…That’s 64 “Megajoules,” to be double clear. Enough to power one of these.*
I’d like to avoid using nuclear power, if possible; and limiting the definition of “aircraft” to anything that actually flies, today (So two An-225’s joined together by a common wing in the middle won’t count).
*You keep close air support in your way, and I’ll keep it in mine.
What’s your repetition rate? You could do this with a portable generator and a large capacitor bank if you didn’t want to fire very often.
Ach, good point. Let me see…
Let’s say two rounds a minute? 3-6 would be great, if possible, but 2’ll work.
I suppose using a GE90 as an APU would be problematic? 
If you just want Megajoules (and not Megawatts) then the onboard APU with capacitors should do fine. You won’t get much rate-of-fire, but you could do something like run a very large gas turbine in the back of a KC-135 (using the huge onboard fuel stores), or strip the lasers out of Boeing’s Airborne Laser and replace it with your weapon of choice.
64 megajoules is just 15,296 Calories, so 28 and a half Big Macs (540 Cal each) would do the trick.
Expendable thermal batteries are often used in applications (like rockets or tactical missiles) where a large power demand is required in a short period of time. These are basically closed galvanic reaction cells in which the salts are isolated until a barrier is broken by an electrical impulse or squib, and then the battery produces a large amount of energy until the working salts are all neutralized. This often produces a more even amp supply than lead-acid or dry cell batteries, but most thermal batteries are use-once-and-dispose (or at least use-once-and-refurbish), and so are limited to the above military and rocket launch applications.
Another option, albeit one that is still more in the realm of speculation rather than practical application, are high energy metastable systems like excimers, stabilized amorphous or crystalline matricies (like solid ozone, cyclic O[sub]4[/sub] and O[sub]6[/sub], and C[sub]6[/sub] cubane) where the energy is stored in internal strains, and highly energetic neutral free radicals. You could make a fuel cell with an energy density ability two or three orders of magnitude greater than normal dry or liquid galvanic reaction cells. The trick is keeping the above systems stabilized. Something like metastable helium has a duration of around two hours before degredation via spin-orbit coupling in a vacuum; however, interaction with the walls of a real storage container or reactor vessel gives a lifetime on the order of a tenth of a second.
In any case, you’d probably need to also carry a bank of supercapacitors as described by Jurph to get the actually power supply that you need to operate this thing. That alone is a lot of mass (though the actual capacitors can be pretty compact). It’s also a pretty hazardous thing to carry around in an aircraft; burn one of those babies and you’ll need to be venting the fuselage quickly before the crew starts melting.
Stranger
You know, I was kind of afraid I might be using the wrong type of figures for the weapon in question, so I did a little more digging…
[quote=Military.com]
According to Dr. Amir Chaboki, the program manager for Electro-Magnetic Rail Guns at BAE Systems[…] firing the 64-megajoule weapon six times per minute would require 16 MW of power
[quote]
Another bit on that same article estimated a need for 6 million amps per shot, for what that’s worth.
Yow.
So, how bad of a deal-breaker is this? (Even assuming, as before, that I could do with a lower rate of fire?)
6 megaAmps is pretty bad, but sort of silly to talk about since it’s instantaneous current. Until you’ve specced the other parts of your power system (specifically the voltage and the duration of a single shot) you won’t be able to turn that current number into a requirement for your capacitors anyway since capacitance is tied to all three terms. The high current number does mean that you will need some very heavy cables. And even with heavy cables you might want to examine the need for cooling the cables so they don’t melt… or maybe make the cables annular and just run non-conductive coolant through them at high speed. I’m not an EE, though, so this is out of my league.
I am a rocket scientist, however, and heartily endorse Stranger’s idea of using chemical/thermal batteries.
At the amperages involved and the fact that it all has to be mobile and fit side the fuselage of a plane you might just need to go ahead and use superconducting cables; even with the cooling system required it’ll be a lot more compact and efficient, and you’ll probably want to use superconducting magnets anyway.
Stranger
I was once shown a piece of electrical machinery that had been driven by a turbine from an F-1 rocket engine. That’s the engine that powered the Saturn V. Wikipedia says the turbine was good for 55,000 hp (41 MW). I believe the intended application was some sort of airborne directed-energy weapon.
Well, just for the fun of it, I did some more digging, and found a university engineering report on the weapon, which quotes an “operating voltage of 30 kV.” That’s to accelerate a 20kg projectile to 2500 m/s over a barrel length of 10 meters.
Thermal batteries, superconductors…sounds like fun.
I’m just wondering if I can get it all under 35 tons.* Not to mention the issue of recoil.
*Hey, you know what they say…“When the last B-1 and B-2 are flown to the bone yard in the desert, the crews will be picked up by a B-52!”
A related question -
How much recoil does a railgun produce?
A lot. How much impulse depends on the exit speed and mass of the projectile and the length of the gun, but at suborbital velocities, the impulse for any practical length gun is huge, probably as much or more as the 16" guns on the Iowa-class battleships. You’re probably going to need some kind of recoil compensation and damping system to spread out the impulse to something manageable.
Stranger
Even superconductors have a limit to the amount of current you can force through them: Currents produce magnetic fields, and superconductors stop superconducting when the field gets too strong. For currents in the megaAmps, you might even be better off with good old fashioned ultrapure copper.