Making a Coilgun - Part 3: Reconsidering Everything 

It’s so incredible to me that this guy was willing to challenge the concept of a coil gun instead of attempt to perfect the pre existing designs that have flaws we’ve all been avoiding. I always wondered why we have to use capacitors when their discharge is so strange for this sort of application, and that a constant high current would do much better. However, I never thought of changing the coil shape to a curve following a log equation, but after seeing his results and considering his reasoning, it looks infinitely better than most modern designs. Great problem solving man, this kind of work is truly fascinating!

12:54 .. Just imagine.. in a few years someone is on the battlefield with such a thing. .. and then it fires backwards .. whoops .. "Yeah it tends to do that." I laughed so hard.. :D

I think you made a few fundamental mistakes, though perhaps you corrected them in subsequent videos. From what I recall when I was working on a coil gun (I only ever simulated it using a custom solver, didn't build it), the force at the center (in the transverse direction) of the coil is zero. The reason being how the projectile is being magnetized and where the force comes from. So when you said that the projectile rests where the force is the strongest, that's incorrect. The projectile rests where the forces are in equilibrium. The force profile looks more like a sin^2 function, so you want to pulse the current before the projectile enters the coil and then remove the current before the projectile reaches the center. Using a 1/x profile is an interesting idea, for it would simplify the control circuitry, though it would still have a low efficiency. What I found (this was 10 years ago), is that the most efficient geometry is to essentially do what is done for super-electro-magnets. You want to have your coil be a big as possible in the radial direction, but a small as possible in the transverse direction (direct of the barrel). That allows you to maximize the magnetic field. Obviously though, there is an optimal size, based on the wire diameter and how much current you can run through it. There's a reason why super-electro-magnets have coils made of solid metal rather than wound wire. For control circuity, you probably don't want to use optical sensors at all. Instead, you want to effectively tune timings down to the microsecond using some computer control system (a MCU like a Arduino could reach that level of control). In simulation, the two stage controller with that I was working on was able to generate about 3T of magnetic field at peak discharge, and would accelerate a 5mm ball bearing up to about 2x the speed of sound (which I think would vaporize it due to the short barrel distance). Although, it's likely that there was an error in the simulation - I never built it due to safety concerns. Capacitors were used (as part of the simulation), along with shunt resistors, and fast switching IGBTs for control (it was a combined kinematic, electrical, and E&M solver). From the tumbnail of your follow up videos, it looks like you did something similar to what I described above (I haven't seen your next video before writing the above).

I see that Dr Pepper product placement

Keep it up, man! This channel is brilliant

Having experimented with coil and rail guns I can give you this tip that will help greatly. Wind each layer of each coil with a separate wire in the same direction and connect the all the positive ends of each layer together and same with the negative ends. Also aluminum plates at each end of each stage will help to isolate the stages and focus the field. An outer reflector can be added to each coil as well. It would be interesting incorporate these ideas into your design. Mine was a 2 stage with half coil length slugs and my coil triggers were actually at the front of each coil instead of at the end.

I think you're vastly over estimating the discharge rate of capacitors. At higher speeds (above 100 m/s), you're actually going to have the opposite problem. You wont have enough time for the current to build up and rise to it's maximum through the inductors (coils), before you have to switch it off abruptly when the projectile is in the middle to avoid suck back. Secondly, the currents you're trying to force through these coils are way, way higher than batteries can handle, which is why people use capacitors. More current, gives you more force, more acceleration and a higher exit velocity. I absolutely understand your reasoning behind the offset coil, but you won't get the constant acceleration you're expecting. It does give you a gradient which moves the projectile towards one end, but it is a less efficient/powerful way to do things. The low number of turns in the start results in a weak magnetic field there, and the magnetic field from the denser part of the coil at the other end is going to be weaker due to the inverse square law. Therefore you're better off with a more uniform, high density coils along the whole barrel that you commutate to move the projectile along, this makes full use of the length of the barrel to apply the maximum amount of force the whole way along. Secondly, the voltage and current rating of a battery pack are almost unrelated. Your high voltage battery pack will likely not have the current capacity you're expecting, there's a reason people use capacitors. Sorry, not trying to be negative, just wanted to correct some misconceptions before you sink too much time and money into some of these and then learn they wont work the way you expected. Coil guns are cool, and hard to make efficient/with high exit velocity, and I love seeing new approaches and ideas.

