# Feasibility of coilgun system for sub-luminar interplanetary transport

My story is set within one solar system, which can be assumed to have a similar composition to that of our own: one Class G star, a handful of rocky inner planets and a handful of gas giants on the periphery. The backstory is that this was a colony from Earth that arrived on a generation ship many hundreds of years ago and has since lost contact with Earth; though this isn't pertinent to the question.

The setting calls for relatively hard-science, so normal Sci Fi tech like inertial dampeners, FTL, and artificial gravity are off the table. In a book series with a similar setting, interplanetary travel was achieved with "fusion drives" which allows for subluminar travel between planets in a few weeks or months. However, while fusion reactors are feasible within my story, a fusion drive still seems like a hand-wavey solution to the problem.

What I would like to use instead is a series of mega-structures in orbit around the various planetoids that act as large gauss cannons to "throw" vessels from planet to planet. Essentially, large interplanetary vessels would be designed to be loaded into the cannons and then fired as a projectile. The cannons themselves would of course need to be massive, on the order of 25km+ in length with massive power requirements to run the magnetic coils. When the ship is arriving on the other end, a similar mechanism captures it and uses the same rail gun assembly to slow it down.

Some issues I've identified with the tech and how they could be addressed:

• Failure condition: in the event that the receiving end of a transit could not capture the craft (for whatever reason) each vessel has enough fuel on board to perform one orbital injection burn as an emergency measure to prevent catastrophic failure.
• Human survival: the accelerations associated with launch and capture would be immense. I haven't done any math, but I'd guestimate on the order of 30+ Gs for a minute or so. Passengers would spend launch and capture unconscious in specialized "crash-couches" that would prevent death during acceleration.

What I'm looking for is an assessment of this method of transit for interplanetary travel compared to other semi-realistic methods; like fusion drives, ion propulsion, or giant chemical rockets.

To be clear, my narrative requires that this is the method of interplanetary travel I use, so I'm not looking for alternatives. What I want to know is how realistic this method is and/or if there are any tweaks I can make to improve it.

Some things to consider:

• Economic viability of maintaining the mega cannons vs. other methods of transport
• Safety of both vessels and launch/capture destinations using this method
• Practicality when considering method requires infrastructure at both ends of the transit path
• Whether orbital paths can be calculated with sufficient accuracy
• Aiming your ships precisely at the receiver seems fairly difficult to me. You may need some thrusters for aligning the final approach at least. But that shold not reduce feasibility of such a system. – Burki Dec 4 '17 at 8:15
• The cost of lifting that much mass to orbit is probably a thousand times greater than the savings on fuel from all the interplanetary cargoes during the whole lifetime of the gaussian cannons. 95% of the fuel requirements of any ship is escaping from the planet's gravity well. Once in orbit, you don't need that much fuel to go anywhere in the solar system. – Rekesoft Dec 4 '17 at 9:03
• I'd argue with that - once you have space based industry, the whole mass-to-orbit problem no longer applies; you build everything from materials already available in space. And although it's correct that the minimum amount of fuel to go from A to B in the solar system is usually quite small, to do so in short timescales requires a lot more fuel. You might as well say that because sailing ships are far more fuel efficient than powered ships (let alone aircraft) on Earth, no one would ever use powered ships or planes. – Andrew Dodds Dec 4 '17 at 10:23
• May I recommend changing "railgun" to "coilgun" in your title? While I haven't been able to find any argument to say that low acceleration rail guns can't exist, the only ones I know of are... well... slightly higher acceleration. You could avoid some confusion by using a term which the reader doesn't already have preconceptions about. – Cort Ammon - Reinstate Monica Dec 4 '17 at 14:55
• 30G acceleration for 1 minute produces a velocity of 17km/s (faster than any Hohmann transfer, but fairly sedate by sci-fi standards), but requires a track length of 529km. Practically speaking, for this to work the acceleration has to be significantly higher- the technology that allows passengers to survive this process is by far the most speculative element of the question. – Catgut Dec 4 '17 at 18:19

# Yes but no

Theoretically yes, the space gun concept has been rattling around since the days of Jules Verne, it's certainly not new.

