Rail gun experimental results due to varying bore and arc materials, and varying the number of barrel turns

We have built rail guns in a range of square bore Configurations (5, 8, 16 and 25 mm) and tested them with one, two and three turn barrels. We have achieved 4 km/s with both the 5 mm and 8 mm barrels. We use these guns for routine hypervelocity impact testing, gun development, and SDI launch environment testing. We have fired a total of over 300 shots, accelerating both polycarbonate cubes and projectiles in sabots, in vacuum and at ambient pressure. Our capacitor power supply is capable of a maximum of 190 kJ at 5 KV. Our routine diagnostics consist of a Rogowski coil and integrator for the gun current, eight dB/dt coils along the barrel, and a breakwire velocity screen. When the velocity predicted from integrating the area under the measured
\int i^{2}
dt curve was compared to the measured velocity (using projectile mass and barrel L'), a consistent trend was observed. While there was considerable scatter in the data, the maximum velocity achieved always agreed with or exceeded the predicted velocity until a velocity of approximately 3 km/s was reached. Beyond this velocity, the maximum velocity consistently fell below the predicted velocity. Within experimental error, this trend appears to be independent of the inductance gradient of the barrel. The cause for this deviation from predicted behavior is unknown, but the arc armature is suspected as being the prime cause.