The preliminary design of a direct current distribution system is presented. The framework for the design is based upon NASA’s report, "A Fresh Look at Space Solar Power: New Architecture, Concepts and Technologies," written by John Mankins of NASA’s Advanced Projects Office. The object of the solar power satellite, as stated in the report, is to "generate solar power in space for wireless transmission to receivers on the ground." Power requirements are determined. Transmission of this power from the "SolarDisc" solar collectors using High Temperature Superconductors (HTS) to Antennae Disc magnetrons is discussed along with circuit protection and slip ring elements. Costs and weights are included along with suggestions for future equipment development and design.


This preliminary design report’s goal is to arrive at a first-order cost and weight estimates based upon currently available commercial products. Where such products require modification to be made suitable for the project it will be so noted. All products mentioned are available today unless otherwise noted. The preliminary distribution system is of straightforward design. Power is collected from groups of solar panels connected to provide the voltage and power ratings required and connected to circuit breakers in the field of the "SolarDisc" (See Figure 1). These circuit breakers are connected to circuit breakers at the hub via 1460 one-thousand amp HTS. Hub circuit breakers are connected to a slip ring assembly at the hub of the "SolarDisc." (Figure 2)

From this slip ring assembly 1460 HTS are routed through the dark side fork of the "SolarDisc" to a slip ring assembly 3000 meters away at the junction of forks and Antennae Disc. From here power is distributed via circuit breakers to control boxes and converters placed throughout the dark side of the Antennae Disc and thence to the magnetrons.

Design details and comments follow under the headings of Power Collection and Management and Platform Support Systems. R.F. Power Generation and Transmission do not fall under Module 1.13 and therefore is not a part of this report.


Power Collection and Management

To find total power requirements a 30-kilowatt magnetron was selected

with built-in phasing capabilities and an efficiency of 71%. Because the Antennae Disc

must radiate seven gigawatts to meet design requirements, 233,333 magnetrons are required. Load calculations are based upon this number. Quantity of devices is based upon 235,000 magnetrons. This allows for loss of output from a few magnetrons during operations.

Magnetron anode voltage is 16.5 kilovolts, current required is 2.7 amperes. Total anode current for 233,333 magnetrons is 630,000 amps. It was assumed that the solar panels could be connected so as to obtain this voltage. A photovoltaic manufacturer confirmed that it could be done with no harm to the panels (1).

Magnetron cathode heater power was 60 amps at 11 volts. To reduce wire size a converter was selected with an input of 270 volts and output of 28 volts. It can supply eleven magnetrons at once when the heater voltage is changed to 28 volts and the current dropped to 23.6 amps. The manufacturer was asked if such a heater voltage change could be adapted and they said yes.

A characteristic of these magnetrons is that the heater voltage must be removed after about three minutes. If satellite transit time through an equinox is seventy-two minutes, then there are twenty-four groups of 9723 magnetrons to be energized in that time. Thus only one group’s heaters are energized and total heater current required, from the input side of the converters, is now about 30,000 amps at 270 volts. This 270-volt system was selected because it alleviates the voltage drop problem, reduces wiring weight and considerable aerospace equipment is available operating at this voltage. Potential future heater improvements may allow elimination of this high current for cathode heating and increase tube life significantly as tube life is determined by cathode heater life. Current estimated heater life is 20,000 hours.

Magnetron anode high voltage relays were used as the writer was unaware of suitable high voltage solid state controls. Latching relays are used to reduce total power requirements and to eliminate relay coil heating and any subsequent cooling requirements. These magnetrons, converters and relays are not space qualified and must go through a qualifying program.

Cooling the magnetrons requires removing about three million kilowatts. This could be accomplished using radiator panels mounted on the forks with an appropriate cooling fluid suitable for use in a hard vacuum (2, 3). By redesign of the magnetrons, cooling could be accomplished using pyrographite radiators (4).

Platform Support Systems

Many assumptions were made as little data was available regarding specifics. It was assumed that a Supervisory Control and Data Acquisition (SCADA) system would be installed. It would incorporate autonomous control artificial intelligence program running on dual computers at the "SolarDisc" and the Antennae Disc. Communication between the computers would be by radio, line-of-sight light beams or by fibre optics. All control boxes incorporate computers communicating with SCADA, SCADA transmitting operational conditions to some remote central control center and operational controls to the control boxes. Much of the communication between control boxes and SCADA could be by light beams. SCADA must handle more than one million input/output lines and operate on 270 volts. The current design estimates the control system will require about 20,000 amps. Another 50,000 can be used for lighting, battery charging and unknown future loads. One unknown but expected motor loads is that required to rotate the Antennae Disc and "SolarDisc." No attempt was made to investigate or design an SCADA system or determine drive motor requirements.

Heater controls are non latching space-qualified relays. They are on for only three minutes at a time. Solid state controls could replace these relays in future design iteration.

Specific Equipment Comments

Two major concerns are the high current circuit breakers and slip ring assemblies. Each is discussed below.

The two pole one thousand amp circuit breakers are a new design. The two pole units’ dimensions are 216 x 279 x 127 mm and weigh about 2 Kg. The unit is not space qualified. Under short circuit conditions, photovoltaic panels develop up to 120% of their rated output (5). Under these conditions it will be imperative that solar panel outputs are configured to meet breaker-rated current and voltage to reduce stress on the breakers.

Magnetron life must be improved to avoid magnetron replacement costs. The improvement in cathode heaters mentioned above is essential to reduce these costs.

