Author Options:

High efficiency coilgun design Answered

Hello there I need some (a lot of) help with my project. I've started to design a high efficiency multistage coilgun. The main feature of the design is using the self-inductance formed in coils that were shut down for accelerating the following coils. To perform this, we can divide each coil into 3 segments (lets call them subcoils), connected with a wire and power all them 3 with a single capacitor. At first, the current from the capacitor is flowing through all 3 segments and a projectile is being pushed through the first segment. Than the first sensor is being activated with the projectile and it switches the power flow to only second and third subcolis As the current in the first subcoil is changed, the self-inductance directed in the opposite direction is appearing. In theory, we can use this power to increase the current in the following segments. The same thing is happening between second and third segments. And I want to place 3 of this stages (9 subcoils in total) I will use 8mm caliber projectiles, 40mm or 50mm length. I chose 5200mkF 450V or 6800mkF 400V capacitors, 2 of them in parallel for every stage (6 capacitors in total) Yes, I know it sounds way too powerful, but that's my actual goal. It's not my own idea. I have learned it from here http://gauss2k.narod.ru/adf/gs3seg.htm (русские вперёд). So, some problems had appeared. I can't figure out how to properly calculate the required inductance of the coils. In the source it is said that the inductance of the first subcoil should be equal to two of the second and the third subcoils (L1=2*L2=2*L3) to achieve the best performance. Also, my rough estimates show that peak current through the coil will be about 1,5-2,0kA for 3,0-4,5ms. This is quite a lot! I'll have to manage this energy and choose the power switches rightly. To sum up, I will be glad to hear your opinions about this idea. It's kind of controversial, but I hope it's not a certain failure.


The forums are retiring in 2021 and are now closed for new topics and comments.

1 year ago

I played with those concepts a bit when the topic gained popularity during the cold war times.
My russian is a bit rusty, so please correct me if I got it wrong: You want to build a railgun for a magnetic projectile - not a railgun that uses the discharge energy to create the required field with a lot of sparks and melting down the rail!?

If so than you are faced with basic problem of controlling not just high energy levels but also extreme magnetic field acting only to contactless move a projectile.
This basic idea is far better than the wasteful approach the military uses with their high energy discharge option but is far more complex to control and calculate.
So far I have been unable to find viable and working prototypes for this.
But I wouldn't be me if I wouldn't have tortured my brain some years ago to come up with other ideas ;)

Here are some of them...
Switching needs to be eliminated when reaching certain power levels as the requirements for the switching hardware will blow out even the highest budgets.
At least when thinking about conventional switching.
My theory might be a bit hard to follow and in a few ways contradictive to common science, so take it to think about it or just enjoy the read and move on ;)

A magnetic railgun needs to work more or less like an induction motor.
You want the magnetic field to act on the mass only where it is strongest and only for the amount of time until the positive force gets too low to be effective.
Imagine a relay that you want to stay half open or half closed.
Totally impossible until you consider the spring force and frequency.
Carefully timed impulses can keep the lever in the half way position....
To upscale something similar for a railgun you need to factor in harmonics and self resonance.
To stick with your three coil design:
Ideally the coils would overlap as the projectile can only be pulled towards the coil but will stop once it reached the center.
Doing this however means you already create a magnetic field in the next coil due to the field you create in the first coil.
Even worse for number three because the created fields would always be opposing.
Placing individual coils next to each other with magnetic shielding means you need fields strong enough to fully act on the projectile already when the projectile reached the center of the previous coil.
The last only works with heavy duty switching devices and insane losses.
However using the coils in a harmonic configuration that allows then to form a resonant circuit with the capacitors would eleminate both the switching needs and the corresponding losses.
Lets start your problem in reverse for a better understanding:
If your rail is 1.2m long with the coil section being 1m in lenght than you can calculate the acceleration required to reach a set speed within 1m.
Factoring in the mass of the projectile plus friction losses gives you a guesstimation of the required energy to accelearte your projectile.
With that data you can then calculate by how much each stage needs to accelerate the projectile to get the target speed, plus the time the projectile requires to travel from a fixed point to where the corresponding coil becomes useless.
This is "On time" that is required in a perfect world for that coil.
The real world however means you have to add some extra due to not breaking the laws of physics and other minor problems.
But, now getting back the the asyncronous motor:
Ignoring normal science you would now be able to get the rotational frequency for this imaginary motor.
And funny enough this frequency is the same (in terms of single impulses) your rail gun would need ;)
So the real key is to bring distance, mass and acceleration into relation with frequency - in a resonant way.
The energy you need for the projectile plus some extra is what the coils need to transfer to the projectile.
In a resonant system however this energy level can be far lower as there is no waste so to say.
Only problem that you never get the full power level right when switched on.
And any miss means the projectile won't even pretend to move.
A possible way out would be a discharge cascade configuration.
A capacitor is charged and once it reaches max voltage a spark gap, like on a tesla coil would discharge the energy into the coil and form a resonant circuit.
A secondary coil like used on simple flashlight for camera flashes (please look it up if not familiar how these flash tubes are operated) will react to the discharge.
Both the initial and polarity changes when resonating.
The energy from this secondary can then be used to provide a "kick" for the next stage.
Here the capacitor is already kept charged to just below the trigger point of the spark gap.
The extra from the first secondary will tip it over and fire the spark gap.
This continues in the same way for all following stages.
Coil distances need to be matched so that the first impulse from the initial actiavtion of the spark gap is exactly as long as it takes to get the projectile in reach for the next stage.
With enough stage you start to recover some of the energy from previous stages.
For example by splitting coils.
Each half is for a seperate stage so that the resonant second or third impulse will happen at the right time for the projectile.
Means for later stages you always half one half coil from a previous stage plus the stage for that point itself.
And if the timing and distances are correct than both magnetic fields will combine, resulting in far higher energy levels in the later stages without the need for much bigger wire diameters or capacitors.
The only real downside of this approach is the complexity and intense EMF produced.
Initial and quite costly tests with system similar to this were ended quite quickly in the 70' due to the problems caused by the high RF interference when firing.
It is like activating a powerful broadcast station that also eliminates the use of all electronics close enough.
However a 40m long prototype was claimed to get a 14kg projectile to an exit speed of close to Mach2...


Reply 1 year ago

In theory, it's a good idea
But in practice... In Russia we call it "spherical horse in a vacuum".
Technically we can precisely calculate all the timings, but it will be way to complex.
Switching the power isn't defined by the charge level of the capacitors but by the actual location of the bullet.
You will have to calculate, for example, the change in coil inductance due to the bullet moving through it, or the fractions of self-inductance current that will do useful work in the following coil. You can make estimates but can't make precise calculations to achieve such efficiency without using any sensors, that will give you actual location of the bullet