A blog dedicated to helping writers and worldbuilders create consistent, plausible Science Fiction.

Monday, 14 March 2016

The Big List of Propulsion Failures II

We continue the big list of propulsion failures.


Also known as 'plasma drives', these engines use magnetic fields to accelerate specialized propellants. The VASIMR fits inside this category.
Basically, they're electric motors that push on plasma instead of coils, and just like electric motors, they aren't prone to exploding.

VASIMR proposal.

They do, however, have a wide variety of ways to fail. The exact methods depend on the type of electromagnetic rocket you use (a little research goes a long way!), but they all have four common components: electric supply, propellant preparation, magnetic accelerator and nozzle.

The electric supply handles the transport of high amounts of electrical energy through wires and into the engine. It might seem simple, but the actual complexity is astounding. You could have transformers for converting current from the reactor's voltage to the engine's voltage, you can have banks of capacitors to handle the peak loads of a pulsed engine, you can have internal radiation shielding for the critical nodes...

This means that an 'electric supply failure' can be written as a transformer melt-down, a capacitor blow-out, a mechanical failure of the wiring, with consequences ranging from a burnt plastic smell to arcs of lightning frying the propulsion section.
Electrical explosion at a powerplant.
The 'propellant preparation' section is what handles the entire process of putting propellant in the engine chamber. It might be seeding helium with potassium or vaporizing alkali metals into plasma, or simply pumping hydrogen into the chamber. In here, propellant might leak, burn through the walls of the engine or form bridges to destructively conduct electricity to where it shouldn't go. Or, the heating coils melt through their casing, the seeders malfunction and choke the engine with unprepared propellant, or radiation damage causes a lower than normal voltage leading to a loss of thrust...

The 'magnetic accelerator' and 'nozzle' components work and fail depending on the type of engine in use. If they are superconductive, they might be vulnerable to warming up and massively increasing their resistance.This could shut them down... or melt them on the spot before control systems can respond. Variances in the power supply to the magnets can also cause the propellant being accelerated to expand, narrow, vibrate, flare up and damage the interior surfaces of the engine's chambers. The magnets themselves can apply too much force and break out of their bracings, or if they've been damaged by radiation, fissure and crumble by themselves.

Worldbuilding tips: These engines rely on the juxtaposition of cold and hot, lots of electricity and conductive materials... they might not explode like a chemical rocket, but they will spit fire. They will also rely on continuous monitoring by an engineer. An unscrupulous Company's might reduce costs by replacing their spaceships' magnets long after they're safe to use, or a pirate might get a risky 'boost' to their thrust by running their engines hot and risking a meltdown.


Engines that rely on electric fields to repulse ions or charged droplets out of a nozzle. They share many of the vulnerabilities of the electromagnetic propulsion systems, but instead of magnetic failures, they have an increased risk of electric failures.

Electrostatic engines work by ejecting charged particles, so by definition, they create a charge imbalance commonly known as static electricity. Generally, this imbalance is countered by ejecting something of opposite charge, in equal amounts, such as through the use of an electron gun. 

If this electron gun breaks down, or issues more or less than expected, a charge imbalance will cause the spaceship to become negatively charged. The propellant will bend around and return to the nozzle. Eventually, like in a lightning cloud, lightning bolts will start striking the hull. The propellant will hang near the spaceship, and finally become unable to leave the nozzle entirely.

The more ions are emitted, the quicker this process occurs.

Some electrostatic designs rely on physical anodes or cathodes in contact with the propellant. These will be eroded away and have to be replaced... and this can be irregular, leading to an unexpected propulsion failure.  Putting too much propellant or foreign material in the rocket chamber can reduce the impedance to the point where electric arcing occurs.  

Here is a guide to spark plug failure, relevant to physical-electrode rocket engines.

Worldbuilding tip: A slip in the crew's vigilance might be all it takes. It can be the basis for a scenario where an electron gun sensor malfunction leads to the a charge imbalance. The crew would become trapped inside their own spaceship, unable to spacewalk due to the electric discharges, and unable to change their trajectory because running the engine creates no thrust. If an essential burn is coming up, they'd have to resort to hazardous solutions for cancelling their charge...

