Wednesday 5 October 2016

Space Warship Design IV: Complete Examples

In this post, we'll use the numbers we've put together so far, and the baseline spaceship from the last post, to create three 'complete' space warship designs.
Each warship will have a datasheet and a commentary on design decisions and tactical considerations.

Martian Interceptor

Description: A high thrust, high performance warship that uses drop tanks to achieve enough deltaV to intercept interplanetary fleets. A solid-core nuclear thermal drive minimizes the need for radiators, but requires an on-board electric powerplant to power the lasers. Expendable mirror-drones are used to out-range potential targets.


Name: Martian Interceptor

Role: Intercepting interplanetary Terran fleets
Dry mass: 1986.3 tons
Mass percentages: 68% Armor, 32% Components
Component masses:
50 ton solid-core nuclear thermal engine
100 ton nuclear-electric generator
100 ton armored radiators
38 ton Gyrotron-VECSEL laser generator
85 ton heat pumps
1 ton optics
112.5 ton mirror drones
11.4 ton life support
46 ton sensors
8.8 ton avionics
45.1 ton structure
33.6 ton tank mass
Propellant mass: 3359 ton internal, 34201.4 ton inner, 253035.6 tons outer

DeltaV: 5km/s internal, 15km/s inner, 25km/s outer.

Acceleration averages: 1.28g, 0.39g, 0.05g
Propulsion output: 100GW
Electrical output: 500MW
Ammunition: 8x Long Range mirror drones, 50x Short Range mirror drones

Standard laser performance factor: 5.7

Laser armor depth: 163mm, rotated, sloped.
Kinetic armor depth: 400mm, 600mm

Shape is a narrow cone on top of a water cylinder.

Component design:

100GW 50 ton nuclear thermal solid-core engine. Propulsion waste heat is absorbed by 8000kg/sec mass flow. Thrust is 36MN (90% efficiency).

100 ton nuclear electric power generators produce 500MW. At 33% efficiency, they put out 1000MW waste heat at 1500K.

Power generator waste heat removed by 2000m^2 of double sided radiators. Mass is 100 tons with 50kg/m^2 (armored). 1000 tons of water can be boiled off in 37 minutes if radiators are damaged.

Defensive gyrotron produces 333MW beam, 83MW waste heat, pushed from 500K to 1500K by 83MW heat pumps.

Offensive Gyrotron-VECSEL produces 158MW beam. Heat pumps move 170MW waste heat up to 1500K, consuming 170MW.

Gyrotron masses 34 tons. VECSEL masses 4 tons. Heat pumps mass 85 tons.
Aiming is through a 1 ton side beam window.

6m radius mirrors composed of 113 hexagonal 1.12m diameter segments of 1m^2 each, massing 4.9 tons for the mirror, 100kg for communications, resistojet RCS and electronics. 15.8kW is absorbed from the laser. It is removed by 35g/sec of liquid hydrogen boiloff. The heat exchanger masses 13.5kg/m^2, based on a microtube design. 1.53 ton exchanger. 100kg liquid hydrogen provides operation for a maximum of 1 hour of continuous shooting. Power for pumps and the rest is provided by a 1kW 200kg electrostatic-thermoionic RTG generator. It is equipped with 12x2kg thrusters at 5 radial directions. 700kg of LOX/LH2 propellant in 10kg tank provides 500m/s deltaV.

Total mass is 7.4tons. Eight mirror drones add up to 59 tons, with 0.5 ton onboard LH/OX cracker and cryo-cooler tanks. Volume is 20m^3 per drone.

Closer combat mirrors of 2m radius are composed of 12x1m^2 segments, massing 528kg. 100kg for control, 162kg for heat exchanger, 100kg for LH2, 60kg 300W RTG, 24kg for thrusters, 2kg tanks, 98kg propellant for 500m/s deltaV. Total is 1.07tons. 50 of them mass 53.5 tons. Volume is 5.5m^3 per drone.

Long range mirrors are nine times as effective compared to a 2m radius integrated mirror, and has the flexibility of one-ship flanking of multiple opponents, along all three axis. Close combat mirrors have swarm behavior.

Life support for a crew of three, with 1 month endurance. Pilot/programmer, comms/commander, engineer/teleoperator. Separated into two control centers and an inflatable rotating habitation ring. Consumables: open-ended 180kg of oxygen (72kg) and food (108kg). Water is drawn from the propellant tanks. Water recycling and CO2 scrubbing are handled by 150kg of equipment.

Control center resembles fighter jet cockpit and masses 1050kg.
Inflated habitat masses 60kg/m^3 and has a radius of 6m, and is a tube ring 2m wide. Mass 7.9 tons.

Total crew mass 11.38 tons.

Forward sensors are 10 degrees arc wide-angle x36 to cover the entire front arc, plus one forward facing wide-angle, plus three redundant 1 degree narrow-angle. Sides and rear are covered by three wide-angle sensors in scanning mode. Total mass 46 tons.

