Magnetoplasmadynamic thrusters have the advantage that they can be scaled up to produce large amounts of thrust, while still maintaining the high ISP of ion drives:
Magnetoplasmadynamic Thrusters. "Testing for these thrusters has demonstrated exhaust velocities of 100,000 meters per second (over 200,000 mph) and thrust levels of 100 Newtons (22.5 pounds) at power levels of 1 megawatt. For perspective, this exhaust velocity will allow a spacecraft to travel roughly 11 times the top speed of the space shuttle (18,000 mph)." http://www.nasa.gov/centers/glenn/about/fs22grc.html
The problem is the high amount of power required. However high electrical power has been delivered up to hundreds of kilometers on Earth over power lines. Then this could be used to deliver the required electrical power to the thrusters from the ground.
The impetus for this was this proposal by Launchpoint Technologies to launch small satellites by magnetic fields:
However, there are many difficulties with getting large mass objects up to orbital velocity with EM fields alone, discussed in this thread on sci.astro:
If the launch system is to stay on the ground and for low mass payloads you can just as well use reaction mass methods, i.e, rockets, at high ISP to get the craft up to orbit velocity at short distances. You wouldn't need to have hundreds of kilometers of cable extending into air trailing from the craft. You could have a cable lying on the ground and a short length of cable extending from the craft to the cable on the ground, say 10 to 100 meters long. Keep in mind, just as for the magnetic launch proposal, the main thing is getting that horizontal velocity component required for orbit. To get to the altitude for LEO is just a small proportion of extra velocity and energy of that required for orbital velocity. Note that for large launch systems such as the space shuttle a large amount of thrust is needed just to accelerate that huge mass of fuel that needs to be carried along. But when the exhaust velocity is much larger than the ending velocity, say 100,000 m/s compared to 8,000 m/s then by the rocket equation the mass of the fuel will be about the same small proportion to the mass of the rocket, 8/100. (The exhaust velocity being 100,000 m/s for this MPD thruster means the ISP, specific impulse, actually is a quite high 10,000 s.) The Launchpoint magnetic launch proposal only talked about launching small satellites, 10 kilograms or so. Only one of the NASA Glenn magnetoplasmadynamic (MPD) thrusters would be needed to accelerate a 10 kg mass to 1 g. Five of them could accelerate it to 5 g's at 5 MWatts power. However, I should say key for this proposal is the idea the MPD thrusters could be made lightweight. From the descriptions of the mode of operation, essentially only requiring two electrodes, I'm assuming this is the case. The images of them shown also suggest they would be small and light weight. Assuming that it is indeed the case the weight of the thrusters would stay low when the thrust is scaled up, this might be used to launch most satellites and also astronaut passengers. Most satellites are less than around 1,000 kg. A 1 Gwatt power plant assuming power to thrust scales up could accelerate this at 10 g's. Transportable gas turbine electric generators at the 100's of megawatts scale can be bought in the 10's of millions of dollars range. So 1 Gwatt total would cost in the range of 100's of millions of dollars. NASA documents give the human endurance level for acceleration according to duration, as described here:
At 9 g's it's about 3 minutes for astronauts lying down in acceleration seats. The formula for speed v attained at an acceleration a over distance d is v^2 = 2ad. So for v = 8,000 m/s and a = 10 g's = 100 m/s^2, d is 320 km. They would have to undergo this for t =v/a = 80 s. You could have the craft go in a circle at a smaller radius to reduce the scale of the distance covered by the cable on the ground, but this would result in a higher acceleration according to the formula a = v^2/ r. For a radial distances of a few km's you get accelerations at the 1,000's of g's scale, which would greatly reduce the payload and make it impossible for human passengers. However, for small satellites, a few kilos, it might be easier to use such small linear or radial distances of just a few kilometers.
For larger sized payloads and for manned craft you would probably want the craft to be at high altitude when it reached the high Mach numbers. Then you could have the hundreds of kilometers long, ground lying cable instead be raised in the air, reaching from the ground to the desired altitude. You would as before use a short cable say 10 to 100 meters long to connect the craft to this longer cable that is held aloft. Then a large helium balloon could keep the longer cable aloft if the cable were say 1 mm wide. But you might need a cable 1 cm wide or larger to carry sufficient current to power the craft. Possibly several helium balloons along its length would work to keep it aloft in this case. Another possibility would be to use separate plasma thrusters along its length to keep it aloft since raising a payload to a high altitude requires far less energy then getting it to orbital velocity. It might also work instead to have the long cable be hollow filled with helium to provide its own buoyancy. High altitude helium balloons are typically composed of mylar 20 microns thick, able to reach 30 km altitudes. A hollow aluminum cable with a thickness of 1 micron and a diameter or 100 meters would have the same cross-sectional area as a solid cable 1 cm wide. We already have cables up to 5 km altitude and moveover these carry electrical power:
What Is a Tether? "A single cable called the "tether" maintains the aerostat in its position above the launch point. The tether not only anchors the aerostat in flight, but through electrical conductors embedded in the cable provides power for the electronics payload and other airborne components. The tether incorporates a metallic braid within the tether jacket in order to safely conduct lightning currents to ground via the mooring system." ... "TCOM is the only successful producer of large, lightning-protected, fiber-optic power tethers capable of delivering from 1kW to 80 kW from reliable ground power to the aerostat and its payloads." http://www.tcomlp.com/aerostats_What_teth.html
This company proposes to lift windmills to high altitude, up to 45,000 ft, 15 km, to send electrical power to ground level over aluminum power cables:
Windmills in the Sky. A bold plan to tap the jet stream and boost our nation's energy supply. By Michael Behar Posted 11.21.2005 at 2:00 am. http://www.popsci.com/scitech/article/2005-11/windmills-sky
They propose using a 3 inch thick aluminum tether with a Vectran core for strength to hold these craft in place which will also serve to deliver the power to the ground, at up to 20 megawatts. They argue the rotors on the craft could be used to keep them aloft as well as generate the power. This then could even serve as a "free" power supply for the electrical-cable powered rocket. The VASIMR plasma rocket requires high power such as nuclear. Could it be made lightweight to be used for Earth launch if it did not have to carry its power supply but was only powered by long cables from the ground? Note that if the high altitude wind power techniques do work we might not need to have a tether to the ground. Just have all the power be wind generated at altitude and have the long power cables would be stretched out horizontally from the wind power station, held aloft by lighter than air balloons or thrusters.