In a way, this is the most robust type of electric engine. Like a glorified cooking stove, it uses various methods to heat up propellant and eject it through a nozzle. 

An arcjet intentionally creates an electric arc to heat propellant with.
The lowest performance, but simplest, electrothermal engine is a resistojet. Propellant is passed over a physical heater. This heater has a temperature limit. The accelerated propellant and vacuum evaporation eats away at the surface of this heat exchanger... chemical reactions might also occur, either with the heater or within the propellant. Cosmic radiation and neutron embrittlement might affect it too. All these affect the temperatures it can handle, which might lead to situation where it melts away, or, like a grenade, explodes into shrapnel. 

Another problem is that many of these components rely on the low resistance of its electric conductors. For most metal conductors, increasing the temperature increases the resistance. This causes a thermal runaway which ends up in fuse blowing out... or a complete meltdown of the system. 

An arcjet has similar behavior, with a use of electric arcs eroding the electrodes even quicker, and more randomly if the propellant gets contaminated with conductive elements.

The fanciest electrothermal rockets use methods such as magnetic induction or microwaves to heat up the propellant indirectly. This removes the limitations imposed by the physical components, but adds complexity and therefore different failure modes. 

The wakefield electron beam rocket, for example, uses a very short laser pulse to ionize a cloud of propellant, allowing it to be ejected by electrostatic forces from an electron beam. The laser generator, the internal optics, the electron gun... any of these components can go wrong.

Wakefield electron accelerator diagram.
Fusion Closed:

Ah, fusion. A favorite of scifi authors. In Fusion Closed, the fusion producs are not used to heat the propellant directly, so this deals mostly with reactors being used to provide power for an electric engine.

The W7X Stellarator with its futuristic wavy magnets.
You might already be familiar with some reactor configuration, such as the Tokamak or the Stellarator. A ring of magnets and heaters try to compress fuel plasma until it fuses, then try to contain it. Many would think that it would explode like the conventional SF reactor, but no. The fusing plasma might be very hot, but there's very little of it and it contains little overall energy. Switching off the magnets means it dissipates and cools down very quickly.

What can happen is that the reactor is pierced, and is flooded with atmospheric gasses. This absorbs the heat and leads to a satisfying explosion.

Alternatively, like all magnets generating powerful fields, they can pop out of their bracing, melt down, break up into pieces or short-circuit.

D-T fusion
Most of the easy fusion reactions produce neutrons. This means that the spaceship relies on neutron shielding. An imbalance in the fusion rate might lead it to being extinguished like a delicate candle flame... or suddenly increase and lead to a neutron burst. While they might not have the penetrative power of gamma rays, they may penetrate the internal shielding and cause radiation damage to the reactor. Neutron embrittlement increases the likelihood of the failure mentined above, so this is a self-reinforcing failure mode.
Fusion Open:

Directly heating propellant with fusion products is perhaps the ultimate in SF propulsion. Astounding thrust and exhaust velocity and a staple in torchships. 

Most proposals rely on magnetic confinement of the fusion plasma, with propellant run through the reaction chamber.

Gasdynamic fusion thruster.
In this configuration, there's always a lot of hot propellant inside the engine, so sudden failure of the magnetic fields does lead to an explosion. This means that monitoring the fields generated is a vital activity, and the crew must be cautious of thermal runaway, electrical shortages, conductivity changes, contaminants, interference by external magnetic fields and so on.

The propellant also creates physical pressure within the reactors. Instabilities and unevenness in its heating can lead to mechanical effects on the fusion plasma, meaning disturbances in the magnetic fields, and on the reactor walls. Vibrations can build up and tear it apart. 

A small physical disturbance can lead to a shift in the magnetic fields, making them push in the wrong direction. If they are very strong magnetic fields, they might start pushing on their bracing mounts in ways not designed for those loads... many of you should be familiar with the exponential increase in attractive force as magnets get closer to each other. 