The first problem, as you've already mentioned, is surviving the launch. Given that you're using it as a catching mechanism as well, you've added surviving arrival to your problems.

The acceleration required to achieve a viable interplanetary speed out of such a weapon launch mechanism would destroy anything more delicate than a house brick, and even that is unlikely to come out entirely intact.

# Acceleration

https://en.wikipedia.org/wiki/Space_gun

A space gun with a "gun barrel" of length (l), and the needed velocity (v), the acceleration (a) is provided by the following formula:

a = v2/2l

For instance, with a space gun with a vertical "gun barrel" through both the Earth's crust and the troposphere, totalling ~60 km of length (l), and a velocity (v) enough to escape the Earth's gravity (escape velocity, which is 11.2 km/s on Earth), the acceleration (a) would theoretically be more than 1000 m/s2, which is more than 100 g-forces, which is about 3 times the human tolerance to g-forces of maximum 20 to 35 g during the ~10 seconds such a firing would take.

To bring that down to a "reasonable" or at least survivable, 30g, means you need a "gun barrel" of at least 200km in length, megastructure is definitely the right word here. Economically, building this could be ruinous, as could running it. I'll leave the energy calculations to the reader as target speed has not been stated but you can probably assume efficiencies of 10-25%.

There's the typical minimal safety involved in any crude space travel, you hope to miss, because if you hit your monument is going to be a large crater. However reaching your target to metre accuracy is just a matter of maths and if you can make the launcher work, it's reasonable to target that well. However, and it's a big however, you're effectively trying to thread a needle by tying the thread to a bullet and firing at it from across the room. It might work, but you're more likely to leave a big dent in the plasterwork unless you're really good.

# Actually making it make sense

The basic concept has some advantages, the "ships" can be basic pods, much cheaper and simpler than a self contained interplanetary vessel. More of the mass and volume are dedicated to cargo. If you can make the launch system efficient enough you could possibly work out an economic model under which it makes travel commercially viable. (You can't in practice as end to end has to be effectively free, but never mind that.)

Space traffic control is a nightmare, this is a model that requires a much higher traffic flow rate than any self contained vessel. Remember that for a single ship travelling under its own power, the cost is the same, whether for the first ship or the hundredth. For this system, the massive up front cost means that the effective overall cost comes down for each subsequent launch. It's a 200km long structure, it cost a vast amount of money, you have to get a traffic rate that makes it cheaper than single launches.

• Don’t forget the receiving mechanism needs to be just as large, and you somehow (well developed rocket tech?) have to get the raw materials for it to the receiving site. – Joe Bloggs Dec 4 '17 at 11:27
• @JoeBloggs, You might be able to use the launcher as a receiver as well, just aim around a planet and catch the return orbit. I mean, I'm not trying to ruin the economy here. – Separatrix Dec 4 '17 at 11:37
• @JoeBloggs, if you started by launching rockets with the ability to slow into orbit under their own power it's possible to build the infrastructure one planet at a time. – Separatrix Dec 4 '17 at 11:41
• The question then becomes why not use rockets, but that’s not the OP’s question. +1 btw. – Joe Bloggs Dec 4 '17 at 11:53
• It would be more like 1000km long. It's a total nonstarter. – Fattie Dec 4 '17 at 21:48

There's one critical consideration that none of the answers so far seem to cover (unless I missed it).

Newton's 3rd Law of Motion Commonly summarized as "for every action there is an equal and opposite reaction."

This means you're accelerating your space station in the opposite direction with every launch. This in turn means that after every launch, your station has to return to station by traditional means. You can't escape this reality. Either you have to fire and equal mass object at the same velocity in the opposite direction(really dangerous) or you have to burn engines to counter the thrust.

• An important point. But solutions could be found- big solar sails to slowly-but-passively recoup acceleration lost, firing payloads from different places in orbit so that they sum to smaller orbital corrections, etc. – Maxander Dec 4 '17 at 18:48
• Good points. But definitely something that needs to be considered. Ignoring Newton is one of the biggest breakers of immersion for hard sci-fi imo – Stephan Dec 4 '17 at 19:58

Disclaimer: This is all rampant speculation by an enthusiastic amateur.

