keeping the satellite on station against sunlight pressure was reckoned at eight per cent of its mass in argon over thirty years and 34 megawatts of its power; the antenna had to stay flat to half a metre across nearly two kilometres; and the review panel said confidence in all this rested on hope rather than evidence. | T3 | HIGH |
| CLM-1076 | In 1980 the electronics that would condition the satellite's power weighed, at the state of the art, more than the whole satellite was allowed to; the design needed them fifty times lighter, and the arrays to run at fifty times the voltage anyone had flown. | T3 | HIGH |
| CLM-1106 | The satellite's dark side charges to thousands of volts in the geostationary plasma, and the plastic backing of the solar cells would hold a few thousand volts across itself, within a factor of two of breaking down; the study's fix was a more conductive backing or a bleed path. | T3 | MED |
| CLM-1175 | The design would control its orientation without fuel. Small mirror panels near its front carry a coating whose reflection can be changed electrically. Changing how much sunlight each panel reflects changes the push of sunlight on it, which can damp the satellite's wobble and control it about all three axes. A Japanese satellite, IKAROS, used the same kind of coating. | T2 | MED |
| CLM-1177 | The flat layers that carry the antennas must be stacked at least a quarter of a wavelength apart, but not so far apart that neighbouring antennas are more than about six-tenths of a wavelength away, or the beam grows unwanted side beams. The layers are very thin, less than a twentieth of the gap between them. | T3 | MED |
| CLM-1178 | The antenna is one twisted surface, like a ribbon turned half a turn from top to bottom. Looking down its axis you see a complete circle; from one side it looks like an hourglass, and it presents the same area to the Earth whichever way it faces. | T3 | MED |
| CLM-1180 | The most powerful version adds curved mirrors at both ends of the satellite. Each end's pair of mirrors squeezes the sunlight to twice its normal strength, and because the light comes in from both ends, each layer gets four times normal sunlight in total. | T2 | MED |
| CLM-1181 | A student study built with the designer's input describes a small version, about 260 metres across and as tall as it is wide, for 24 megawatts. The surface is upright at the centre and leans over towards the rim, to about a third of a right angle. | T5 | MED |
| CLM-1182 | In this study the huge twisted surface is tiled with small six-sided panels, each about 0.6 to 1.1 metres across. Each panel is a little staircase of thin strips a few centimetres apart, with the antennas standing between the steps and solar cells on both faces of each strip, except near the centre where the strips would shade each other. | T5 | MED |
| CLM-1185 | The top-power version uses two pairs of curved mirrors, one pair at each end. Each pair doubles the sunlight and turns it sideways into the stack, so the stack gets four times normal sunlight in total. The patent gives the best mirror tilt for each pair's own concentration. It works this out for a pair that concentrates in two directions: about 16.5 degrees for a doubling, and about 11 degrees for a single pair that quadruples the light. The four-times system itself is built from a simpler kind of pair that concentrates in one direction only, and for that kind the patent gives mirrors tilted at 22.5 degrees. | T3 | MED |
| CLM-1186 | The patent describes an existing kind of solar collector whose layers step up like a staircase, with risers that can be panels that shed heat; it does not say its own helical design has such risers. In the helical design the rows are typically 14 millimetres apart, and the whole thing may be hundreds to thousands of metres across and tall, so it has many thousands of rows. | T3 | MED |
| CLM-1187 | The middle of the collector, a tenth of its area seen end-on, has no solar cells, so the mirrors have a hole in the middle rather than reflecting light that would be wasted. The hole's radius is about a third of the mirror's radius. | T3 | MED |
| CLM-1189 | A 1990 NASA study of powering a Moon base from orbit put the satellites in circles over the Moon's equator, beaming down to a flat receiver at an equatorial base that can see a wide cone of sky. The higher the orbit, the fewer satellites it takes to keep one always in view: 30 at 90 km, 4 at about 2,500 km, 3 at about 10,600 km. | T3 | HIGH |
| CLM-1190 | The study's reference design uses four satellites about 2,500 km above the Moon, each about the size of a space station (148 by 58 m): a 58 m square microwave antenna between two solar wings. Between them they beam a steady 1 MW to the base at 100 GHz, with batteries on board for when a satellite is in the Moon's shadow. | T3 | HIGH |
| CLM-1192 | With 1990 technology the orbiting system came out heavier per kilowatt than solar panels on the ground with enough storage for the lunar night. With the technology the study projected, the four-satellite reference came to about 330 kg per kilowatt, below the roughly 500 projected for surface solar but far above the roughly 20 for a large nuclear plant. | T3 | HIGH |
| CLM-1194 | The company's first planned project is sized at 200 MW to the grid from a 732-tonne satellite, running nine-tenths of the time over a 20-year life; a later project is 40 times bigger. | T4 | MED |
| CLM-1195 | The design builds one thin, nearly kilometre-wide disc out of many metre-sized flat satellites, each with solar cells on one side and radio transmitters on the other; the first project's disc is 967 m across. | T4 | MED |