I will be submitting some work of my own before too much longer, but I will point out that coilguns can be made far more efficient by changing the nature of the projectile. A ferrous projectile suffers in two ways. First, it can saturate. Second, it can only make use of around half of the coil's length for acceleration, as any magnetic flux present in the coil after the projectile has passed the center will be a loss. You can bias the coil to help mitigate this to some degree, but even so, you will have to match the resonant LC frequency of the coil and capacitor to the acceleration profile of the projectile to get maximum efficiency, and nearly half of the cycle has to be managed as a loss. You can use an IGBT or similar to trim what can't be resonance matched - but you have to contend with the collapse of the magnetic field, as well. You can improve efficiency greatly by inserting a magnet into a solenoid. A magnet inside of a solenoid will move exclusively in one direction based on its orientation to the direction of winding and current flow. This means that, up until saturation, the magnet's response will be acceleration under any present magnetic field. But magnets are a poor projectile choice and quite expensive for the more powerful varieties. Cheap magnetized iron/steel could be used, but will easily saturate. There is something, however, that is cheap and will not saturate. A closed loop secondary - a shunted solenoid inside the solenoid. Configured as a sabot, nearly anything could be accelerated so long as there is a changing magnetic field. The full power cycle of an LC circuit can be used (provided the projectile is not accelerated out too quickly). There are two things to consider in this approach. First is that the speeds many want to achieve with a coilgun can be translated to a distance over time equation and optimal efficiency will be achieved when the LC resonant frequency completes its full oscillation within the time it takes the projectile to leave the coil. A full analysis would use a differential graphed equation to show how the increase in power in the coil translates to acceleration of the projectile - you can have a slightly longer cycle than what is implied by a length of final velocity over time - but it's a decent rule of thumb. This means that faster projectiles mean lower inductance, higher current coils with lower capacitance, higher voltage capacitors - and/or longer overall coils. However, the longer the coil and the higher the power, the more that coil is going to self-repel and the two ends are going to want to bend back onto each other. You'll want a nonferrous, nonconductive, rigid material to use as an insert or jacket to prevent coil distortion. These will also be significantly more powerful than any existing coilgun you see shown. 80% efficiencies have been recorded in airforce and navy research in using these as primary boost stages for orbital rockets and the like. Which - I found the claim curious when I read over it because every action should have an equal and opposite reaction capping electrical energy to acceleration at 50%... But then I consider the efficiency of electric motors and wonder if I am missing something about physics... Or if we are all missing something blatantly obvious and profound.

Very cool concept thanks for sharing

Just so you know I had a class member in school do this on a bit bigger scale over 40 years ago and fired things out of it quite well. If I remember correctly it was about 3 or 4 feet long and took a lot of insulated motor wire and I think the big end was around 4 inches in diameter. So I think you are on a good track.

I Think you forgot to consider the coils characteristics. When the capacitor has been discharged, somewhere in the beginning of the coil, the coil itself will have some energy stored in form of an electromagnetic field. This means that even though your capacitor has been almost discharged, there will still be an electromagnetic field pulling on the projectile. This is by the way also where you have your efficiency problems... when the capacitor is then fully discharged, the coil will try to hang on to its electromagnetic field, and does this with a large negative voltage spike, which is usually removed with some diodes. The diodes shorts the negative spikes, and waste all of the energy stored in the coils. There is no apparent way to solve this, besides not using coils.

This is something I was thinking about for a couple of years. I was thinking that any effect gained would be lost by the lower number of turns in each successive layer, though I suppose you would just increase the number of layers for any given loss of width. Also, I wonder if the length of your projectile should be at most the length of the final coil such that the last coil acts as a regular coilgun and all previous layers are just to accelerate the projectile to that "stage." Glad to see I don't have to work it out myself.