This is a very aggressive sensor layout. It relies on a planetary network to detect flanking threats. Multiple redundancy helps mitigate damage and blinding. Fixed positions allow for better detection of incoming stealth projectiles (zero scan delay).

Avionics represent 2% of dry mass, excluding armor. Mass 8.84tons.

It is an arbitrarily worse figure than the Terran's 1%.

Internal structure adds 10% to dry mass. Mass 45.1 tons.

Total internal mass is 597.8 tons. Volume is approximately 1800m^3, based on a 0.33ton/m^3 figure, derived from modern jet fighters. This is a cone 8.4m wide and 100m long.

External mass contains armor and point defenses. The Martian Interceptor is in some ways a battering ram: no stealth, high closing velocity, no regard for force disparity. It breaks through enemy defenses and creates an opening for missiles launched by a stealthy arsenal ship. The protection of an armored nose cone is a vital element, one of the reasons it does not have a keel-firing laser. It does not use laser point defenses, relying on kinetic PD instead to survive the one and only pass.

The rotating laser armor must survive the onslaught of three standard 250MW/2m/400nm lasers from maximum to 100km range. It uses a monolithic cone of DLC.

A standard laser penetrates 1mm/s of carbon armor starting from 2750km. This can be considered maximal effective range. At 100km, it penetrates 20.9m/s.

At a closing velocity of 10km/s, the MI spends 275 seconds under laser fire. If the closest approach is 500km, it accumulates 4133mm of carbon removed. If it closes to 100km, it takes 115532mm.

Failing to take out any targets leads to a 24803mm penetration flyby. Taking out one target at the closest intercept reduces this to 20699mm. An interceptor should take out at least one target before the closest intercept, and another after intercept, so 20699mm is the worst design case, and 24803 is the catastrophic failure case. We design to survive catastrophic failure.

The internal volume allows for a 2.41 degree cone. This sloping increases armor thickness by a factor 23.78, reducing requirements to 1043mm.

The cone has an average circumference of 13.2m. The standard laser's spot size averages 0.347m in diameter. An armor shell rotating twice a second can spread laser damage by a factor 76, reducing armor requirements to 13.72mm.

Actual armor requirements lie between 13.72 and 1043mm. In the best case scenario, the interceptor faces a tightly clustered opponent. At worst, the opponent has set up an un-escapable flanking shot. A compromise is necessary. If the interceptor can keep opponents within a 60 degree frontal arc, then it needs 163mm of armor.  

The laser armor masses 655.5kg/m^2, for a total 438 tons.

Kinetic armor consists of two armor belts and two internal bulkheads. The front bulkhead is a 40cm hemisphere 4m wide. The main belt is 40cm thick, 20m long, moving up from the base of the armor cone (437 tons). An inner cylinder with 60cm armor protects the crew (5x4m cylinder, 86.7 tons). The rear belt is 40cm thick, 8.4 m wide, 9.08m tall cone, sloped at 65 degrees.

Front bulkhead is 8.18m^3 and masses 18.8 tons. Main belt is 437 tons. Crew tube is 86.7 tons. Rear belt masses 202 tons. The reactor cap 172.4 tons.

Total kinetic armor mass is 916.9 tons. 

Total armor mass is 1354.9 tons.

Total dry mass is 1952.7 tons.

Internal water tank mass is 1% of water mass.

To achieve 5km/s deltaV, we need a mass ratio of 2.72. This can be achieved with 3359 tons of water propellant, requiring a tank mass of 33.6 tons and 8.4m wide, 60.6m long. Total mass is 5345.3 tons.

Inner drop tanks provide 10km/s. This requires 34201.4 tons of water. The water is held in balloons 60m long, 28.2m wide. Total mass is 39546.7 tons.  

Outer drop tanks provide 10km/s with 253035.6 tons of water. It is held in a balloon 42m tall and 78.5m wide. Total mass is 292582.3 tons.

Martian Arsenal Ship

Description: An unmanned missile carrier that relies on liquid hydrogen boiloff and directional stealth to stay undetected. Cold missiles are delivered by a coilgun. An efficient nuclear-electric propulsion system allows it to catch up with Interceptors and build up an intercept velocity to compensate for the missiles' low deltaV capability. 


Name: Martian Arsenal Ship

Role: Delivering missiles undetected.
Dry mass: 1176.3 tons
Component masses:
15 ton Pulsed Inductive Thruster engines
100 ton nuclear-electric generator
1.8 ton droplet radiators
122 ton coilgun
82.5 ton energy storage
173 ton heat pumps
354 ton missiles
9 ton sensors
6.5 ton avionics
171 ton stealth shroud
68 ton structure
73.4 ton hydrogen tank
Propellant mass: 734 tons
DeltaV: 24.6km/s

Acceleration average: 2.1 milligee
Propulsion output: 1.5GW
Electrical output: 1.5MW
Ammunition: 10000 stealth missiles
Standard laser performance factor: 0

Laser armor depth: 1mm.
Kinetic armor depth: 1mm.

