The spacecraft used to deploy the first cable will have some unique requirements. The spacecraft will have a large payload mass, require a specific set of mechanical actions, have a short lifetime, and minimal power, attitude control, and communications requirements. In the originally proposed scenario [Edwards, 2000] the initial spacecraft was launched to geosynchronous orbit in pieces and would autonomously assemble there. We have rethought this deployment plan and have decided to deploy the pieces to low-Earthorbit first, assemble the pieces there with the aid of astronauts and then send the entire system to geosynchronous (see Chapter 5: Deployment). The difference is we can construct one cable with its spacecraft and send that to low-Earth-orbit instead of four spacecraft with four cables that need to be spliced on-orbit. This will save spacecraft mass and reduce the construction risk dramatically. From our deployment calculations we have determined some basic overall mass constraints (table 3.1). The masses for the subsystems can be determined by comparison with existing spacecraft.
The systems that we will need include: power, communication, attitude control, structures, cable deployment mechanisms, command, thermal and propulsion.Power
The power required by the spacecraft is minimal. Some power will be required for attitude control and communications initially and for deployment control during most of the craft 's life. Batteries or small solar panels can supply power prior to deployment. Once deployment has begun there will be excess power generated by the deployment.
|Item||Initial SC Mass (kg)||Requirements on subsystem|
|Structures||2,000||Support cable mass during launch, act as spindle for deploying cable|
|Power||50||Provide power for the spacecraft|
|Attitude Control||50||Orient spacecraft relative to Nadir and other spacecraft, set spin rate|
|Command||15||Control deployment, cable combining process, and ascent to far end of cable|
|Communications||10||Housekeeping and status reports|
|Thermal||20||Keep electronics within operating range, control epoxy temperature|
|Other and contingency||850||Mechanical system for deployment control, mass on the cable end|
|Orbital correction prop. sys.||1,620|
|Orbital correction prop.||16,200|
|GEO insertion prop. sys.||23,300|
|GEO insertion prop.||107,000|
Communications with the craft will consist of a few commands and diagnostic downloads. Data rates of a few hundred bits per second should suffice. It is conceivable that several very wide-angle antennae may be sufficient.
The attitude control will consist of pointing the spacecraft in a nadir orientation to within roughly 20 degrees and stopping any rapid rotations. Once the cable has begun to deploy, the spacecraft will be pulled into a rigid nadir orientation. Additional attitude control will be needed on the cable end weight to orient it properly and for imparting the initial angular momentum.
In discussions with Composite Optics (a company with experience in composite structures and systems for space) a spacecraft with 10:1 payload to structure mass is currently possible. Since our spacecraft, once in LEO, will not need to survive launch stresses, its structure can be lighter for the same payload mass as compared to current systems. Future developments and the possible use of carbon nanotubes (Since they are required for the cable, we can assume that carbon nanotube composites will be developed prior to deployment of a space elevator) will push the payload mass to structure mass ratio even higher. Our spacecraft design will have a total mass (not including the spent propulsion systems) to structure mass ratio of 10:1.
The propulsion systems were discussed briefly in the deployment section. Depending on how the cryogenic liquids store on-orbit and the ability to transport these systems it may turn out the two systems will actually be one large, non-cryogenic liquid-propellant system. In figure 3.1 this large propellant system is shown with twelve 2.6 meter diameter tanks (5 are not visible in figure 3.1).
In our original space elevator system proposal [Edwards, 2000] both ends of the cable were deployed at the same time and then the spacecraft moved outward to the end of the cable to act as a counterweight. However, deploying both ends of the cable from a spacecraft sitting at geosynchronous has some complexities. After further examination we believe that deploying only the lower end of the cable will work best. In this method the end of the cable is pulled downward and the spacecraft will be maintained at its geosynchronous orbit (see Chapter 5: Deployment). Eventually the end of the cable reaches Earth and the spacecraft continues to deploy more cable and floats outward (see figure 3.2). In this scenario the deployment of the cable, from a mechanical standpoint, is straightforward and has no high tension loading on the deployment mechanism inside the spacecraft.