Some configurations, such the Polywell fusor, use an external power source to initiate and heat the fusion fuels. While these might be prone to failure, they are unlikely to cause accidents. Instead, they will fail to initiate the fusion. Some, like the Z-pinch, can have the electrical system failure modes described for the electric engines above, with the increased risk that comes with accumulating large amounts of energy and discharging it all at once.   

Worldbuilding tip: These types of reactors are vulnerable to both internal and external events. If two spaceships get too close to each other, they might destabilize the magnetic fields of their engines and lead to a mutual suicide. If a close encounter seems inevitable, they might be forced to flush their chambers with propellant to immediately halt the fusion reaction and switch off the current to the magnets. The might even have failsafes dedicated to this sole function. 

Fission Solid:

A nuclear fission rocket is not like an atomic bomb, and when things go wrong, they don't happen the same way.

What it does it push liquid hydrogen over a very hot rod of uranium and eject it out the nozzle. Due to its low exhaust velocity, it has a large mass flow - in other words, the piping system that takes the liquid hydrogen from the propellant tanks, over the uranium and holds it until it reaches the nozzle has to large, durable and complex.

To simplify the handling of the propellant as it transitions from cryogenic temperatures to pressurized vapor, the reactor is divided into a series of cylinders. Add in pressure release valves, insulation, pressure vessels and it'll end up resembling a locomotive's steam boiler.

Providing too little coolant (in this case, cold propellant) or too much heat causes temperatures and pressure to rise. This can melt the propellant lines, causing to leaks. If the propellant is liquid hydrogen, it can supercool its surroundings and lead to thermal shock and hydrogen embrittlement. The effects of these can manifest long after the rocket has stopped running. 

Most dangerous is when this damage affects the nuclear rods and their control mechanisms. Dislodging them can put them closer together than intended, increasing their reaction rate. If a control rod gets stuck in an 'ON' position, it would make it very hard to open up the reactor for repairs. Similarly, damage to the neutron reflectors can lead to a very neutron-poor environment and a 'cold' reactor. This situation might rapidly change if attempts are made to speed up the reactor. 

Some rocket designs rely on the propellant acting as a neutron moderator. While this simplifies design and the most effective moderators (water) are easy to come by, it also leads to sudden changes in reactor state if the moderator's flow, temperature and contents are affected. For example, if a moderator supply line is cut by a projectile, the section of the reactor might find itself in a 'dry' state, with no moderator between its uranium rods. The neutrons available will decline and the reaction rate drops, and the temperature with it.

The nozzle is another vulnerable system. It is permanently exposed to high levels of radiation, high velocity propellant and extreme temperature gradients. This imposes strict service lifetimes on it. If ignored... it might suddenly shatter and lead to assymetrical thrust that flips the spaceship. More slowly, the heat will warp it into non-circular shapes, causing reaction control systems to fire to compensate for the torque. The hottest part is the throat, which is also the narrowest and hardest to cool. If the engine is run too long, the nozzle might drop off entirely, cut off at the throat. 

Temperature gradient in a nuclear thermal rocket nozzle.

Out of the nozzle,  a tiny amount of radioactive particles can be ejected. If the uranium rods are 'old' or have a rough, poorly manufactured surface, the will drop much more radioactive particles into the exhaust stream. 

The propellant itself might be a source of failure. Water, if run over very hot uranium, can decompose and reform into superoxides, hydrogen peroxide and various nasty oxidants. These can eat through the various engine components and chambers like acid. Methane and other hydrocarbons can similarly decompose into carbon-rich solids that, like soot from a fireplace, coat the interior of the engine. This reduces the efficiency at which heat is transferred from the reactor to the propellant, leading to lowered thrust and efficiency. This has to be cleaned out or it can completely trash a reactor.  

Worldbuilding tips: NTRs can be seen as the 'problem engine' of developing space programs. It is expensive due to the uranium required, has a short lifetim, generally bathes a huge volume in space in radiation and the juxtaposition of cryogenic propellant and 2000K+ temperature parts can always lead to problems. Despite its performance, it might be forbidden for use in densely populated orbits or on human-carrying spaceships entirely.

