Home
Programming
Chemistry
Physics/Electronics
Images
Go back
Introduction
If any of this is seen as wrong by anyone, please contanct me as it will help others and stop me
from making mistakes that could cost time, money, and mislead others as well.
I now have aquired some large capacitors. I ordered eight 350V ones off of
Ebay from a ctr plus seller, and those should give me a base to work with. I would have been much
better off had a gotten that lot I saw for 450V ones as the whole thing with shipping would have only
cost 15 dollars. While the voltage might be a little low, I am getting the leftover capacitors someone
else had from their railgun, and although they are of all different sizes, they are nice 430V ones.
I also have a lone 450V @ 2500uF capacitor, so I have an odd assortment to get me going. I think my
total power is around 7kJ if I used it all. The trick will be how to arrange the capacitors in the
most efficient fashion.
As far as my understanding of the rail gun, this is how I now see how things fit together.
Basic Schematic
Armature
The armature is what pushes the projectile forward. Often in amature designs, the armature and the
projectile are the same thing. The projectile is often made of aluminum, brass, or carbon.
Aluminum and carbon seem to be much more popular. While aluminum has good conductivity, it highly
damages the rails. Therefore, carbon is a much better projectile for amature design as you will have
longer rail life.
Capacitor Related
Capacitors(power source)
The capacitors power the project. Two styles of capacitor arrangement seem common. Most people use
capacitors around 450V and stack them up in series to get several thousand volts(maybe four to eight
in series) to reach the high voltages necessary to overcome rail resistance, however small it might be.
Multiple arrangements like this are put in parallel to create a powerful capacitor bank. Other people
have acccess to large pulse capacitors. These giants cost a couple hundred dollars each used, however,
so are hard to come by for amatures. They seem to be around the 10kV range and the capacitance is 50uF
on the largest ones but usually much lower. These are just put in parallel to increase the
capacitance. The storage capacity in joules in found by the equation C=1/2 * F * V^2, where C is
the storage capacity, F is in farads, and v is in volts. Since voltage is squared, higher
voltage but less capacitance can mean more actual storage capacity.
Capacitor charging
The charging power supply for the capacitors is usually home built. A step down transformer is wired
in reverse to step up the voltage from the main line or, more commonly, a microwave oven transformer
is used. It is then fed into a rectifier which converts AC to DC. Finally, a resistor is used to
limit how much current is put through it. I have a large amount of transformers, so it shouldn't be
a problem to find one. I also have some very large power resistors so I think I'll have everything
pretty secured except for the diodes. I might have some SCRs coming, so hopefully this will be under
control. I also have many large transistors laying laying around which I could
use as power diodes.
The switching is accomplished by IGBTs. These seem to effectivly be a type of high power transistor.
They need to have a very high current, specifically pulsed current(needs to be checked in the
datasheet, will use 60Hz increments), and high voltage ratings in the kilo volts.
Bleeder Resistors
These are high wattage, medium to high ohm resistors across the capacitors. They serve two purposes.
The first is to discharge the capacitors for long term storage. The many Joules of stored energy
are quite deadly and need to have a safety measure to dispose of them. The second is that they help
even out charging. In the even that one capacitor charges faster than another, it helps to make sure
the less charged capacitor can continue to charge by bypassing the first. I have some 50 watt, 4kOhm
resistors that I hope work for this. I haven't yet done calculations to determine if this resistance
value is too low. There are two major disadvantages of having the resistors. The first is that they
waste power during charging. They will not only put resistance across your charger, but also will
discharge the capacitors. Since they are discharging the capacitors, all that energy has to go
somewhere, right? You have to make sure you can safely dissipate the amount of heat generated on these.
Diodes
The diodes are not technically necessary but you could be sorry if you don't use them. When a magnetic
field collapses, it generates a large voltage spike(of oppisite voltage of what was put in?).
This feeds back on your equipment and what really comes into danger are the capacitors. This can
go from a simple ruining of all of your capacitors to causing an explosion, sending shrapnal and
fluid all over your work space. Diodes need to also be able to handle massive amounts of current for
short durations. I have heard that often people underprotect their capacitors and end up losing
their diodes, some even as large as fists. This spike generated from the rail gun is combatted
usually using SCRs which are effectivly high power diodes. Multiple SCRs, it seems like 12 or
so good quality ones where common, are put in parallel to handle this large spike. This ensures
long life of the SCRs/IGBTs and protection of your capacitors. An additional important factor to watch
is how quickly these switch on. If they can not switch on very quickly to handle the large surge,
the first part of it could hit your capacitors and cause extensive damage before even the main surge
hits. I plan on using a mixture of fast diodes and regular diodes to minimize cost. I'm hoping that
I'll get the fast diodes to get the opening pulse, while the regular diodes can take the majority of it.