The mechanical system for deploying the cable will need to closely control the tension and speed of the deployment and insure that no tangling or twisting occurs. In figure 3.1 the cable is shown as a single, large-diameter core spool (3m x 2.75m O.D. x 1m I.D.), however, a longer spool (6m x 2m O.D. x 1m I.D.) or multiple spool design should be investigated to reduce the chance of cable damage on launch. With deployment speeds of 200 km/hr and our proposed spool size we are talking about the spool rotating at less than 1000 RPM. This rotation rate is comparable to that of wheels on automobiles, the linear velocity is about four times as fast as commercial spooling machines.
One additional consideration is the technique for beginning the deployment. There will be no forces initially pulling the cable out so propellant may be required on the cable end weight to initiate deployment.
On the end of the cable will be a small self-contained craft. This small craft has two purposes. First, this craft is to impart a small amount of angular momentum to the cable as it is initially deployed. Once this initial angular moment is imparted and the cable is deployed to a few hundred meters to kilometer length gravitational torques will keep the cable aligned. The second purpose of this craft is to transmit a beacon signal as the cable reaches the end of its deployment so the end of the cable can be found and retrieved on Earth.
No power is required to deploy the cable (except for the first few kilometers) but a considerable amount of power will be generated. Depending on the deployment rate, the power that we need to dissipate could be roughly 20-40 kW for one to two weeks. This mechanical power can be converted to electrical with a system (DC electric motor and controller) having a mass of 15 kg per 10 kW to be dissipated (see electric motor discussion in climber design section below). Options to dispose of the excess power include:
The climber will be designed similar to a spacecraft with some important differences. The mass, power, reliability and such are comparable to a spacecraft but the launch forces that the climber will be subjected to are minimal compared to that of a spacecraft during launch. However, unlike a spacecraft the climber will feel gravitational forces for most of its life. The climber will also have some unique mechanical requirements spacecraft generally do not encounter. The major components of the climber are the locomotion, cable deployment and power systems. There will be little 'thinking' to be done on the climber and minimal communications. The basic climber can be seen in figure 3.3.
From our deployment calculations we have determined a cable mass to spacecraft or climber mass for our proposed cable length and the mass capability of the initial cable. The cable to climber mass ratio is 0.87 and the total climber mass is 619 kg. This gives us 331 kg for the spacecraft and 288 kg for the cable. The cable we will be deploying on the first climber will be shorter (91,000 km vs. 117,000 km) and slightly stronger (9.7 kg vs. 8 kg) than we had originally proposed.
The primary job of the first 207 climbers will be to deploy cables as they climb and attach it to the existing cable. This will need to be done at high velocity (up to 200 km/hr) with very high reliability and in no case damage the existing cable. The tension on the cable being deployed will need to be controlled carefully to insure there is no breakage and the new segment is attached at a comparable tension to that on the existing cable. In our scenario we believe that these additional small cables should be added to the edge of the initial ribbon to widen it at least until the cable is roughly 30 centimeters wide. By widening the cable we reduce the likelihood of catastrophic meteor damage. Once the cable reaches a 30 centimeters width then additional cables should be used to thicken the ribbon.
A second aspect of the cable splicing is the epoxy and its application. The epoxy must have some adhesion immediately such that the deployed cable will remain in contact with the existing cable and the epoxy must also cure in the environment of space (vacuum, solar radiation, temperature variations, etc.).
The locomotion system for our climber must be designed around the operational constraints of our entire program. The basic performance specs that we require include:
The first six of these are pretty clear from a simple understanding of the space elevator. The last comes from the fact that our power beaming system will operate at a constant output. To best utilize the input power the climber's locomotion motor should run at a constant power. Since our load will be decreasing as the climber goes from a 1 g environment to zero g, a constant power implies that the speed of the motor will be constantly increasing during its ascent or a variable transmission is required.