Everything aft of the shadow shield has to be replaced regularly.
In the final part, we'll look at the gas-core fission rockets and pulse-propulsion systems.


  1. “The effects of these can manifest long after the rocket has stopped running.”

    For how long I wonder? Could you have the heroes hide from the antagonists in a graveyard/boneyard of old spacecraft, only to find that it’s more radioactive than they first thought? Dun, dun, duuh…

    “The propellant itself might be a source of failure. Water, if run over very hot uranium, can decompose and reform into superoxides, hydrogen peroxide and various nasty oxidants. These can eat through the various engine components and chambers like acid.”


    I guess all those nuclear electric designs of mine will be scrubbed from the world building. You could probably have an evil dictator even weaponise this if they were fond of particularly over-elaborate methods of destruction….
    I did wonder if the propellant could be used as a limited shadow shield (with some propellant always kept as a reserve to serve for emergency manoeuvres/ radiation shielding). Probably unlikely, but I thought I’d ask.

    P.S.: I wrote a comment on the air war section, just wondering if you might find it useful.

    1. I'm very happy to see commenters!

      -Long after the rocket has stopped running:

      Three forms of long lasting damage don't have immediate effects. The first two are neutron and hydrogen embrittlement. Neutrons bombard metals to make them weak and brittle, and radioactive leaks and spills can cause this over time. Hydrogen leaks are the same, creating bubble in metal that undermine its structural integrity.

      Next time you switch on a reactor and apply pressure, it'll snap without warning.

      The third time-bomb-failure is temperature effects. You may know that metals expand and contract after being subjected to high temperatures. Over several cycles, microfissures become larger, add up and spontaneously split a component in two.

      Oxygen is nasty, yes.

      Nuclear electric designs have their uses though, especially in terms of power density. But, just like you don't expect your car to run forever, just make sure to remind your chaaracters that nuclear technology isn't a black box that can be forgotten about.

      Propellant is an excellent form of radiation shielding... it's just that as the mission goes on, you'll have less and less of it. A propellant reserve means you need more 'real' propellant, so it'll necessarily be small.

  2. Has anyone ever looked at linear accelerators as a type of electric drive? Like a mass driver only instead of buckets you use a magnetically susceptible propellant like powdered iron or even powdered LOX.

    1. Mass drivers can use just anything as propellant. However, they have three major disadvanatges as a propulsion system:

      -Exhaust velocity is the muzzle velocity. Generally, this means poor efficiency and having to use lots of propellant.

      -Very low power-to-weight ratio. The magnets, electrical systems, structural bracing, capacitor banks and reactor all add up to create a propulsion system that is very heavy for its mass.

      -Expense and complexity. Compared to a chemical rocket or solid core nuclear thermal rocket, a linear accelerator is a huge piece of electrical engineering that uses tons of expensive equipment, such as superconductors, supercapacitors and high-tesla magnets.

  3. I recently started reading your blog. I think it's quite good, but my 1st comment will be to point out an error.

    In a nuclear reactor the moderator does NOT slow down the reaction. The moderator slows down the neutrons so they are more easily captured by the fissile material to cause fission. So removing enough moderator stops the nuclear reaction.

    OTOH the moderator often also absorbs neutrons & with poor design a loss of coolant (which may also be moderator) can result in the reaction increasing. This case is called 'positive void coefficient'. The Chernobyl reactor got most of its moderation from graphite rather than the water coolant & under the odd conditions of an experiment being run had such a runaway reaction.

    In a nuclear thermal rocket, using the propellant as (some of) the moderator means the nuclear reaction does not run unless there is propellant going through the reactor. Possibly sufficiently poor design could make a NTR have a run away reaction with the propellant flow cut off, but it seems unlikely to happen.

    Jim Baerg

    1. Welcome to the blog. I hope you find more of what interests you!

      Thank you for correcting me on what a moderator does. I'll edit the post in accordance.