Small-Scale Coilgun

[0111narwhalz]

I've been fascinated by electromagnetic slug-throwers for quite a while, but have only recently actually looked into the possibility of a small-scale, amatuer grade coilgun. However, the only sites I've looked at are concerned either with items the scale of the US Navy railgun or gigantic megawatt accelerators powered by various nuclear reactors.

Anyways, I'm looking for a set of maths tht would let me calculate a feasible coilgun from parameters such as target depth and material, projectile mass and cross section, et cetera.

Concrete Example:

Suppose I require a coilgun to fire a .3g needle of cross-sectional area 12.5mm^2 (barrel, it's a tapered needle; tip 3.1mm^2). at a Styrofoam target 5m away, and pierce it to a depth of 1cm. What kind of magnetic field strength would I need to achieve thin in a single-coil piece of .4m length? How many turns of common copper wire, with what wattage, would this take? What number of coils would be optimal in this length?

Edited August 14, 2015 by 0111narwhalz
Example now has some more reasonable numbers.

[spudman2]

>>BEGIN UNHELPFUL REPLY<<

I'd suppose the best way to start would be to figure out how much energy, and how fast the needle needs to be moving in order to achieve desired impact characteristics. You can use some of the kinematic equations to figure out barrel acceleration. Knowing final speed, length of acceleration, and initial speed (zero), you can solve for acceleration needed. I've never had a chance to study the equations that govern magnet-magnet interaction, but I will say that I think you need a magnetic projectile. But I hope I'm wrong.

[Tex_NL]

The projectile does not need to be magnetic. Even non-ferrous metals would work.

Any rapid change in a magnetic field will induce an eddy current in any conductor inside that field. This eddy current in turn induces its own magnetic field opposite to the original field. The opposed fields will repel each other, ejecting the conductor from the magnetic coil.

[Charzy]

The challenge is not in powering the thing, it's in programming a microcontroller to switch the coils (of which you should have several, separate) as the projectile passes through the barrel. You want each coil to be on when the projectile is before it and off after the projectile reaches it.

[SomeGuy12]

And it's not programming the microcontroller, it's designing the high power switches that can cut the flow of current instantly upon command from that microcontroller. You don't even need or want the microcontroller, this adds latency. You want a photocell sensor and a diode, located at the entrance to the magnet. You'd design electronics to delay an instant and then cut the power.

[0111narwhalz]

Why bother with the photodiode when you can math it out and use a controller? Ideally you'd have the same latency between "on" and "off" both ways (if not, compensate), and there's no real reason for the latency to be different between different coils. So the biggest issue is for the switches, as SomeGuy pointed out. They have to... switch... quickly.

But enough of this. I just want the maths at the moment. Once I have maths, I can generalise to different scales. Changing projectile mass/cross-section, material, et cetera. I also would like some sort of reference for different materials' resistance to being needled, or a relatively cheap but accurate force-meter.

Oh, and assume a ferrous steel needle.

[Shpaget]

Power MOSFETs are designed for high power and fast switching.

As for the original question, I don't think you'll be able to calculate that easily.

I'd be surprised if there was extensive data on styrofoam penetration. My guess is that you'll just have to build any coil gun (or any other ejector) and test it. Measure the muzzle velocity (and preferably impact velocity) and penetration depth. Then graph it with varying velocities. You probably won't get linear graph.

After you have that data for a range of velocities and calibers, you can go about designing the coilgun to suit your needs. Start with the energy required to achieve your goal (speed and mass of the projectile). That should give you an idea about the capacitors you need to use.

[0111narwhalz]

If I had a force-meter, I could take the measurements myself. Force times speed, or force-distance per time, equals energy, right? And energy is mass times speed, which gives us our impact velocity.

Edited August 7, 2015 by 0111narwhalz

[Shpaget]

Kinetic energy E_K= (m x v²)/2

Work W = F x s

Power P = W / t

However, force meter (dynamometer) won't be able to measure force of a projectile (that doesn't make sense). It's used for measuring continuous forces.