Shape is a cylinder inside a cylinder, behind a hydrogen cylinder. 

Design comments:

Low exhaust temperature nuclear electric propulsion. 100 ton 1.5GW nuclear reactor, 15 ton PIT engines, 5000s Isp (49km/s) exhaust velocity, 30kN thrust (98% efficiency).

3030MW of waste heat is removed by four 1x5m liquid droplet radiators, using carbon-darkened tin at 1200K (emissivity 0.8, 1mm droplets). 146kg of fluid are used in operation. Radiator mass 0.8 tons, fluid mass 1 ton.

Missiles have 1km/s deltaV. They are launched by coilgun to 5km/s.

Missiles are all equipped with stealth capability. 
The penetrator and armor is a Diamond-Like Carbon cone 40cm long, 6.6cm wide, massing 10kg. It mounts a 1kg sensor and communications package. An 8kg, 10MPa (1450psi), 80K tank made of kevlar contains hydrogen gas for use as a propellant in a cold gas thruster, achieving an exhaust velocity of 2700m/s. Propellant density is 33kg/m^3. Thruster mass is 3kg. 9.86kg of hydrogen are required to provide 1000m/s deltaV.

Launch mass is 32kg. Total length is 1.22m, width 0.82m.

Missiles are delivered in 10-missile clusters massing 330kg, including a 10kg 'sabot'.

Coilgun efficiency is 90%. It requires 4.125 GJ to launch a cluster to 5km/s. The projectiles are accelerated at 2000g, reaching the muzzle velocity in 122m of acceleration. Coilgun mass is 122 tons.

Energy is stored in by SMES. Several superconducting loops with carbon bracing hold 4.125GJ in 82.5 tons.

A coilgun generates waste heat too quickly to be removed directly by a radiator. It must be transferred to a heat sink first. 413MJ is released per shot. Non-corrosive liquid helium at 20K is used as the coolant fluid. If it operates at 300K, it requires 278kg of helium gas to be flash-heated. 10 shots (one missile wave) require 2.8 tons. 

The waste heat from the coilgun has to be pumped from 300K to 1200K. This can be handled by 173 tons of heat pumps. They consume 346MW. The power draw of the heat pumps reduces firing rate to one shot per 3.6 seconds. 

Stealth firing and moving can be performed. Waste heat is removed by boiling off liquid hydrogen propellant. The reactor operates on a temperature difference of 3000 to 20K instead of 3000 to 1200K, reducing waste heat by 40% (from 3000MW to 1812MW.  

Heat exchangers increase the propellant's temperature to 100K before ejecting it, thereby removing 1575kJ/kg of hydrogen. Stealth firing consumes 1222kg/s. At the cost of 33.6 tons of liquid hydrogen, the Martian Arsenal ship can drift into position and fire a full missile wave without breaking stealth. 

A standard enemy fleet formation can output 750MW of laser power, and take down between 75 and 150 missiles in the terminal 100km stage. Stealth projectiles evade detection long enough for long-range laser defensive fire to not matter. A wave of stealth missiles detected at 100km range must therefore number at least 150 to get through the target's defenses. 

To defeat an enemy that has survived the first wave, the missiles must deplete kinetic defenses, and attack frequently enough to force lasers into operating at high temperatures. 10 waves against 3 targets requires 4500 missiles. An ammunitions load of 10000 missiles will therefore be sufficient to take out dozens of unaware targets, three aware targets, or one target in a massive simultaneous attack. This will mass 322 tons and require an ammunitions bay 30m long and 4m wide. 10% is added for storage and handling, totaling 354 tons.

No crew.

9 wide-angle sensors: 3 front, 3 rear. Mass 9 tons.

Avionics add 2% to the dry mass, minus ammunition. Masses 6.5 tons.

The 4m wide internal cylinder is enclosed in a 30m wide, 170m long insulated tube. The far end is a radiator emitting at 1200K. The opposite end is open and limits emissions to 10 degrees of the sky. This is Mars's diameter at 38400km. Heat removed is 74.5MW. It allows 25MW of electricity to be generated, allowing one shot per 165 seconds, or 2.5kN thrust at 10km/s exhaust velocity. Mass is 171 tons. 

Structural mass adds 10%. Dry mass of the warship, minus ammunition, is 748.9 tons. With ammunition, it is 1102.9 tons.

The arsenal ship is not expected to come under laser fire. The insulated cylinder is no armor, but acts as a whipple shield. 

734 tons of liquid hydrogen are kept in the nose of the ship. It is contained in a cylinder 10m long and 30m wide, massing 73.4 tons. 

At worst, the Arsenal ship can absorb up to 3.2MW of sunlight side-on. This requires up to 7kg of liquid hydrogen to be boiled off per second to keep the ship thermally invisible. Front-on, this is reduced to 417kW, requiring only 0.9kg/sec.