The problem as I see it is the acceleration is literally killer.

So the solution is to ramp up to speed slower with a gauss cannon taking the form of a mag-lev track running in a wide circle with a switch-track to curve off gently to the final trajectory. (it'd have to be a gentle curve rather than sudden straight simply because the change of trajectory would itself affect acceleration)

Your Catcher would have the same configuration, you'd simply enter through the switch-track and decelerate in the ring until you reach a velocity gentle enough that you can be safely released from the system.

This has some advantages, if you accelerate at 1g you would get the benefit of centrifugal forces on your passengers until you join the switch-track, making the probable weeks of acceleration/deceleration much more tolerable to the human frame.

Some problems would be that your ship would be the only ship that could be in the loop at any time, any later ships would rapidly catch up and collide with the first one within the first loop around. So your massive infrastructure would have potentially a very small payload and quite low turnaround.

The possibility of using this for cargo is much more likely, you could apply much higher changes of velocity in the system without worrying about pasting fragile fleshy people.

• I really love this answer, incredibly creative – enpaul Dec 4 '17 at 15:29
• Problem: the centrifugal forces felt only care about your angular velocity, that is to say the faster you go around the circular track the greater the centrifugal force, regardless of how slowly or quickly you reach that speed. This requires an immensely huge track or the centrifugal force at the final launch speed will still flatten squishy humans. – Joe Bloggs Dec 6 '17 at 7:57
• If you want an intuitive explanation for this: as your payload goes around the track it’s accelerating up and down. Let’s assume the switch track points straight ‘up’. At some point the payload still has to do that vertical acceleration from the ‘lowest’ point on the curve up to the end of the switch track, which has to equal accelerating vertically from a standing start up to your launch velocity, except with a quarter curve in the way to add complications. – Joe Bloggs Dec 6 '17 at 8:02

I don’t think a gauss cannon is a good solution in part because of the tremendous acceleration required and in part due to the massive infrastructure needed. However it is almost certainly possible.

I would suggest a large number of wide torus shaped electromagnetic stations aligned so as to enable a craft to be accelerated through each one. When the cannon was needed these electromagnetic stations would be positioned along the proposed trajectory and the interplanetary craft would be accelerated through each one in turn.

After the craft had left the “recoil” would have to be damped by using the electromagnetic stations to repel one another as the end of the series would probably need to be free floating stations due to the great lengths involved.

There are too many variables to provide any exact numbers but I imagine the cannon would need to be of the order of tens if not hundreds of kilometres in length if you didn’t want to crush the crew. A nasty but survivable 10g applied for 40 seconds would give a space craft delta v of 4 km/s, enough to get from Low Earth Orbit to Mars. But it would need 80km of launch “tube”. Assuming your brutally punishing 30g the same velocity could be achieved in under 14 seconds with a tube length of around 26km, however I think you would need a lot of back story hand waving to explain why the crew generally survive rather than generally die.

One big issue would be alignment, especially if the far end accelerating stations were free floating. That said with high tech I would have thought this should not be insurmountable. There would also be the matter of recoil the free floating stations might crash into the other parts of the cannon if the breaking failed. Perhaps some emergency thrusters could push them out of alignment if this happened to avoid collision.

The economics of this would depend on too many variables. All I would say is that a massive amount of electrical power would be needed as would massive orbital stations so both power and space construction need to be relatively cheap.

The safety is in part dependant on the forces used even 10g is not really “safe” and 30g is probably deadly in many cases.

The infrastructure would be huge and not very practical. A new cannon would be required for each destination whereas conventionally propelled craft could go anywhere.

I doubt that there would be any serious problems with calculating the correct trajectory and orientation of the cannon, although mid-course correction would probably still be required by the use of thrusters of some sort (ion drive would be my suggestion here).

I suggest that the craft carry ion engines as backup. They will already have powerful electrical generation capability for use in conjunction with the gauss cannon so it would make sense to use this for the backup plan. The propellant required could also be much lighter as the exhaust velocity of an ion engine is very high.

In summary its an interesting idea but not really that practical. You might want to consider other variants depending on how wedded you are to the static cannon idea. Perhaps a hybrid cannon/ion drive, a multi stage cannon or a series of cannon like accelerators in continuous orbit used by craft in both directions for acceleration and deceleration?

