A comment provided on my SA blog 17 asked several questions. This blog provides my responses.
=====
As the former NASA manager who ran the SSPS studies in the late 70s, a few words. SBSP will never get off the ground, to coin a phrase. Why? Here are a few reasons.
1. On the equinoxes SBSPs in GEO spend some time every day in the shadow of the Earth, no power transmission. Back up?
Response: Twice each year, for several days, satellites in GEO pass through the Earth’s shadow. The shadow is about 12,800 km across and the satellites are traveling at about 3 km per sec. The transit time has a maximum of about 1.2 hours. The continental U.S. spans about 50 degrees of longitude. At GEO, the length of the 50 degree arc is about 31,000 km. At maximum, the shadow will cover about 40 percent of this arc meaning that about 60 percent of the SSP satellites will still be producing power. The transit of the shadow happens at local midnight when consumer demand for power is reduced. This combined with the ability to transmit power through terrestrial power grids over long distances will enable the power reductions to be managed. Unlike terrestrial power losses due to droughts, storms, cloud cover, strikes, etc., these SSP down times are very predictable in terms of timing, length, and number of satellites impacted.
2. Using reasonable conversion factors, to deliver 1 GW will require a 10 square km satellite. A 5 GW SBSP would require 30-50,000 tons in orbit. Assuming a 100 ton payload for your “aircraft-like” system (~X a shuttle payload) the number of launches to LEO is very large, number of launch sites needed, many if one is planning 100s of SBSPs! Frequency of launches at each site, weekly, monthly?
Response:
a. My assumption is that the SSP satellites will be divided into two payload classes where 95 percent will fit within the payload bay of the fully-reusable, two-stage aerospaceplane and 5 percent will require a Saturn-V class partially-expandable launch system. The assumption used during the study was that the satellite mass would be 20,000 tons. This value is doubled to add mass for propellants, on-orbit assembly support, payload holding structure, etc. Assume 40,000 tons for each SSP satellite needs to be transported to LEO and 20,000 tons to GEO.
b. Assume each aerospaceplane transports four 10-ton modules. Each module will fit within a 40-foot intermodal container for ease of land or sea transport from the manufacturing site to the launch site. Each 40-foot container contains about 2,500 cubic feet. The 20,000 lb module has a density of about 8 lbs per cubic foot – about the maximum for general cargo. 1,000 missions are required to transport 40,000 tons to LEO for one SSP satellite.
c. Assume each aerospaceplane orbiter (second stage) is capable of flying once a week ,with two weeks in depot each year, on average. In other words, each second stage can fly about 50 missions per year. (This will be a second generation aerospaceplane designed specifically to transport the modules.) Assume that one SSP satellite per year is needed. 1000 missions / 50 missions per year per orbiter requires 40 orbiters. (The number of needed first stage systems, which turn around faster, will be somewhat less.) Assume 25 percent extra orbiters to cover those out of service for depot maintenance or repairs. 50 orbiters are needed to support one SSP satellite per year. 40 missions per week are flown, on average, or about 6 per day.
d. Assume two LEO space assembly locations, 180 degrees out of phase with each other, located in a logistics orbit. (A logistics orbit is a circular, constant altitude orbit that, by selection of the inclination and altitude, has a repeating ground track. This is not a microgravity orbit like the ISS or Hubble.) This provides two launch windows per day from the primary launch site and one from a suitably-located secondary site. This provides for 6 missions per day to be flown. The primary site would be at KSC and the secondary site would be on the coast of Texas. (Note: the aerospaceplanes may not launch from sites on land, but may use floating platforms that move offshore to minimize sound and enhance safety. For LOX/kerosene fueled systems, they would be fueled at the launch point from sea-anchored terminals similar to those used to off-load commercial tankers. This will enable the platforms to be moved to sea at a reasonable speed and enhance safety near the shore.)
e. SSP satellite production rate scale up would require a larger aerospaceplane fleet size, increased capacity at the terrestrial launch sites, and added LEO space assembly depots. At some point, shifting to extraterrestrial resources/manufacturing will be needed.
3. Then the LEO tonnage must be boosted to GEO for assembly, not in LEO. This will require additional thousands of tonnes of propellant and boosters in LEO. Are you counting the launches needed?