Terran Advanced Battleship with interplanetary stage

Description: An advanced two-stage warship that travels across the Earth-Mars distance on a large interplanetary stage, then detaches to use its high thrust open-cycle gas core engines to dodge incoming missile waves while a massive laser allows for re-usable offensive firepower. Missile complement its offensive role.


Name: Advanced Terran Warship

Role: Destroying enemy battleships at 1000km distance.
Dry mass: 4729.4 tons
Mass percentages: 32% Armor, 68% Components
Component masses:
200 ton open-cycle gas-core nuclear rocket with reverse thrust
750 ton 3.3GW MHD generator
354.6 ton tungsten wire radiator
330 ton 600MW free electron laser
450 ton heat pumps
33.7 ton optics
115.2 ton PD lasers
10 ton PD casaba howitzers
630 Silver Bullet missiles
33 ton crew and life support
168 ton sensors
30.8 ton avionics
310.7 ton structure
80 ton tank mass
Propellant mass: 7997 tons

DeltaV: 20km/s

Acceleration average: 0.1g
Propulsion output: 100GW
Electrical output: 3.3GW
Ammunition: 10 Silver Bullet Casaba-Howitzer missiles

Standard laser performance factor: 3.2

Laser armor depth: 291mm, triple rotated, sloped.
Kinetic armor depth: 300mm, internal. 

Name: Interplanetary stage
Role: Transporting four Terran battleships to Mars.
Dry mass: 15262.8 tons
Propellant mass: 409879 tons
Component masses:
10000 ton nuclear reactor
437.7 ton liquid droplet radiators
1500 ton PIT rockets
119.4 ton avionics
1205.7 ton structure
2000 ton tank mass
DeltaV: 116km/s
Acceleration average: 0.1g
Propulsion output: 150GW

Design comments:

100GW 200 ton open-cycle gas-core nuclear rocket delivering 8.7MN of thrust at 2039 Isp exhaust velocity. It is 85% efficient, and produces 5GW of waste heat.

66% efficient MHD generators produce 3300MW of electricity and mass 750 tons, producing 1.65GW of waste heat.

A 30% efficient free electron laser produces a 600MW beam at 300nm wavelength, massing 330 tons. It operates at 600K temperature and is mostly immune to reduction in beam quality due to thermal effects. 

1800MW waste heat is moved to 1200K by 450 tons of heat pumps consuming 900MW. It requires 248kg/s of hydrogen coolant operating between 100 and 600K. 
Incredible render of this design's combat stage by u/Grokodaemon.
The laser is focused by a 4m wide adaptive dielectric mirror, shooting through a fused quartz conical window with 96% transparency to 250nm light at when shooting off-center by 5 degrees, up to 99% transparency at further diagonal angles and through the flat truncated section straight ahead. It achieves nearly 90% absorption of visible light wavelengths. It can survive a 900K temperature increase. It is 32mm thick, 4m wide and truncated at 12.5m length, sloped at a 4.5 degree angle to achieve 400mm relative thickness. The truncated window is 2m wide. 

The mirror masses 0.5 tons. The cone masses 8.3 tons. The warship is equipped with 3 onboard replacement cones and mirrors.    
Find more here.
Comments: This cone serves to protect the main mirror. Acting as a blacklight, it prevents the Martians' lasers from attacking the mirror directly, while letting shorter UV wavelengths through. The 400mm straight-head thickness prevents penetrations from regular-sized kinetic impactors. Firing diagonally is preferred, as it decreases the laser power being absorbed. The truncated window is a 3.14m^2 weakpoint.

The cone can absorb up to 24MW during operation. This would bring the cone to dangerous temperatures within four minutes, with localized failure along the beam path. A liquid hydrogen cooling system is required to absorb this heat, with a throughput peak of 8kg/s, operating between 200K and 300K. 

3.45GW of waste heat is removed by tungsten tensile wire radiators operating at 1200K. The wires are 1mm thick and 10221km long, massing 154.6 tons (15.1g/m). They are extended into 100m wide loops in space. At 70m/s, they maintain a 10G centripetal force and will cool down to approximately 360K in 1.3 seconds. At 0.057m separation, they only block 0.6% of each other's radiating area. At 3mm separation, it is 10%, but they risk touching each other. 

The radiator loops can be stacked 50m long, in four sets, extending 100m into space. They are mostly resistant to laser fire and kinetic damage due to their small cross-section. 200 ton replacement wire increases survivability.

Anti-missile protection relies on point-defense lasers. 6x6 turrets with 60 degree traverse, massing 3.2 tons each, putting out 100MW maximum each and can take down 384 missiles. 10 Casaba-Howitzer wide-angle nuclear charges, massing a ton each, can be used to clear out a 30 degree cone of missiles at 100km range. However, this destroys the radiators. 