Switch
There are several techniques for switching a rail gun on. Each is described below.
Explosive Thrown
I've never seen any hobbyist use these. Forget it. Expensive and one time use. Not even sure
if they are good, but I suppose so.
Insolated Gate Bi-polar Transistor(IGBT)
Basically high power transisors. Often have protection diodes built
in, but you'd still bewise in protecting them with additional diodes. Sensitive
to electro static discharge(ESD). I am considering using some of these and have a couple. 200A and
400A amp at 600V or 1200V are common rating for higher power "brick" packages. There are also
transistor sized ones, but they can only hold up to around 40A and I think lower voltages. These
can be switched on or off at will. Current only in one direction.
Silicon Controlled Recifier(SCR, aka thyristor)
The main advantage of these is that they turn on faster than IGBTs and thus can (less expensivly)
carry more power. However, once turned on, they must go below a certain voltage to turn off. Their
normal use is in AC circuits, so this normally isn't a problem since the 0V region will be crossed
many times a second. For rail gun purposes, it generally doesn't matter that they don't turn off.
Armature Switch
The most cost effective: in theory, free. The idea is that you somehow inject the projectile onto
the rails, thus closing a switch. This is often done pneumatically. However, there are some major
disadvantages. The primary is that the projectile will initially make poor contact with the rails,
thus cause damaging arching on the rails. The second is that this can cause the projectile to be
forced in the wrong direction. The Lorentz force pushes the slug to try to maximize the size
of the current loop. If the loop is best enlarged in the direction from which the slug is coming
from will slow it down.
Inductor
Finally, an inductor is sometimes used Its purpose seems to be to shape the pulse generated by the
capacitors. If too much power is generated by the capacitors initailly you lose a lot of efficiunecy
as the projectile has not had time to speed up. By extending the pulse time, it becomes more efficient.
There doesn't seem to be any way to quicken the pulse except by lowering the resistance of the rail gun,
so people try to keep this as low as possible. The pulse shaping occurs because inductors have high
resistance to DC voltages, so it acts as a large resistor. Another way to limit power is to undercharge
the capacitor bank.
Advanced Techniques
Advanced Schematic
Field Augmentation
I am unfamiliar with this area, but would like to learn a lot more about it. This involves the use of
external magnets to increase the stength of the magnetic field which is suppose to make the process
more efficient. There are apparantly two types(parallel and series augmented), but I don't know much
about them.
Plasma Sabot
More advanced projectiles(especially military class modules I believe) use something called a sabot.
These are commonly used in tanks. A sabot is something that helps propel a projectile, but isn't
actually part of the projectile. In tanks, the sabot is a discarding sabot because it normally doesn't
acutlly travel all the way to its target. The most common sabot in a railgun is alluminum. An example
of the actual projectile is tungstan. The aluminum is placed in a finned fashion so that it can
quickly heat and turn to plasma. This forms an evenly conductive barrier and is relativly efficient
at pushing a slug along.
Injection(usually pneumatic)
Its more efficient to have the slug already moving when it enters the rail gun. The first is that it
helps prevent welding the slug to the rails. If you can at least clear the rails without power attached,
theres a good chance you'll be able to clear them with voltage across them. Another is that the faster
a projectile goes, the more efficient it will be.
Timing
If you are using an electronic switching device and the slug is being pneumatically injected, you will
need to use some sort of timing to turn on the electronics at the right time. Usually, this is achieved
through a laser sensor at the start of the rails. Appropriete timing code is used to make sure that the
slug will be well onto the rails before the switches trigger. You must not only consider the injection
speed of the projectile, but also the minumum distance you'd like it to be into the rails and the
delay of the turn on of the switches. Additionally, it is a good idea to put a second laser at the
end of the rails. This allows you to compute average projectile speed along the rails(seperation of
diodes / delta t). Or, you could put two sensors at the end of the rails to get a more accurate
reading. Finally, if you are using IGBTs, there is one final advantage to timing code. You can set a
timeout to check if the slug clears the rails in a certain time. It is pretty safe to assume, for
example, that the slug should spend MUCH less than a second in the rails. If it takes longer, it would
be to your advantage to switch of the power ASAP. If the projectile is going slow, its a pretty good
sign that its either already or is soon to weld itself to the rails. SCRs can not do this because
they have to discharge completly and have no way of shutting off.