Response: The additional propellants are factored into the total tonnage discussed above. The current expectation is that conventional rocket-propelled space tugs will give the assembly modules an initial Delta-V and then the modules will use electric propulsion to complete the transfer to GEO. The space tugs are fully-reusable and based at the space assembly depots. Electric propulsion may be provided by a separate tug or this may be integrated into the assembly module using the solar arrays.
4. Power transmission will require a steerable antenna ~ 1 km in diameter containing millions of elements. You can get one from Radio Shack, right?
Response: Well, certainly not from just one Radio Shack store. 
Seriously, my focus is on the spacefaring logistics needed to support the assembly of the SSP satellites. The design of the antenna is being looked at by the appropriate experts. I believe that a phased array approach may be the preferred configuration. Either way, the antenna will be assembled in LEO at a special space dock and then moved to GEO. This will certainly be a challenge but we have about 15 years to work out the design, construction, assembly, and transport details.
5. Then there is the small problem of receiving the transmission and integrating it into the power grid. The rectenna size, including a buffer zone, would be ~ 15 square km. Lots of empty space in US, however not near the consumers. Solution, long transmission lines = power loss plus problem of right of way for transmission lines.
Response: Recent information that I have found indicated that transmission distances of several thousand miles can be performed with acceptable loses. Also, the U.S. currently has a highly integrated electrical power transmission system that will be increasingly used to move electricity from distant sources to the market near major urban areas. Whether we use SSP satellites or not, the U.S. electrical power generation system already had hundreds of GW class generation stations integrated into common power networks. The U.S. will need to add many new generation sites. I do not see any specific problems with these being SSP terrestrial receiving arrays.
6. Environmental impact? Many SPSBs in orbit will not be accepted by the astronomy community and others. I could go on. Suggest you obtain a copy of Don Rapp’s draft report to SMD “Assessment of Concepts for Utilizing Lunar Resources.” In the meantime, of course, there are much cheaper alternatives that require no R&D like renewables, nuclear power, and perhaps breeder reactors. I don’t place much hope on fusion power but potentially it may be much easier and cheaper than SPSB to finally make it cost effective.
Response:
a. I do not see the environmental impact issue as being significant. Modern astronomy is mostly digital which would imply to me that suitable computer controls can be used to eliminate any adverse impact from the SSP satellites. While the satellites are large, they will still occupy a very small fraction of the observable sky. The primary question is how do such considerations stack up against the nation’s need for new acceptable energy supplies. Also, the same spacefaring logistics infrastructure that enables SSP satellites to be built will enable large space telescopes to be built and serviced. By the time the SSP satellites become numerous, current large terrestrial telescopes will be 3-5 decades old.
b. I agree that extraterrestrial resources and manufacturing of SSP satellites are likely to be used in the future. Building SSP satellites from primarily terrestrial resources will constitute the first era of space industrial operations. Building them from extraterrestrial resources will constitute the second era.
c. In one of the other SA blog entries discussing SSP, I address the issue of magnitude of energy needed by the U.S. While there are other technological options for new energy production, it is not clear that they are suitable for new electrical base load power generation of the magnitude needed.
– Terrestrial solar and wind have the disadvantage that they are non-continuous and would require some form of substantial storage to enable continuous power generation under varying weather conditions. They may prove to be useful, under some circumstances, for offsetting CO2 production, but they do not provide a replacement for the coal-fired powered plants which would still be needed to provide power at night and/or when the wind velocity is low. Few people live in a climate where terrestrial solar and/or wind can meet there energy needs 365/24.
– Nuclear has issues associated with the cost and availability of uranium, the disposal of nuclear waste, the location of the nuclear power plants due to geography and cooling water availability, and the production of plutonium in some fuel cycles. The extent of the acceptance of a significant growth in the number of nuclear power plants is not yet clear.
d. While there are engineering issues to be solved with SSP, the general technology readiness level is substantially higher than any form of fusion at this time. This means that the engineering development of SSP can proceed while further fundamental research into fusion is needed. There have been, however, some interesting new fusion technology concepts. This may be the ultimate energy source of the future — how far in the future is not clear.
e. As the world economy grows, the demand for high quality energy – electricity – will increase significantly. For non-hydrocarbon energy production of base load electricity, only hydroelectric, geothermal, biomass, nuclear, and space solar power will be able to be used. SSP provides one renewable energy production system that can be readily exploitable as a turn-key process
Mike Snead