The Advanced battleship is equipped with another 10 'Silver bullet' missiles with 1200kg Casaba Howitzer 10000km range warheads, 4439kg of carbon armor (400mm thickness, angled at 30 degrees) and 10km/s deltaV with hydrolox propellant. Each missile masses 57.3 tons, for a total of 630 tons. 

The crew count is 9 individuals: 1 commander, 3 tacticians, 2 mechanics and 3 programmers. They inhabit two 10-ton habitation rings rotating to provide 0.3G. Life support is fully closed-cycle, with 3 tons of equipment. Control is through two CIC units massing 6.5 tons. Included is radiation shielding and a 20mm armored shell.

The Advanced battleship has a heavy sensor suite. It has three rings of 36 narrow-angle sensors, and six lines of 10 wide-angle sensors along the flanks. Total mass is 168 tons.

Comments: Terrans fear the roving stealth missiles deployed by the Martians. They believe Martian space to be infested by thousands of smart groups of such missile. A 360 degree view sensor suite is necessary to prevent a surprise attack by cold projectiles.

Avionics are 1% to the dry mass, adding 30.8 tons.

Structural mass is 310.7 tons, increasing total internal dry mass to 3417.5 tons.

This fits inside a cone 200m long and 14m wide. It is sloped at 2 degrees.

Comments: Battleships match orbit with their target and burn it down with their lasers. Zipping past would expose them to a cheap kinetic attack. Attack runs include closing the distance for a targeted laser strike, then pulling back on reverse thrust (to keep the armor cone pointed at the enemy). Armor requirements are calculated based on how long the armor must survive laser fire.

A standard fleet equipped with three standard lasers can remove 3mm/s of laser armor at 2750km, increasing to 60mm/s at 1000km.

Based on the setting's objectives, we can set the 'time to kill' duration as 1 hour at 1000km. This leads to an armor thickness of 216000mm. 

The armor cone's 2 degree slope can reduce the armor required by a factor of 28.6. However, battleships with low acceleration and locked down trying to stay out of one target's weapons' effective range can easily be flanked by another target. This means that the slope cannot be relied upon. 

In a typical 3 vs 1 scenario, two laser ships bracket the defender while a third rises above the plane and attacks from an angle. The Advanced battleship's cone would reduce one laser's penetration rate by 28.6, the second by 15.5 (the first two targets are 100km apart for mutual anti-missile defense) and the third by 1.006 (90 degree flanking). The sum of their penetration rates is 21.48, leading to a thickness requirement of 78748mm.

The Advanced battleship uses a nestled rotating armor scheme. Average diameter is 7m. Rotation is once per second, leading to the laser being dragged along at 22m/s. With a second shell under the first, the laser can be spread out over 44 meters of armor per second. With a third shell, it is 66m/s. 

At 1000km, spot size is 0.244m in diameter. The armor rotation reduces laser effectiveness by a factor 270. The armor thickness required is 291mm.  

Each shell is 97mm thick. Laser armor mass is 981.9 tons.

Kinetic armor is internal. A 4m wide, 60m long 'citadel' tube protects the crew during combat, life support equipment, backup power generation and redundant avionics, as well as the vulnerable the free electron laser. It is 300mm thick and masses 520 tons.

Internal dry mass is 3147.5 tons.

Armor mass is 1501.9 tons.

Total dry mass is 4649.4 tons.

A deltaV of 20km/s requires a mass ratio of 2.72. This requires 7997 tons of water, contained in a cylinder 14m wide and 52m long. It masses 80 tons. Actual deltaV is 19661m/s. Launch mass is 12727 tons.

The interplanetary stage permits rapid movement from Earth to Mars. It moves 4 battleships as a payload of 50908 tons.

A 10000 ton 150GW nuclear electric reactor produces 300GW of waste heat. It is removed by two 60x60m droplet radiators operating at 1200K, massing 285.7 tons. They contain 52 tons of liquid tin, and 100 tons of replacement fluid are kept onboard.

For propulsion, 58km/s exhaust velocity PIT engines massing 1500 tons produce  5.17MN of thrust. 

The interplanetary stage's dry mass is 13151.7 tons, including avionics and structural components. It departs at 64059.7 tons.

Liquid hydrogen is the propellant. 409879 tons are required, to be contained in four spheres of 132m diameter. They mass 2000 tons.

DeltaV is 116km/s. It would take it 87 days to cross the average 225 million km distance between Earth and Mars.          


  1. Feel free to ask about the numbers or design choices made.

  2. If you'd like, I could put together some 2d representations of these if I have time on Saturday. I offer some of this work below as examples of other spacecraft ideas (and evidence I can actually draw).

    1. I don't doubt your skill. I especially like how you pulled off the 2D curves for cylinders on the Velocitor. Feel free to use these designs as you wish.

  3. I applaud your worked example. I start from a different set of assumptions based on Rocketpunk Manifesto's discussions, so any worked examples I come up with will be different, but I think I can import some ideas.