Our motor study has come up with a motor design that fulfills all of the stated requirements. The motor would be based on permanent magnet brushless multipole technology to achieve a high efficiency with low mass. Cobalt-steel alloy and Neodynium-Iron-Boron magnets would be used along with a liquid cooling system and a two or three stage transmission. During most of the ascent these motors will run at greater than 96% efficiency and above 90% for most of the remainder. A 10kW motor of this design would have a mass of 14 kg, require 5 kg of control electronics and could be produced in quantity for under $9k. A 100kW motor of this design would have a mass of 105kg, require 20kg of control electronics and could be produced in quantity for under $50k.
The track and roller system to grab the cable must be designed to hold without damaging the cable. The frictional properties of carbon nanotubes are not known. They will need to be examined before the track part of the locomotion system can be designed. In considering our cable design it is important that the track system grab the small structures of our cable uniformly. This would imply that the track in contact with the cable must be uniform and deformable on the micron scale.
This part of the system will also need to have a braking system in case power is interrupted and also a method must be available to release the track from the cable externally in case there is a malfunction (see malfunctioning climbers below). Tests and experiments will be required to optimize this part of the locomotion system.
The climber half of the power transmission system consists of photovoltaic cells for receiving the incoming laser power and a power conditioning system. There are a couple options for the photovoltaic cells and the choice is dependent on the performance of the photovoltaic cells and available lasers (see laser beaming section). The most likely scenario is to have a 3 m diameter array of photovoltaic cells located on the bottom of the climber. An alternative suggestion has been made (Hal Bennett) that the arrays could be located on the side of the climber and the power beaming stations could be fairly distant from the anchor (example: Mojave Desert). There are positive and negative aspects to this possibility and it should be examined further.
For optimal use of the motors it is best if our power system outputs voltages greater than 2500V.
In addition to receiving and utilizing the power beamed to the climber the power system must also be capable of dissipating excess energy that will be generated once the climber passes geosynchronous orbit. Depending on the velocity maintained, it will be necessary to dissipate kilowatts of power for up to several weeks. Various methods for dealing with this situation are addressed earlier in this chapter.
One issue related to the climber that must be considered is thermal. If we consider a laser beaming system we will have solar cells where possibly 20% of the incident energy will be converted to heat. For our initial climber this is 17 kW. In addition we will generate 1 - 4 kW in the locomotion system. If the solar cells are isolated from the rest of the climber and exposed to space it is possible with our 3 m diameter array the panels will come into equilibrium at 400 K. (This assumes a blackbody emissivity for the panels and a 293 K ambient environment on the under side of the arrays.) This is too hot to run the photovoltaic arrays efficiently. If radiators of equal area to the arrays were added then the temperature would drop to 340 K which is much more reasonable for photovoltaic operation. This adds mass however. Since carbon composite structures are thermally conducting we may get the additional radiative cooling we need simply by having good thermal connections to our structures. This will have to be investigated further. It may also be possible to design the photovoltaic arrays to minimize the heat generated.
In addition there are thermal considerations in the motor design. Initially the motors will be operating at extremely low speeds which dramatically increases the heat production. This initial heat load affects the motor designs and can cause damage to the permanent magnets. To improve this situation we might consider a low-mass "tug" that takes the climber up the first few hundred meters until the climbers are easily pumped with the power beaming system and so the climbing motors would not need to start at zero velocity.
The support structures of the climber will require much less strength than that of a standard spacecraft since it will face no launch forces. The structure of the climber will be designed for a slowly varying load in the vertical direction with the primary structural loads existing between the cable and locomotion system. Design of the structures must consider thermal issues as well (see thermal discussion above).
The control system on the climbers will be required to monitor the speed of ascent, the tension in the cable, the splicing process and the climber location. During most of the climb the system will be in a slowly changing system with little complexity in the control required. Beyond geosynchronous, the climber will need to switch modes from a climbing mode to a braked descent. The final, and probably most important, responsibility of the climber controls is to stop the descent and lock the climber in place for use as a counterweight at the far end of the cable.