Amazing!! This is science, love it. You may need to patent some ideas soon.

You could also try super capacitors, which are more like batteries that can be charged super fast

Can we make the bullet more conductive to the magnetic field? In ac induction motors you put grooves in the rotor can a similsr solution be made for coilguns?

1800 gravities in the mass driver III at SSI in Princeton, NJ.

Nice work, butttt. A coil is a big inductor and it resists current flow so yes a capacitor wants to instantly discharge but an inductor had resistance to that. This is why good dc power supply’s use capacitors and inductors as it smooths out the dc voltage.

Hi bro , any updates? i am eagerly waiting for a new video for this Coilgun

You changed the way I now think about this

Good job ... I've been wiring up the same coil ... but for a lot bigger project ... as soon as I get my patent, I'll share the completed project.

I doubt this will be more efficient. Single stage coil guns have terrible efficiencies (1-2%), the only way to get a coilgun to be efficient is to use multiple stages, that way the degrading magnetic field caused when one coil is switched off is cancelled by an opposing field (in the correct direction) in the next coil. The magnetic coupling of coaxial coils with this geometry would be terrible.

I think you're wrong by not using capacitors as you can modify the time constant of the circuit as your driving current timing. Look into how both DC and AC motors work using sinusoidal distributed windings and you'd probably be better off applying some sort of decaying sine wave decaying at your desirable time constant in order to push and pull the projectile. You simply need to unwrap a 3 Phase AC motor and use a single projectile instead of a squirrel cage.

I agree; capacitors are not the way to go, they're much too slow. Imo, there are six options other than caps.

I need more of this

I feel like you were on to something with that design, but I also thought of one while watching this. using the standard coil design, what if when energized it pulls the projectile to the center, but instead of just cutting the power and dealing with that not instant decaying field, what if you then reversed the electrical flow in the coil? this would require a magnet of some kind on the back/front of the projectile I would think, but essentially have the coil both Pull and then push the projectile? that way you dont have the slow down/pull back from the decaying/collapsing magnetic field. I feel like some photo sensors and solid state switches could handle that power switching quite easily.

Very good video man, keep it up

I may actually have a go at experimenting with this design, I think I'll 3d print a barrel and toy around a bit

Wouldn't a phased pulse be highly efficient in this application? TKS. GODOG

I recommend doubling the length of the coil and rapping it one third every level instead of one half every level.

I literally thought about this yesterday, thanks for doing this work!

I have been wanting to make a coilgun that doesn't use alot of components that weigh it down, the cone design already is proven to be more effective (although I think it looks like a sci-fi gun), I think the cone coil plus a transformer or ballast with an AC power supply will give it enough power but as for the timing it would be a lot faster to use a momentary switch instead of a sensor. a sensor I don't think would shut off the coil fast enough although I've never worked with a sensor. one issue I have though is heat and my only solution for that is to use liquid helium or liquid hydrogen. also I love this design because it is the fibonacci sequence

At some point it seems like the inductance of your one coil will become a bottleneck. I wonder if using a much bigger inductor to store the energy as opposed to a capacitor would work. As in, charge up a bigger coil slowly to a decent current then shunt it over to having your coil in series. The bigger coil will want to remain constant current so will energize your firing coil. Basically a single shot of a boost converter. Then you should be able to scale up indefinitely because your staging and active coils can both be scaled up together and any charging circuitry won't be interacting with the active coil at all.. hopefully that made sense. A high current battery may not have the voltage to immediately bring your firing coil up to current but with the staging inductor boost effect that shouldn't be an issue for even very large coils.

As faster the projectile becomes the shorter and stronger the next impuls has to be. Otherwise it is like a break.

USE 2” aluminum spheres you can pull apart, line up nine of them in sequence with tungsten electrode, copper electrode and nickel chromium, and lamp wire. Feed it out your device put it all in a housing. That would me a phased array microwave plasma gun. Or a maser phaser!!!!