If measuring the speed of the projectile is not practical for you, you could measure impulse of force. If you hit a small piece of styrofoam that has more mass than the projectile, and is big enough so the projectile doesn't penetrate it completely, you could measure the speed of the styrofoam after the impact and calculate the speed of the projectile that way.

[0111narwhalz]

By "measure force" I meant "push the needle through at a steady speed" not "shoot the needle at the dynamometer." And thanks for the maths.

[SomeGuy12]

Why bother with the photodiode when you can math it out and use a controller?

Because empirical data trumps theory. Speaking as an engineer here, your premise is bad. You can't "math it out" - too many variables. Slight differences in the projectiles, air resistance, slight asymmetries in the magnets, heating of your power MOSFETs, etc etc etc. Much more reliable to actually detect the projectile using photocells and switch your power control switches based on this information.

I do take back what I said about microcontrollers - your sensor rig, you would want a photocell sensor associated with every magnet, but you can determine your real acceleration and real velocity by measuring the exact times it crosses 3 of the sensors in a row.

Anyways, the currents involved are enormous. I am uncertain what kind of power mosfets can even handle the load. And your power supply...yikes.

What kind of money are you looking to spend?

[0111narwhalz]

Enormous power for a 1g needle to stick into Styrofoam? I'd expect you need only a couple joules in the projectile for that, but I have no sense of relevance fore that number. As for the money, preferably less than USD50, but I might be able to scrounge up more. I'm looking for the absolute minimum cost, like PVC pipes & tinfoil capacitors kind of cost.

And point taken about the theory vs evidence.

Edited August 7, 2015 by 0111narwhalz

[Shpaget]

Yes, coil guns need to be capable of delivering very high power for a short time.

However, for this application, this MOSFET would probably be more than enough.

It is rated for 60 V and 30 A continuous current. Also if you look at absolute maximum ratings, you'll notice that pulsed current goes up to 128 A. That's more than 7 kW.

In any case, it's not the MOSFET that's the weak link here. Entire wiring needs to be capable of supporting that current, not continuously, obviously.

Also, looking over the internet, some use thyristors or IGBTs instead of MOSFETs.

[K^2]

Suppose I require a coilgun to fire a 1g needle of cross-sectional area .01mm^2 at a Styrofoam target 5m away, and pierce it to a depth of 1cm.

A steel needle with these parameters is 13m long, so it already pierces the styrofoam target to a depth of 8m from 5m away. Done.

Always check your numbers for being sensible. Don't just make them up. If it's important that needle is 1g, estimate dimensions from that. If you want a very thin needle, with that kind of cross section, then compute mass from that.

Next up, what exactly do you want to build? I assume you understand distinction between railgun and coilgun. There are still a whole bunch of different coilgun types. An induction coilgun, for example, is much better at firing rings than needles. It's also kind of easier to build, because it does not require switching the magnet off. That also makes all of the math easier. Though, you do have to understand a bit about how induction works.

If you don't want induction, then material of the needle becomes critical. A coilgun won't launch a thin aluminum needle no matter how strong the field is. (Ok, not quite true, but you won't be able to get THESE kinds of mag fields at home.) So you need to make sure your needle is ferromagnetic. Furthermore, you'll need to have coil(s) switch on and off as the needles moves through. Pulling the needle in as it moves towards the center, but not pulling on the needle as it tries to leave. This is a bit problematic, because of the induction in coils themselves. The maximum velocity you'll be able to impart on the needle, and therefore, its penetration capacity, will depend in large part on how well the coils are switched on/off. And that will factor into all of this math in a non-trivial way.

So I would strongly recommend to forego needles for a start and concentrate on accelerating a copper ring. It's a much simpler device with much simpler math.

[Bill Phil]

Kinetic energy E_K= (m x v²)/2

Work W = F x s

Power P = W / t

However, force meter (dynamometer) won't be able to measure force of a projectile (that doesn't make sense). It's used for measuring continuous forces.

If measuring the speed of the projectile is not practical for you, you could measure impulse of force. If you hit a small piece of styrofoam that has more mass than the projectile, and is big enough so the projectile doesn't penetrate it completely, you could measure the speed of the styrofoam after the impact and calculate the speed of the projectile that way.