    If the Terran navy believes stealth in space is an issue, they need to eliminate the possibility of the enemy using it on them. They can either boost large nuclear bombs well ahead of the fleet to detonate ahead of the constellation and illuminate the area with a massive pulse of radiation in all wavelengths (from X-ray to Infrared) to be analysed by sensitive scanners in the constellation. If that is not advisable for tactical or treaty related reasons, then one of the laserstars will be using a "wide" beam to illuminate the enemy Hill sphere from a light hour to a light day away, looking for anomalous reflections. This can be thought of as using an active sonar.

    I also see a fully developed constellation assuming a formation one light second in diameter, seeded with thousands of sensor drones to provide a highly detailed 3D view of the battlespace. This also allows the laserstars to attack from multiple angles (since I assume a Ravening Beam of Death [RBoD] laserstar, this can happen from a light second away , with scorching attacks coming from even greater distances).

    This would also change the Martian strategy to building massive laser battle stations on Phobos and Deimos to take advantage of the natural armour protection and huge heat sinks available, while seeding Martian space with Kineticstars to break the attacking formation with thousands of Soda Cans of Death (SCoDs) targeting the Terran laserstars.

    Naturally the battlespace will be awash with ECM, decoys and nuclear pumped weapons as well, making space warfare a true exampel of 99% boredom and 1% stark terror.

    1. Just quickly reposting Francisdrakex's Scod illustration here.

    2. @Thucydides: Nice to see you back here!

      It is difficult to eliminate stealth through active means, especially at long range. The best 'illumination' would be a device that can make sure the most amount of watts can reach the potential target, so it can be bounced back. Ultimately, this will be the main laser weapon. With the largest focusing array, megawatt output and shortest wavelength, it can serve double function as an active sensor and eliminate the need for an equally powerful radar.

      A terran battleship can put out 600MW of laser power at 300nm. It is focused through a 4m diameter mirror. Let's say the threshold for detection is 3e-15W/m^2, considering the amount of background noise there is in a warzone (a 100 times more than my realistic figure for a cold sensor and blank background).

      A floating mirror might bounce back 99% of the incoming light, a warship only reflects about 50% or less, but a stealth warship with a coating of Vantablack will absorb 99.99% or more.

      3e-15 is 3e-25 in MW/cm^2, so using the laser calculator, I reach this value at about 2500 trillion km.

      However, the signal has to bounce forth and back, so the actual distance is 1250 trillion km.

      The reflectivity of a regular warship can be 0.5, so distance is reduced to 975 trillion km.

      This assumes that all energy is bounced back to the transmitter. Compared to a cylinder, a flat plate returns 4/λ more energy, with λ being the wavelength. λ here is 300nm, so a cylinder produces a signal 13.33 million times weaker. This reduces the distance by a factor 3651, to 267 billion km.

      At such a distance, the spot size is 357 million km in diameter and covers the entire Martian system.

      My unassuming conclusion is that an active beam can spot warships from anywhere in the solar system... but is still blind to the stealth warships that have extremely low reflectivity Vantablack coverings.

      A formation a lightsecond in diameter will be able to flank any closely-defended constellation, but beam effectiveness at the power levels mentioned is terrible.

      As for deploying sensors into the battlespace, I agree. They would provide excellent information on the regular warships. However, no sensor will detect a perfectly back warship with a 3K surface, slowly exhaling hydrogen at 20K. Its even worse with 'thousands of sensors', as they become necessarily small and therefore even less effective at detecting low-temp warships.

      Considering all that, I'd assume that Vantablack and radar-absorbent coating will become standard in all warships, even those that are not hoping to achieve thermal stealth.

      Ravening Beam of Death laserstars attacking from 1 lightsecond are practically immobile. I don't see them as ever working, for two reasons:
      -Stealth projectiles will take it out the moment it fires and reveals itself, if it could ever hide its bulk in the first place. It cannot accelerate out of the way of thousands of dumb, low-deltaV projectiles.

      -Planetary defense lasers will always outperform RBoDs. If you suspect the Terrans have deployed one, then you can always pull back within range of your planetary defenses. You'll suffer the swing-by-and-dump-kinetics-at-interplanetary-velocities bombing, since you did not intercept the fleet early, but you will catch the RBoD. And due to it being mostly immobile, it can be taken out by high-acceleration spaceships of much lower mass and much shorter-ranged weaponry. It is not cost effective against targets that accelerate at 0.1 to 1.8G.

      I did not think of Phobos and Deimos as perfect laser bases! Good idea.

      Nuclear pumped weapons are presumed to be extremely expensive.

      Considering that all regular, 'hot' warships are afraid of a sudden strike by stealth kinetics, it's mostly 99% terror and 1% anger. 99% when scared of a strike, constantly changing vectors and chasing faint signals barely above noise level, and 1% anger when they reveal themselves and you get to obliterate them with superior weaponry.