I actually had this idea myself but i never actually put it into practice. Might try it on a larger scale in the future, and now i see one of the issues i should put a steel plate in the back.

An induction coil might also help your project instead of capacitors. So long as you're using the inverse of shit, logs and fun maths you might as well start thinking with fun shit in mind. Go get a starter coil from an old big ass truck and see what you're new coil can do with that. The collapsing of the coil has the opposite geometry of a capacitor so I'd think that you might have more luck with one than you would with capacitors. You might also think about using a relay for both discharge and cutting power, if you get the thinking thought in it you don't need a sensor and can greatly reduce the complexity of your circuit.

What about using a high capacity tazer to power the coils on off really fast or for more power output use the Subaru flyback or other vehicle spark coil for on off freq?

I feel like you're on to something here. I think try a pancake coil for the last coil(punch the barrel through a piece of paper with glue on the paper, and add on to what you already have) to pull the center of the field more to the end. Also, when you did the sketchy battery test fire thing you had the battery connector facing the same direction that the gun was firing(, try to align the battery connector so that when you touch the coil lead to the battery lead, the recoil will pull the other coil lead out of the battery lead for an auto disconnect. I've never liked the cap design, direct power is the way to go.

What if you used two capacitors on each coil, would it help keep a steady high energy output rather than a spike and fall? The second capacitor should give it another high energy output right as the first capacitor drops in output. One capacitor to pull the object in and one capacitor to push the object out.

When you speak of the barrel is the projectile going to travel through a rifled barrel. It seems you understand the electromechanical ideas but to be accurate it needs to travel the length of the barrel with rifling to put a spin on it to make it accurate. If you want to replace the chemical process with and electromechanical process it needs to be fast, accurate and generate speeds of at 1000 ft per sec which to me is the bottom speed. Let me ask you a question have you ever fired a real rifle to understand the foot-pounds involved in just a 40-grain bullet.

Please explain if I am wrong, but at about timestamp 3:05 you draw a V vs T diagram of a capacitor discharge discharge over what you call "a CONSTANT load". Isn't your drawing the discharge over a resistive type load?? Shouldn't the discharge curve over a constant load usually linear? For instance, for a simple linear voltage regulator, don't you see a linear discharge at the input? I have done several computer simulations (in BASIC) of this kind of coil. I graphed the magnetic field for countless windings. I got the peak IIRC mag. field far beyond the coil center, but never got it past about 80% the coil start. Have you looked at the winding of what is called a (zeeman slower)?

imo, you are close but you do want a cap, the reason being. your coil design needs just a little more "layers" maybe bring each down 10%-25% rather than 50% of the previous and add back the cap but you will want to size it properly for the coil, as i am not mathematician i can not give you a equation for it. but basically you will want the cap voltage drop off to align with the projectile reaching the end of the coil, this way the "pull-back" effect is minimal (of course you can still control the cut off with MC/likely high current SS relay) but you DO want the spike in the beginning because the law of momentum, you want to give a big push to get it moving and then the power can taper off while still increasing the projectile momentum as a whole. you will also want atleast 100v if you want considerable speeds from the projectile, but if you couple the cap with battery in parallel then the instant current can come from the capacitor and as the voltage drops off it will start limiting at the current the battery can supply. but remember it will pull from the highest voltage source so it would still ultimately be limited by the current the battery can supply. with carefully placed diodes you could in theory have a 2 stage setup where instant shot is high voltage and then drops to the lower voltage, similar to how a tazer functions but i dont think their circutry would work due to the closed loop in the coil plus quite difficult to implement. you can ultimately put a shunt on the output to the coil to limit the current from the cap causing the voltage drop off to be somewhat extended (less power drawn/less drop) or you can arrange some 18650/21700 cells to supply a good amount of amperage. but i am interested to see what you come up with.

Just a thought how about two or three such coils wired together the same as a electric motor so you have a strobe effect similar to a set of Christmas lights and to fire an old door bell chime to push the projectile into the barrel to engage the first coil

The development is inertial confinement via microwave Helmholtz resonators!!! Hello! Step it up!!!