What's "s" in the work equation? I know work is force times distance(like Potential Energy...) so is that it?

[Shpaget]

It is distance.

Throughout my education I was thought s denotes distance. Our countries' system may have different conventions. I've seen weirder stuff.

[K^2]

More precisely, s is used for distance along the path, to distinguish it from d, which is displacement. The distinction is crucial once you go to 3D. The correct expression for work becomes W = ∫ F·ds = ∫ F·v dt

[0111narwhalz]

Always check your numbers for being sensible. Don't just make them up. If it's important that needle is 1g, estimate dimensions from that. If you want a very thin needle, with that kind of cross section, then compute mass from that.

Ah, yes. I didn't have a scale of any accuracy at the time, so let's call it a sewing machine needle, about 3cm long (eyeballing). Point taken about the "sensible numbers" bit. I'll edit the OP example to reflect.

If you don't want induction, then material of the needle becomes critical. A coilgun won't launch a thin aluminum needle no matter how strong the field is. (Ok, not quite true, but you won't be able to get THESE kinds of mag fields at home.) So you need to make sure your needle is ferromagnetic. Furthermore, you'll need to have coil(s) switch on and off as the needles moves through. Pulling the needle in as it moves towards the center, but not pulling on the needle as it tries to leave. This is a bit problematic, because of the induction in coils themselves. The maximum velocity you'll be able to impart on the needle, and therefore, its penetration capacity, will depend in large part on how well the coils are switched on/off. And that will factor into all of this math in a non-trivial way.

I'm pretty certain the needles in question are a ferrous steel. Not sure the particular variety. They're held in a box by a magnet, so... Also, I already have the needles, but not the ring. And a relatively steady supply of needles (My mom is a textile artist). So, I'd prefer to make a more complex item to fire the projectiles that I already have, than make a less complex item to fire projectiles that I'd have to make.

Also, am I correct in assuming that I want a needle-firing coilgun's barrel to be nonferrous?

[QuesoExplosivo]

A lot of people on this thread mention using power MOSFETs. I would think that SCRs would be a better idea, given that they are capable of handling higher current loads, and that there is no need to stop the discharge once it starts.

[K^2]

Also, am I correct in assuming that I want a needle-firing coilgun's barrel to be nonferrous?

That's a tricky question. You definitely don't want one, long continuous chunk of ferous material. But you might want ferous rings in your coils. That can dramatically increase the mag field strenght, but it will also increase induction by the same factor. So if yo can build a sufficiently sophisticated switching system that counters induced EMF, you can get better output with ferous rings. But if you are simply switching coils on and off, induction will probably do more damage to final speed than what you get from stronger field. So if you are using a very simple switching system, then you should probably just have a plastic tube for a barrel, and have wire coils wound around it.

[AeroEngy]

Check out this guys website. It shows a decent amount of detail in his design, construction, and testing of an 8-stage coil gun.

[0111narwhalz]

The OP's example has been updated. These are numbers gained from measurements of the needles in question. I'm not sure how to measure the required projectile energy. Just to be clear, I need the energy of the projectile at impact, before I build the gun. I'm trying to get ball-park estimates of how powerful I need the gun to be.

[Shpaget]

At the range of 5m the energy on impact will be pretty much the same as at muzzle.

You measure it by testing. Get yourself to a height, take your needle and add some fins to stabilize it and drop it on the Styrofoam that you've placed flat on the floor. Measure the depth of penetration. If not enough, add some weight.

The energy will be:

E_P = mgh

mass in kg

g is 9,81 m/s²

h in meters

[Chiboko]

From my limited knowledge of coil guns you could theoretically use a single long coil wrapped around a ferrous barrel in in order to greatly simplify the design. This would unfortunately drop the efficiency through the floor and you'd have the problem of your gun also being a large electromagnet with all the issues that entails.

[SomeGuy12]

The OP's budget is too low to even consider fancy designs. He should stick with ball bearings and some kind of simple barrel like a plastic PVC tube. If he takes some courses or reads some electrical engineering textbooks, builds some simpler devices and gets them working, and gets a budget more like 5 grand, then he can consider doing things a better way.