• A series of rings, push/pull stations like the HyperLoop by Elon Musk. Pointing at multiple targets might be complicated so may need dedicated ring series that stay pointed at other planets as they orbit the sun. I like the idea but they would have to shift angles between launching and receiving since you'd be firing ahead, where the planet will be in 2-4 days while receiving from where the planet was. Can we call them stargates? (please, please, please?) I suppose if the planets are in the same ecliptic a fast orbiting series of rings would work but require more energy for alignment. – Hebekiah Dec 5 '17 at 8:36

Following up my original answer, I went looking for statistics on 1G accelerations.

I turned up this gem https://space.stackexchange.com/questions/840/how-fast-will-1g-get-you-there

By my estimation, at a steady 1g, passengers would spend 11 hours in the loop at each end, and about two days of flight-time coasting between earth and mars under optimal conditions at 401,235 m/s (estimated for a 45 hour flight based on the optimal 1G powered flight profile with a turnover in the middle)

3 days to mars sounds cool, but it's highly contingent on the loop's ability to divert that kind of velocity into a curve. the more powerful the magnetic containment the smaller the loop can be.

Scale affects everything, your projectile is moving 4000 times faster than the fastest bullet, so that forces your loop to be very big indeed, very likely the loop would need to be a ring all the way around the moon or earth itself, under those circumstances your best option is probably to build multiple orbiting facilities with their own sections of gauss coils. far easier than building a contiguous structure and it'd be inherently much easier to upgrade and replace parts. you could also use the facility from day 1 of construction at reduced capability and improve it as funds and resources permitted.

Highly unrealistic.

Economic viability: Maintaining rail gun systems is extremely costly. As it stands with Rail guns right now. The Barrel has to be exchanged every couple of shots because it melts due to the heat generated by accelerating the payload.

Safety: Extremely unsafe. The passengers probably wouldn't survive the extreme acceleration of the cannon. Having fuel on the vessel to correct the flight path would certainly result in an immediate explosion inside the cannon.

Practicability: There's no way to effectively catch the vessel. Also, as @Rekesoft already said in the comment, there are way better launch systems if you already start from orbit.

Calculation of orbital paths: Should not be a problem.

• I agree with a lot of this but the barrel should not need changing if we are talking orbital stations. Ground based operation would be totaly unrealistic due to frictional effects.They way to catch the vessel is to align the incoming craft with the gun and run it in reverse. – Slarty Dec 4 '17 at 12:43
1. Loops

Multiple answers to this question have mentioned the idea of using a loop rather than a straight barrel to ramp up speed over time rather than turning the occupants of your craft into a G force induced paste. I would wholeheartedly agree with their idea as it allows us to bypass the need for special safety measures on the occupants, and adds an interesting quirk to your space travel in that travelers would be forced to wait in these acceleration loops for a period before leaving and after arriving at planets.

b. Modularity

The system can be built incrementally making travel faster and more efficient as each new module is installed. This will assist in offsetting the structures gargantuan cost.

2. Psychology

This is going to sound weird, but I think one of the largest barriers to entry here is psychological. If I put you in a car and give you the controls, you feel like you are in control. However, if I place many individuals into a plane where they have no control, they become anxious even simply unable to function. This problem is magnified greatly if I am going to be sending people millions of miles at speeds approaching the speed of light with no way of helping themselves in the event that something goes wrong with the catching system, or a miscalculation occurred with their trajectory. This brings me to my next point.

3. Targeting

Space craft missing their marks when simply trying to hit a planet is a concern, let alone hitting a small, say few hundred meter wide, target. For instance, Curiosity's landing zone was a 12 mile by 4 mile ellipse. Did it land within that ellipse, yes, but that kind of uncertainty simply will not do in a system like this. Accuracy is getting better, but errors can always occur.

Granted, the spacecraft in this system should never have to deal with any sort of drag or air resistance pushing them off course, but as vacuous as space is, you cannot assume your path will always be clear.