Multiple discharges from many caps along the distance of the rails like a high voltges AC half wave form. using IGBT Transistors to switch the caps on in a time pulse maybe BSM75GAR120DN2 Datasheet PDF Viewer. - Datasheet.Directory

what if you wind the coils in a cone so that the projectile does not just pull forward, but also pushes .. alternately the cone and the reverse cone ..

make it a dc emf coilgun railgun, no brushes, no control circuit, just dc power, and turn the magnet sideways

This is extremely interesting. I'm not entirely sure if this could be useful, but what about a flyback transformer, I know in CRTs TV sets they get up into the kilo volts.

I am gonna wind a coil very similar to this in about a month but a little different, that i would say is scaleable

Some suggestions: Use an IR Photodiode infront of your coil to trigger an Inverting-Fetdriver for super fast activation and deactivation of your coil. (gets rid of the backwards flying projectile) For the powersupply you can try to use two 6S Lipo batteries (1000mah 100C) in combination with 35V caps for maximum power delivery. (saves chargetime and increases the power output)

this is a very good idea

Jesus, Idk what to do with my brain. I'm not an engineer, or electrician, or any kind of professional science person, but about 20 seconds before you drew those newly shaped magnets, I was thinking, 'it's the fields, you need to see a map of the fields, and be able to tune them to the shape you want. lol... and I have ideas I know are useful, but there is really no reason I should think that is th e case, lol.... I'm certain, in a final triumph, a torus will be involved. Also, I think this is somewhere on the path to perpetual motion. - For projectiles, how about a 2 stage bullet, or 3 stage. So, use one accelerator that gets trapped, but separates from a non-ferrous bullet, and the inertia keeps the front stage moving. - another idea is a pressure mechanism. Something about compressing the magnetic field, and repelling the projectile, instead of attracting it... I've watched a few coil gun vids looking for my own info to emerge(p.s. a dude just did one recently, he really nailed the projectile part), and I can't shake the feeling that this is very close to PM. Without electricity though, just magnets. Cz think about it, anything like this, all you have to do is circle the track back to the starting point, and if it can get there, you have done it.... So, any magnet accelerator can accomplish this. I don't understand why they don't just do that. I have never built one, but it's easy to see they get a bearing, or even a puck, moving fast enough that it could return to the starting point, easily, if on a circle track. I really don't get why that isn't perpetual motion??? I have to assume I'm missing something, cz it seems so easy. Peace ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-id90kjYh-Qw.html

Use capacitor but had resistor to the end, u can make their discharge time to the infini.

Could you use just the high voltage part of the capacitor, and when it gets to a lower voltage switch to the next capacitor and repeat? Id imagine this would require alot more capacitors than normal circuits.

This is very interesting dude !!!

Cool It’s like an electromagnetic slingshot

There is always going to be force pulling the projectile from behind, limiting acceleration, if the length already passed through is still magnetically active.. chained coils w capacitors is still the current best practice for this application

If you dispose the sensor using just one coil would it work? Well, apparently yeah, because you tuned it with your hands and it still bent metal but I just wanted to make sure

Someone please make a coil pistol with a wearable power pack

Super good idea

i have made a coil using 32 gauge enameled copper wire but it does not work can somebody help fix this i have tried different batterys but still does not work do i need to perfectly coil the wires side by side, please i need help

best moment: when he drops the pen.

Great video (and series). I'm starting to plan out a nerf gun driven by a coil and I'm gong to use your coil arrangement here... it actually solves a lot of problems for me quite nicely. Is there a message board or something where I could bounce some questions at you if they come up?

Keep researching! Very dope work (even though I did rip on you r design in a previous comment haha) and most definitely looking forward to more of your content on this topic

Awsome video!!

it's not more efficient as the copper wire has micro fractures in it's coating.. you need to fully cover each coil with a rubber resin to prevent any interference with the next coil layer. :) some use electrical tape but this doesn't work as well as a rubber resin :)

I'm curious and I only just thought of this has anyone ever tried to replace the capacitor with a spark plug and having it so that the circuit is a AC circuit rather than a DC one

What if you used a compressed gas to accelerate the projectile before it reaches the coils? Would that eliminate the backfire issue while also increasing the velocity?

