Value chain · CLM-1433
In plain terms
In the 1975 study, at 300 dollars a kilogram to launch and assemble, a 1 GW station cost about 1.7 to 3.5 times as much per kilowatt as a 20 GW one. The large stations were cheapest at about 1.25 to 2.25 GHz, but their curves are shallow: about 1,900 to 2,300 dollars per kilowatt across the whole band for the largest.
The claim as recorded
Raytheon's 1975 Figure 12-6 (transport and assembly at 300 $/kg, power source at 500 $/kW) plots direct capital cost against frequency from 1 to 5 GHz for 1, 2, 5, 10 and 20 GW of ground output. Graph-read to about +/- 50 $/kW (+/- 150 on the steep start of the 1 GW curve), cost per kW falls with plant size at every frequency where the lines are apart. The 1 GW curve falls from about 6,740 $/kW at 1 GHz to about 3,940 at 5 GHz; the 2 GW curve falls from about 4,070 at 1.05 GHz and is flat at 3,010 to 3,040 from 2.5 GHz on. The 5, 10 and 20 GW curves have shallow minima of about 2,240, 2,080 and 1,900 $/kW, near 2.0 to 2.25, 1.5 to 2.0 and 1.25 to 1.75 GHz, and rise to about 2,550, 2,410 and 2,290 at 5 GHz.
Evidence behind this: T3-the programme's own 1975 contractor trade study, held as a scanned plot; the costs are graph-read by us from the printed curves, not printed numbers.
Scope and limits
GRAPH-READ FROM A DEGRADED SCAN, NOT PRINTED NUMBERS. The printed lines are about 100 $/kW thick and each value is a line centre. Between about 3.5 and 4.3 GHz the 5, 10 and 20 GW lines merge, so the plot does not show which is which there; the readings at 4.5 and 5 GHz assume the three keep the order they have at 1 GHz (5 GW highest). The vertical axis starts at 1,000 $/kW, so a chart drawn from zero changes how large the gaps look. The power source mass reads "1 5" with no visible decimal point and is read as 1.5 kg/kW; it is not used in any value here. The minima and their frequencies are our reading of the plot; the study's own characterisation is "favorable in a broad range near 2 GHz" (CLM-1086).
What it assumes
Linear axes, fitted on the printed ticks; a scan rotation of about 0.008 rad taken out; each reading is the centre of a printed line about 100 $/kW thick.
Where it comes from
PDF p.95 / printed 12-14, zoom-5 crop. Caption: "Figure 12-6. SPS Capital Cost vs Frequency - 300 $/kg". Vertical axis "DIRECT CAPITAL COST" in $/kW, ticks 1000 to 7000 every 1000, the horizontal axis drawn at 1000, not zero; horizontal axis "FREQUENCY - GHz", ticks 1 to 5. Curves labelled at the left under "GROUND OUTPUT POWER": "1 GW", "2 GW", "5 GW", "10 GW", "20 GW". Legend, transcribed: transportation and assembly 300 $/kg; power source 500 $/kW and "1 5" kg/kW, the decimal point not visible; "MPTS", "BEAM TAPER 5 dB", "BEAM EFF 90%", "CONVERTER AMPLITRON", "MATERIAL ALUMINUM". Graph-read, $/kW at GHz (+/- 50 $/kW, +/- 0.03 GHz). 1 GW: 1.0 6740 (+/- 150), 1.05 6650 (+/- 150), 1.25 6100, 1.5 5390, 1.75 4920, 2.0 4600, 2.25 4410, 2.5 4290, 3.0 4130, 3.5 4060, 4.0 4010, 4.5 3980, 4.95 3940. 2 GW: 1.05 4070, 1.25 3830, 1.5 3550, 1.75 3340, 2.25 3080, 2.5 3030, 3.0 3010, 3.5 3030, 4.0 3040, 4.5 3040, 4.95 3030; the printed line has a gap near 2.0 GHz. 5 GW: 1.0 2620, 1.05 2600, 1.25 2490, 1.5 2370, 1.75 2290, 2.0 2250, 2.25 2240, 2.5 2270, 3.0 2360, 3.5 2410, 4.5 2490, 4.95 2550. 10 GW: 1.0 2160, 1.05 2150, 1.25 2120, 1.5 2090, 1.75 2080, 2.0 2080, 2.25 2110, 2.5 2150, 3.0 2230, 4.5 2380, 4.95 2410. 20 GW: 1.0 1930, 1.05 1920, 1.25 1910, 1.5 1900, 1.75 1900, 2.0 1930, 2.25 1960, 2.5 2010, 3.0 2110, 4.5 2260, 4.95 2290. Where lines touch: 10 and 20 GW together span about 2,130 to 2,340 at 3.5 GHz; 5, 10 and 20 GW together span about 2,180 to 2,490 at 4.0 GHz. PDF p.94 / printed 12-13, zoom-4 crop: the results "are given in Figures 12-6 and 12-7 for the low and medium cost combinations of power source and orbital operations, with microwave costs at the medium level. All costs are expressed in 1975 dollars."
Basis: 1975 dollars per kW of ground output; GHz; GW of ground output
Related records
- Answers question
- QST-017 not published in this release
Cite this
energyaperture.com/c/clm-1433
CLM-1433, Energy Aperture, as of 1975-12.
Connected records
Other published records linked to this one. A list, not a judgement.
- Also from SRC-451
- CLM-0954The 1975 designers worked out how much their kilometre-wide antenna would bend, precisely so they could decide how good the beam-steering...CLM-0959The original 1975 calculation found the satellite's radio noise would be fine for ordinary and troposcatter radio links and a problem only...CLM-0960The 1975 study found the satellite's harmonics - the multiples of its transmit frequency - would trouble radio astronomy at the second and...CLM-0961Everything in the 1975 interference analysis is for one satellite. The study itself says that a hundred of them would raise the noise and...CLM-0965In 1973 an astronaut on Skylab put together two long poles from short sections in just over two hours, about six minutes per join. A 1975...CLM-0966In 1975 the antenna alone was expected to take one to two years to assemble in orbit with about fifty people, and robot arms were judged...CLM-0968The 1975 antenna study worked out how hot the middle of the transmitter gets depending on how efficient the tubes are and how tightly they...CLM-0969If the tubes are only 70 per cent efficient, the transmitter's own face is not big enough to get rid of the heat they make, even spread...More on the source page.