4. Energy Harvesting

An interesting side note, and not really a problem, in fact it is a benefit. Your catching system could use the incoming space craft to generate electricity in the form of the world's largest electrical generator. As the spacecraft passes through the electromagnets it could be used to generate a charge. The resulting charge could then be used to launch other spacecraft significantly reducing energy costs. (It wouldn't be 100% efficiency, but with a system this big, every Joule counts.)

Happy world building!

• See: Large Hadron Collider. But I'm not sure the catch and throw mechanisms can be the same unit in this design. Usually, speed is generated by shortening the radius as in a spiral. – Jammin4CO Dec 4 '17 at 15:51

There is an additional reason why coil guns don't scale up to high speeds that is seldom mentioned.

Let's start by taking a quick step back and saying what a coil gun is. A coil is a big loop of wire. We have a metal projectile moving at some speed towards the hole in the middle of the coil. We run a current through the wire. This creates a magnetic field which pulls the projectile towards the center of the coil, speeding it up. When the projectile is through the coil we turn the current off, because we don't want it to be pulling the projectile back towards the coil.

We then build many stages so that at each stage the projectile is sped up a certain amount.

The problem therefore is:

• Current has to be as high as possible, because the magnet pulls harder if the current is higher.

• Current has to be on for precisely the right amount of time. Too short and you're not giving enough pull; too long and you're pulling backwards.

The initial stages, where the projectile is moving slowly, are easy. It's the middle and final stages where the projectile is moving at several thousands of meters per second, that are tricky. If one coil is a meter long then the object is in the coil for less than a millisecond, and thus we need to trigger a high current very precisely.

So think about how to do that. We have technology in our computers for triggering extremely small currents down to a fraction of a nanosecond and we can put billions of them on to a tiny silicon wafer. Now think about how hot your computer gets when manipulating those extraordinarily tiny currents. This is the technology you want to scale up to the point where the currents are equivalent to the current draw of a small city. High speed switches are made out of semiconductors. They're by design not perfect conductors, so they make heat.

If you work out the math, the heat sink you need to keep the on-off switch of each stage of your coil gun from melting is enormous, and sheer amount of semiconductor material you need for what would easily be the world's biggest transistor is unfeasible at today's prices. For one stage; and you'll need thousands of stages.

In short: the electricity that powers a big coil gun is cheap and readily available at scale; it's the electronics in the control system that we don't at present know how to scale up to the currents required to attain space vehicle speeds.

Must admit I've been toying with a similar idea, except at interplanetary scale.

The huge advantage is that you are free of rocket equation constraints. As all the energy is delivered externally, you don't have to hand wave about Bussard Ramjets and the like. The presence of a catcher is also required, unless you are going to try and carry all the fuel for the slowdown phase, which pretty much defeats the point. If you are doing that then you might as well have a detachable 'acceleration phase' stage.

So what are the problems:

1) As mentioned above, massive sustained acceleration, unless you have an amazingly long rail gun assembly. In a zero-g environment, a structure that is 100s of km long is not impossible. I would presume that in a zero-g vacuum environment a barrel would not be required, if your magnets can keep the projectile going straight. Indeed, at the speeds involved a physical barrel would just be a hazard.

For the payload, you'd have to go beyond acceleration couches. Think full immersion in a fluid (no air gaps, even in the lungs or ears); possibly freezing solid. You want to avoid any density contrasts within the ship in the acceleration/deceleration phases, to allow for the 100+g loads. Once in cruise phase, you could come out of vitrification.

2) Safety - Well.. you've accelerated a ship weighing hundreds of tonnes or more to, perhaps, 1000km/s. Going by the kinetic energy formula, e = 0.5*m*v^2, that would be ~10^17 J According to the Wiki, that's about a Tsar Bomba). So if anything goes wrong and you hit a planet (never mind a KBO) you will leave a noticeable dent. Even under good conditions, your catcher has to dissipate that amount of energy in a very short time. You'll need some very, very big capacitors.

3) Practicality - As long as the system is reliable and the energy requirements are solved, then it avoids sending huge fuel tanks out on long, tedious orbits. So in many ways better than chemical rockets or solar sails. It also requires some form of hibernation/sleep tech, as above.