Communications like control will be minimal. The only communications that will probably be present in the climbers are low baud rate diagnostics and emergency contacts. If the climber stalls, if there is a loss of power beaming or a problem on-board then the climber should send a prompt communication.
|Epoxy and Cable Splicing||40|
|Photovoltaic Arrays (7 m2, 50 kW)||21|
|Motors (50 kW)||70|
For doing what we need to do we have 331 kg of climber mass. The communications, and control system will require minimal mass (15 kg each). The structures will have less constraints on them than on a spacecraft so we can use a value that is aggressive for a spacecraft (10:1 or 58 kg). Thermal control is important in our situation but also fairly difficult to estimate a mass for at this point. For the moment, we will assign 40 kg to thermal control. If we assume we will have epoxy of the mass of 1% of the cable mass and a system to apply it including rollers we get a value for the epoxy and splicing system of roughly 40 kg. The actual control system for the power may be 15 kg leaving us 123 kg for the locomotion and solar panels. A 50 kW climbing system (7 m2 of photovoltaic arrays and five 10 kW motors) requires about 96 kg (21 kg for the photovoltaic and 75 kg for the motors). The remaining 68 kg will be contingency at this point. If possible this contingency should be used for increasing the photovoltaic arrays and motors to handle 70W on the initial climber. A summary of the mass breakdown is in table 3.2.
Since we will be sending up hundreds of climbers of varying sizes it is critically important the climber is designed to be expandable. The design must allow for addition of motors, increasing the strength of the structures, larger cable spools, additional photovoltaic panels, higher heat loads, and higher power flow (see figure 3.4). The control and communications systems are the only ones that will not expand as the climbers grow.
In the unfortunate case where a climber becomes stuck during it ascent there must be a method for removing the climber. One fact that complicates removal of stuck climbers is that the cable will probably not be able to support two climbers both at low altitudes. However, there are methods to get around this difficulty.
One option if a climber becomes stuck at a low altitude is to pull the cable down until the climber is retrieved and then allow the cable to float back out to its nominal position. In the current design of the cable the fraction of breaking strength would be pushed from 0.5 in nominal operation to 0.6 if 3000 km of the cable were reeled in to retrieve the climber. The highest stress in this situation is at the far end of the cable; the rest of the cable would be at less than 0.6 of the breaking tension.
A second option presents itself above 2600 km where the downward acceleration on the climber is less than 0.5g. In this situation a second climber without payload could be sent up to release the malfunctioning climber and carry it beyond geosynchronous orbit where it could be released. Climbers that would be sent up to retrieve a stuck climber would not have the cable load and thus have a mass of less than 40% that of a full climber. All climbers would be equipped with a release that could be accessed by a climber coming up from below.
Between 2600 km and 3000 km either of these two options are viable.
In addition to the standard climbers and a rescue climber a repair climber may also be warranted. This repair climber would be sent up with short sections of cable that would be epoxied over weak sections. The climber would travel up the cable searching optically for weak sections in the cable and then place a cable patch on this section. The difficulty is that to do this efficiently the patch work would have to done at a speed of 100 to 200 km/hr.
Repair climbers could also re-apply the coating to protect against atomic oxygen erosion. This could be a metal deposition process or possibly a metal-impregnated paint. The repair climbers would also be less massive than the cable carrying climbers and thus may be able to be sent up in between the standard climbers with minimal schedule impact.
After deploying the initial cable it will need to be strengthened and widened to make it more usable and more resilient to its environment. As the strength of the cable increases the mass of the climber and the cable it carries increase. As each climber reaches the 0.1g point (97 hours after initiating ascent) a second, slightly larger climber can be attached and sent up the cable. At this rate a 20,000 kg capacity cable can be built in 2.3 years or a 1x106 kg in 5.1 years. The mechanical power utilized by the climber primarily depends on the size of photovoltaic arrays and motors that can be carried. It may be possible to improve our current design and allow for more power per kilogram of climber (currently it is 50 kW for 619 kg or 81W/kg). If we can improve our power to kilogram ratio by 40% (70kW for our initial climber or 113W/kg) then we can strengthen our cable to 20,000 kg capacity in 1.7 years (saving 8.5 months) or to 1x106 kg in 3.7 years (saving 18 months). In addition to getting a large cable on-line faster the quicker schedule also reduces the risk of cable damage.
|ç Chapter 2: Cable Design and Production||Table of Contents||Chapter 4: Power Beaming è|