| CLM-1196 | A separate, very light mirror flies beside each array and turns slowly, once a year, so the solar cells always see the Sun; for the first project it is about twice the area of the array, but the design does not count it as concentrating the light. | T4 | MED |
| CLM-1201 | Caltech's design has no single large structure. Many separate square craft, each 60 metres across, fly side by side in geostationary orbit and act together as one antenna 1.6 kilometres wide. Each craft is launched rolled up into a drum about 2 metres wide and 1.5 metres tall, and unrolls in orbit. | T2 | MED |
| CLM-1217 | In a 2023 talk, the designer said four satellites in a stretched 8-hour orbit could keep three receiving sites supplied day and night all year. The slide gives the figures but not how they were worked out. | T5 | LOW |
| CLM-1218 | SPS-ALPHA's design catches sunlight with a few thousand units of large, thin plastic-film mirrors. The mirrors are grouped into several hundred assemblies, and each assembly points on its own, like the mirrors around a solar tower plant, but the frame they sit on stays still, so the design needs no big moving joint. | T3 | MED |
| CLM-1219 | SPS-ALPHA's mirrors concentrate sunlight about three times on the array. The report estimates that five times would save almost 15,000 tons, if the array could stand the extra heat. | T3 | LOW |
| CLM-1248 | By December 2025 Japan's space roadmap said hardware for a large-structure test had been handed over for a supply-ship flight, and a satellite to beam power from low orbit to the ground had been built. | T3 | HIGH |
| CLM-1274 | NASA said its programme had built a test solar panel that it expected to lead to arrays more than three times lighter per watt than the best of the day, kept its power beams to about a tenth of midday sunlight for safety, and found that the very high voltages assumed in 1979 would not work in space. | T5 | MED |
| CLM-1321 | The 2012 design study for SPS-ALPHA put the weight of its 2 gigawatt satellite at about 35,000 tonnes, about half of it in the transmitter. | T3 | MED |
| CLM-1407 | Most designs turn their solar panels to face the Sun and their antenna to face the Earth, and pass the power across a turning joint between the two. CASSIOPeiA is meant to fix its panels and its antenna together and steer the beam electronically, so it needs no such joint. | T3 | MED |
| CLM-1432 | Rockwell's chart puts a price in mass on a flatter antenna. The frame for the 1,000 m radius weighs about 85 tonnes if it may deviate by 48 cm and about 226 tonnes if only 12 cm is allowed. The 500 m frame goes from about 16 to about 30 tonnes over the same range. | T3 | MED |
| CLM-1438 | The ring of geostationary orbit lies over the Earth's equator. That ring is tilted by about 23.4 degrees to the flat plane in which the Earth travels round the Sun. So for most of the year the Sun is above or below the ring, and it lines up with the ring only around the two equinoxes. | T3 | MED |
| CLM-1441 | Documents give the size of the CASSIOPeiA satellite design as anything from 1 to 2 km across. For a full-size version delivering 2 gigawatts from geostationary orbit, the best supported figure is about 1.7 km, from a 2021 study for the UK government. A 2020 report gives 1.6 km and credits the designer's 2019 paper, but that paper gives 1.6 km only as a general minimum for any such antenna at one radio frequency, not as this design's size. None of these is a measured size: the satellite has not been built. | T4 | MED |
| CLM-1442 | A 2021 study for the UK government shows one design, CASSIOPeiA, as about 1.7 km across and 2,000 tonnes, sending 2 gigawatts to the grid from 35,786 km up. Its scale drawing puts it beside the world's tallest building, the 830 m Burj Khalifa, and the satellite is about twice that height. It is a design, not a built satellite. | T4 | MED |
| CLM-1458 | The 2019 paper behind CASSIOPeiA's 1 kW/kg figure lists six modelled design variants, and only the two with the most concentrated sunlight reach that figure; the other four come in at 0.79 to 0.95 kW/kg. The paper does not show where the mass goes, and its statement that the design's operating principle was independently validated cites nothing and does not say what the check covered. | T2 | MED |
| CLM-1459 | A 2013 design study for a large constellation of solar power satellites proposes that each one climb to its final orbit on electric thrusters powered by a separate, smaller set of solar panels rather than its main panels, which stay folded during the long trip to limit exposure to space weather and debris. The study expects the separate panels to be damaged on the way and reuses them afterwards to help keep the satellite pointed. | T5 | LOW |
| CLM-1468 | The basic idea, a solar power satellite climbing to its final orbit on electric engines run by solar panels it carries, with no separate fuel tanker, is real: it has flown on communications satellites, though not at the scale of a power station. A 2013 design does it with a separate, smaller set of panels. The stronger version, where the satellite's own main panels run the engines, has been proposed, for example by a 1977 US study, which found the panels used for the trip would be badly damaged on the way, and by a 2021 UK study for the last part of the climb only, after chemical stages carry the parts past the inner radiation belt. None of these designs has flown. | T0 | MED |