4) Orbital Paths - You would probably need some form of adjustment onboard, but the calculations would be pretty trivial. You'd also want adjustment thrusters in case space junk was detected en-route, as collisions at cruising speed would be very very dangerous.

• Just watch out for recoil. – Joe Bloggs Dec 4 '17 at 11:33
• That's quite interesting. You would have to eject mass in some form to allow for conservation of momentum.. and somehow use that ejected mass to transfer momentum to the nearest very large body. – Andrew Dodds Dec 4 '17 at 13:20

The concept of interplanetary travel by railgun isn't entirely feasible. When O'Neill and his associates were developing the concept of L5 space habitats they experimented with mass drivers as a method of launching vehicles. They discovered that there was a limit to the maximum velocity mass drivers or as here gauss cannons could attain. At higher velocities the vehicles wreck their electromagnetic launchers.

This limiting velocity was 4 km/s. Now vehicles could be launched from orbit to this velocity. The vehicles' velocity, due to the gauss cannon, can be added to its orbital velocity. vehicles could be launched from an earthlike planet with a velocity of 12 km/s.

This does produce feasible interplanetary travel. The travel times will be fairly long. Six to eight months travel time to planets in the equivalent of the orbits of Mars and Venus in your hypothetical solar system. Upwards of many years for travel to other planets. It will work but it will be slow.

This form of travel is roughly on a par with chemical rockets. But fusion, plasma and ion drives will perform better and faster.

# Have some numbers!

Okay, so I don't know how to make numbers look pretty, so I'll be using code blocks :(

Also, for the sake of having some real numbers to work with, I'm going to assume a nice simple trip from Earth to Mars at the point when they are closest together. All numbers are taken from NASA.

30g = 300 m/s^2
delta-v mars-earth = ~5,000 m/s
minimum distance mars-earth = 55,700,000


So first of all, how long is this gun going to have to be? You said about a minute, so I'll give numbers for 60 and 120 seconds. For the rest of this answer I will continue calculations for each of these numbers, the first being the craft accelerated for 60 seconds and the second being the craft accelerated for 120 seconds.

d = vi * t + 1/2 * a * t^2
d = 0 + 1/2(300)(60^2)  = 540,000 m
d = 0 + 1/2(300)(120^s) = 2,160,000 m


How fast does this get us?

v = vi + a * t
v = 0 + 300(60)  = 18,000 m/s
v = 0 + 300(120) = 36,000 m/s


Plus the 5000 m/s perpendicularly from the difference between Mars' and Earth's orbits gets us:

sqrt(5000^2 + 18,000^2) = 18681 m/s
sqrt(5000^2 + 36,000^2) = 36345 m/s


At the point that the craft passes Mars. How long does this trip take us?

t = d / v
t = 55,700,000/18,000 = 3094s
t = 55,700,000/36,000 = 1547s


So how much energy are you going to need? For this we need the weight of the craft. You said we need to be able to perform an orbital injection, so the rocket equation is coming into the picture! (YAY!) WolframAlpha has a nice tool built-in for the rocket equation, so I'm going to give all the numbers to it here.. I assume a (very light) dry mass of 1,000 kg and shoot to do nothing more but get the craft below escape velocity of Mars. These are the initial masses we get:

m0 = 15,180 kg  (for the slower craft)
m0 = 523,219 kg (for the faster craft)


Now how much energy will it take to get these crafts up to speed?

E = 1/2*m*v^2
E = 1/2(15,180)(18,000^2) = 2,459,160,000,000 J
E = 1/2(523,219)(36,000^2) = 339,045,912,000,000 J


Actually the energy requirements aren't as horrible as I had imagined, but a 500 km long gun seems pretty ridiculous to me. Play with these numbers as you see fit to try and figure out what works for your story!

• The energy requirements are really low because all you are accelerating is the projectile. The interesting thing to think about is: how long is the projectile in each stage of the coil gun when it gets close to speed? And the answer is "not very long at all"; the problem becomes not the total amount of electricity, but rather how to control it so precisely. – Eric Lippert Dec 4 '17 at 16:28
• I agree and encourage that to be it's own standalone answer. I aimed this answer at tackling just the raw physics; I'm definitely not an engineer ;) – bendl Dec 4 '17 at 16:31