    3. @Geoffrey S H: Warships carry thick armor belts made of near-physical-limit materials. Kinetics impact at velocities where hydrodynamic penetration is still important. Combine the two, and fragments from a SCoD will never amount to more than pitting and scarring. You need long-rod penetrators.

    4. Yes, I always thought Rocketpunk's assumption that armour would never work was a bit odd....

      I presume there would be no use for such a device on a high velocity pass (when anything could be used to mission kill a craft)? Firing out buckshot on suggested trajectories would probably be cheaper...

    5. Rocketpunk manifesto assumed gigajoule impacts.

      The balance between armoring a projectile and using that same mass to launch a new projectile is a simple cost/benefit ratio. The cost is mass and the benefit is increased survivability, or alternatively, increased number of projectiles on target.

  4. The use of thousands of sensors, even very small ones, spread out across a one light second diameter circle, would be the equivalent of having a mirror of that diameter, using interferometry. And thousands of probes can be optimized over a huge speed of wavelengths, effectively giving the constellation a multitude of high resolution sensors. This is a variation of the Server Sky idea developed by Keith Lofstrom: The constellation is also getting high resolution data from a similar constellation of sensors in Earth orbit, so there is triangulation with a huge baseline in addition to multispectral scanning.

    I actually don't see stealth in space as being an option, unless you can somehow beat the laws of thermodynamics. Even the shielded ships wrapped in a cylinder of liquid hydrogen will be revealed by the boil off at 20K (I am assuming there is no refrigerator with its own heat source to reveal itself). You may have a more difficult time to get a lock, but you will get one.

    As for the RBoD, the paradigm is different. You are not as worried about tactical mobility since your effective engagement distance is the same as from Earth orbit to the Moon, and while Luke Campbell had given the parameters of the RBoD having an electron beam accelerator of a kilometre in diameter, wakefield accelerators and deliver that sort of energy in a matter of meters, so the RBoD will be from airliner to supertanker sized, not the size of a small asteroid.

    I am actually getting more inclined to think of nuclear pumped devices as being the way to go, since even though they are expensive, they are not as massive as powerful laser generators, rail or coilguns and other weaponry. A small spacecraft carrying a battery of Casaba howitzers and nuclear pumped weapons may actually mass far less than an equivalent laserstar or kineticstar, yet still have as much firepower. Lower mass is the key to manoeuvre in space if you are looking for tactical space manoeuvre. In that case we are back to the sorts of distances in your example, but can reorganize the ships to have less mass devoted to heat rejection and electrical generation.

    It really is all about the starting assumptions.

    1. The small sensors will give good resolution, but catching a stealth ship relies on minimum sensitivity. Not sensitivity over time, but with the amount of photons captured per second as the ship moves across the field of view. At 3K, you do not emit more photons at the background. I think, in practice, sensor sensitivity will not detect you even if you emit at hotter temperatures (20, 30K?).

      Liquid hydrogen does not beat the laws of thermodynamics. Boiling a liquid keeps it at the boiling point, so with nothing running onboard, a tank of liquid hydrogen will keep the ship at 20K. If you use a heat pump, you can go lower, slipping under the metaphorical nose of your enemy. The cost is that the amount of hydrogen you have to throw overboard is multiplied. Going from 20 to 3K temperature, with near-100% efficient heat pumps, requires 5.7W to be removed for every 1W absorbed. If we include 20% energy generation, it increases to 28W. Sounds bad, but you have thousands of tons of cold hydrogen sitting in your propellant tanks.

      A very slow missile ejected at 5km/s by a mass driver, cooled by liquid hydrogen boiloff, will cross 300000km in 16.7 hours. In that time, a milligee laserstar will generate up to 588m/s of deltaV. This can be matched by a cold gas thruster. A fast missile accelerated by a staged booster to 15km/s will cross that distance in 5.6 hours. With hypergolic thrusters, the warheads only need a mass ratio of 1.06 to match maneuvers with the target. The missiles can be literal blocks of armor with a small fuel tank hidden behind them. Tactical mobility is important, because even a fast missile cannot catch a 0.1G warship (it can put out 19.62km/s).

      Wakefield accelerators might be the final nail in the coffin for exciting space combat. However, the only article I found that mentioned wall-plug efficiency cited a 0.1% figure.

      In the end, its all up to the assumptions. Personally, a setting which ends up with pellet guns as the primary weapon system would be great. Nuclear shaped charges criss-crossing the battlespace and leaving behind beautiful glowing trails of plasma? Be my guest!

      As for the lower mass comment, I disagree. Propulsion, especially NERVA-type propulsion, can generate incredible amounts of thrust with little to no need for radiators. I briefly considered a dual propulsion system for warships, I might have to go back to it and actually work out how much a 200 ton solid-core nuclear rocket would cost a ship, in return for a few meganewtons of thrust.

