The limit of my 6mm coil gun was the switch. It could only take so many amps and was tuned to the point it would need the switch replacing after repeated shots. This is because of arcing in the switch damaging the contacts. To avoid this you need a solid state switch. The problem with these is that they are VERY expensive.
While at University I was asking the departmental specialist on electromagnetics about cheap switching options when he asked me if I could build him something to demonstrate the principles of particle acceleration to his students. This opened doors for me as it gave me departmental backing to develop a coilgun.
My personal mentor arranged for one of his colleagues to give me two very big IGBTs. These switches really are game changers in terms of the jump in voltage and current they could safely handle. This changed this project from fun to serious(ly fun).
I spent quite some time doing lots of research and even more maths to work out what the optimal set up was to get the most out of these switches. I decided that in terms of efficiency, optimising the kinetic energy of the armature was the best choice. It turned out that the most energy is when a .50 caliber projectile is used roughly two inches long.
I have a coil gun simulator that produces the graphs you see here that make optimising the coil gun a lot quicker as you can see where you are wasting energy. Using this simulator I predicted that the projectile would travel at roughly 22.7m/s in reality it came out at 21.25m/s. Considering it is known that the simulator over estimates slightly I think this is pretty accurate at predicting how good your coil gun will be before you bother building it.
You may notice some photos of what appears to be a blunt ended armature stuck in one of those blue material display boards that you pin posters to. This is from when the guy who asked me to build it borrowed it to exhibit at the Crockroft institute stand at the Big Bang in Manchester. I told him only to charge it up to 200v not 400v, use two feet of high density foam at the end of the flight tube to catch the projectile and then clamp down a 3mm thick piece of steel plate to catch it if it managed to get through the foam. He ignored all of this, told the voltmeter wasn’t working when it turns out he had it in AC mode for measuring the capacitor bank voltage which didn’t bode well and then over charged it and shot it through all the foam which he hadn’t bothered to replace, past where the steel plate should have been, and then got it lodged into the display board at roughly head height of the watching children. I re-iterated my points about safety and he decided to brag about this event to the head of the department at Uni. So I got called into his office and told to put a key on it so that only I can use it and not lend it out again. I think this is fair. I made the mistake of assuming a professor at University who is an expert on electromagnetics doing research for a company that works primarily with particle accelerators would be intelligent and responsible.
The full explanation of how it works and how to build one is on my Instructables here: http://www.instructables.com/id/50-cal-Coil-Gun/