Instead of changing the coil it self cant you just shorten the coils depending on the speed of the projectlie

Great work

ATF ENTERED THE CHAT

Genius

Very cool indeed! Do you foresee any situation where multiple non-linear stages would be an appropriate solution? Btw, better Micing would have made a big difference, you have a good speaking voice and avoid excessive uummms and aaahhhs - so use that mike (even if it get's centrally in shot) to capture a less "recorded in a box" recording! I've found that sensitivity to jumpy video is amazingly large, good audio is way more important.

Wire Gauge could affect this? When the projectile stops at end, does it align its center with the center of the outer most layer?

Like you said, you wouldn’t necessarily need to shut off coils which the projectile has already passed, but wouldn’t shutting them off decrease the forces pulling them back, further increasing the propulsion? I feel like combining your design with the other kind would be better than either on its own

What's the resistance and wire gauge of the coil you made?

AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

So has this worked

Hi, I've been interested in coilguns for years, and now I have an actual budget (I'm 19) I was interested if there was anyone posting research like this here, so I can maybe make my own as good as possible, and I'm glad I found this, cause I'm really kicking myself cause I was thinking about putting a magnetic core outside but I neglected changing the coil shape for some reason. Because we don't live in a mathemathically perfect universe, you'd need a much finer wire/bigger coil to get closer to the effect you wanted, and you'd need... well... an infinitely large coil at the front to literally negate the effect of all the coil behind it and have the point of maximum magetic field *almost* at the front tip (wire thickness exists). I have not watched the rest of the series but I'm intrigued, nice brain you got there.

I believe I have an improvement that can substantially increase the force while reducing current flow. I'm not going to post publicly however, but I might be willing to collab if you are thinking of going commerical.

Wouldn't a Super Capacitor of say 1000 farads at about 24 volts, be able to give the coil a good shot of energy? We could use an SCR to switch the power ON, to the coil.

Is that barrel made of iron

I made a coil gun that shoot electricity only it looks like a bullet shape projectile when I fire it. The more electricity you add the more stronger it becomes

How about an induction type with a copper/aluminium projectile.

Tell me about barrel

Hey bro good video after watching i think i have a rough idea to improve your plan how can we chat

That’s how the Germans did it, that’s how we do it, that’s how infernals not aliens do it!!!

use GOLDEN RATIO,,,

希望可以翻译成中文，很感兴趣但是看不懂

i can barly hear u

Hi

0:19

Holy sweet baby Jesus

I made this video no longer have precisely 666 likes (:

Building a hobby gun at home and am at a juncture im not sure how to cross. Some people say go capacitor others say use inductors. Which one should I use for pre shot energy storage? Power supply isnt a problem. Just stuck there. Help?

this is quite cool, im glad to see someone coming up with new ideas, i look forward to seeing more

Hi, the center of force is a weird way to say it. Yes the force is greatest when it is at rest but it also means that the opposing force made by the opposing magnetic field is also of the same magnitude. This design really doesn’t offer any benefits. The only usable force would be on the first half of the coil if it were cylindrical. However it is this cone shape it just moves the region of usable force and makes the projectile travel down the barrel longer with more friction. Think of it like centroids and all you’re doing is moving the location of it. The integral of the force over distance is the energy transferred and is the same for the same area of the coil no matter how it is wrapped. I would love a response!!

@Andries Putter science is about taking constructive critisism in your theoretical ideas and turning that critisism into practicality as with anything theoretical comes practicality and without theory there is no practicality in science. (to think first is to apply ones knowledge and ideas and make all theory practical and logical) and where possible share knowledge thus advancing in our thought of theoretical application via the will to be practical. :) this is science my friend :)

Salt water conducts energy almost perfectly and put water doesn’t do it at all.

Is it just me or did it shoot with more force, backwards?