The slip ring assembly(Figure 3) is composed of two opposing structural plates. The inner plate is stationary and is fastened to the "SolarDisc." The other, the outer plate, is keyed to slide freely along a 254-mm diameter shaft but cannot rotate about the shaft. Bearings on the "SolarDisc" support and allow the "SolarDisc" to rotate about this shaft. Shaft size was arbitrarily selected and is not necessarily correct.

Mounted on the inner plate are four, flat, concentric bands mounted on ceramic standoff insulators. These bands are 25.4 millimeter thick copper with a hard silver plating on the surfaces facing the outer plate. These are the non wearing surfaces of this sliding contact assembly. The outer diameter of the outer band is 1940 millimeters. Bands one and three will carry or transfer 100,000 amps at 270 volts. Bands two and four will carry 630,000 amps at 16.5 kilovolts. The area of the bands was calculated not on current density requirements but on the area required to connect the 1460 one thousand amp capacity high temperature superconductors. This equates to approximately 17 amps per square centimeter, 64 square centimeters per one thousand amps.

The opposing or outer plate supports an identical set of flat circular bands mounted on ceramic standoff insulators fastened to the moveable plate structure via a series of stainless steel springs. These springs allow the concentric bands to "float" in relation to the fixed "SolarDisc"bands and maintain contact pressure. The conductive bands are composed of 25.4 millimeter thick layer of silver plate on a 25.4 millimeter thick copper backing plate. Insulators are fastened to this backing plate. The outer assembly concentric rings with their heavy silver plate are the wearing rings, thus allowing replacement if it becomes necessary.

Calculations suggest this thick layer of silver should last about thirty years, even assuming one revolution per hour. Lubrication between mating concentric bands is provided by a conductive lubricant with a vapor pressure suitable for the vacuum of space. This is currently available. This slip ring assembly is not intended to make or break a "live" circuit. It is to provide a means of power transfer between a fixed and a rotating member.

The pancake style slip ring assembly is one of many attempts to design a sliding contact that can transfer extremely high power between a fixed and moving body. It is not a definitive design and must be subjected to rigorous review and analysis. It is a suggested design that overcomes the problems with other option of using many individual slip rings. However, it introduces problems in circuit protection. To overcome that problem the interconnecting high temperature superconductor cables between the "SolarDisc" and the Antennae Disc requires circuit breakers at each end of the cables.

A cylindrical slip ring assembly involves stacking 1460 rings 12.7 millimeters wide and approximately 173 centimeters in diameter with five centimeters between rings. The brushes would require replacement about every four years based upon the "SolarDisc" rotating once per hour. They also introduce the problem of removing about two megawatts of heat due to brush voltage drop.

The "SolarDisc" pancake style is calculated to dissipate about 420 kilowatts as will the Antennae Disc slip ring assembly. Thermal analysis must determine the best way to eliminate this waste energy as localized heating effect’s superconductivity.

Weights and Costs

Below is a table of estimated weights and costs. Weights are tabulated as to location as this influences future design. Weight and cost data on SCADA and other instrumentation as well detail wiring, shielding, enclosures, cable connectors, mounting hardware, cooling equipment, batteries, communication and computer equipment are not included. It should also be noted that no effort was made to reduce accumulation of static electricity nor was there any consideration given how this equipment was to be stored while waiting for assembly of the Solar Power Satellite.








Antennae Disc

Relays, Circuit Breakers, Converters
High Temperature



Slip Ring Assemblies
(Not calculated)
Cable & Wire
(Not calculated)
30 KW Magnetrons

(235,000 units)

Copper Cable 


Aluminum Bus Bar

instead of




(Not included in totals) 



(Not included in totals)



(Not included in totals)


*Weights do not include supporting structure

** No production pricing available. Costs shown are based upon an estimated price of $7 per kilogram that is approximately twice the current price for copper buses of the same ampacity. The price differential is the writer’s estimate based upon the reasoning that superconductors must be competitive with copper line, making allowances for elimination of line losses and lighter construction methods.



Without high temperature superconductors this project would evidently be extremely difficult to carry forward.  The use of HTS achieves significant weight savings, savings which would otherwise translate to significantly higher weight and cost. Conservative weight savings, as shown in the table, may be significantly enhanced as commercial installation and use of HTS improves with experience.
Fostering development of high current circuit breakers would reduce some circuit wiring with a potential for reducing weight and costs. Rotating sliding contact assemblies are unique in the amount of power being transferred. These assemblies require further study and analysis.


(1) Private telephone conversation with Bryan Spence at Able Engineering

(2) Loop Heat Pipes Heat Transfer Characteristics Using Nomograms. Borodkin, A. A., et al, SAE Paper 961566, International Conference on Environmental Systems, 26th, Monterey, CA, July 8-11, 1996.

(3) Emittance Measurements for a Thin Liquid Sheet Flow. Englehart, Amy N., et al, NASA-CR-204653, AIAA Paper 96-0237, Aerospace Science Meeting & Exhibit, Reno, NV, 15-18 Jan 1996.

(4) Experimental Radiation Cooled Magnetrons for Space Use, William C. Brown, published in Space Power magazine, Volume 11, Number 1, 1992

(5) Private e-mail from John Steven-


 High Temperature Superconductor Links

Note: Further information about current and commercial high temperature superconductors can be obtained at many websites:

November,  2000  Science News article    -

General HTS information                         -

The first commercial HTS installation is now being completed and is expected to enter service at Detroit Edison in early 2001.