    2. Resolution in itself is not sufficient. You also need sensitivity.

      IIRC, a 1W/m^2 low frequency IR emission source will yield a MAXIMUM of 1 photon/m^2 at a range of 1 000 000 km. Higher frequency photons will be even rarer. This means that a source colder than 90°K (producing less than 1W/m^2 under maximum blackbody radiation emissivity) will not be physically visible at over 1 000 000 km range. At below 9°K this minimum detection range decreases to 10 000 km.

      Another factor is that, regardless the range, it is physically impossible for a sensor to detect waste heat sources that are cooler than the sensor platform itself. This is due to the fact that the platform will be flooding its sensors with its own blackbody radiation, at exactly the wavelengths it is trying to detect. Thus, you will only detect the 9°K heat source if you yourself (the sensor platform) are cooled to less than 9°K, even if you are only 1km from the source. Note: this only applies to waste heat emission. If the source is emitting ANY electronics pulses (RF energy) of sufficient strength, it will probably be easily detectable. Also note that RF emissions produce many orders of magnitude more photons for the same energy load, so you will see such emissions from billions of km... potentially. KEEP THOSE RADIOS OFF!!

      Finally, although you only need three platforms at the appropriate spread for interferometry to produce an excellent image, this ONLY works if all three sensors are covering the same point in space at the same moment of time. FOV comes into play here. If you want full coverage, you will need not just thousands of sensor trios, but HUNDREDS of thousands. This is smply not practical. Therefore, such interferometric telescopy should be limited to get tactical/strategic info on confirmed threats.
      Also, this means that all platforms must be able to independently detect the target. If the target is cooled down to less than 90°K, your 1 light-second spread (298 000 km) will not be able to resolve the image, even if it IS close enough to one of the platforms to be detected.

      Okay, one last note: I have been assuming platforms with collectors on the order of single or double digit meter diameters, and single/double digit seconds frame exposure times. 100+m collectors will increase the range by about a factor of 10. More precisely, any increase in range will require the square of that increase in collector diameter and/or in single frame exposure time.

    3. Just because I haven't seen this discussed yet, has anyone discussed occlusion sensing? Detecting the spacecraft physical form occluding stars, sunlight, or reflected planetary light? Using this method of detection, how many sensors in how wide an array would be required for what sort of detection ranges? Occlusion for both the spacecraft body (total occlusion) and the expelled gas cloud (dimming and refraction). I ask only because I haven't seen anyone mention it, and thermal detection seems to have some reliable countermeasures by using very large quantities of liquid hydrogen.

      Speaking of liquid hydrogen or helium, has your mission endurance calculation incorporated leakage from the storage tanks or is it negligible?

    4. The dimming/refraction caused by the grams per second ejected by the Hydrogen Steamer is quite negligible. Even if sensors could pick it up, it would be dwarved by the millions of tons of hydrogen ejected into space by regular space traffic (which can be assumed to exist if Hydrogen Steamers are roaming interplanetary space).

      A spaceship running in between a sensor and a star in the background will certainly be picked up. However, the problem is transitioning from 'oh there's something along this sensor's line of sight, travelling along an unknown vector' to 'there's an enemy spaceship at this location, headed in this direction at this velocity'.

      Hydrogen leakable is pretty hard to estimate as it depends mostly on manufacturing tolerances and defects. Interplanetary Mars missions using Zero-Boiloff tanks do not make a big deal about hydrogen leakage even after holding the cryogenic propellant for 9 months or more, so I did not include it in the calculations here.

    5. Specifically for occlusion by the spacecraft: once the object has passed in front of two radiant celestial objects, you will have their velocity and heading, but probably not their position, from a single sensor. You'll also know from the dimming/blackout curve a single one-dimensional slice in the objects cross section, as well as the opacity (probably complete) of any given point. Celestial radiants are generally treated as point sources because of their extreme distance, hence the one-dimensional slice. The difference between the first and the second one-dimensional slice (if any) could possibly tell you distance, I'm not sure.

      For two sensors, of course, once the object has occluded two celestial radiants for both sensors, they would have four different one-dimensional cross sections that should give a good size estimate of the object. It will give you a high certainty in the velocity and heading. If the readings are separated enough in time, they will tell you if the spacecraft is currently accelerating and at what thrust - knowing that the object is accelerating when you cannot thermally detect an exhaust plume will tell you immediately that it is a stealth vehicle, which means it is a military target. For every extra sensor trained on the anomalous region and extra occlusion event, the certainty about exactly what sort of object it is in cross section, whether it is accelerating i.e. a vessel, and exactly where it is and where it's going will increase.

      I don't know how sensitive a sensor platform would need to be to detect these sorts of events, but I suspect that because they are bright point sources they wouldn't actually need to be that sensitive, maybe even just one pixel per celestial radiant, although that wouldn't be very reliable. You'd want maybe at least four pixels per celestial radiant to minimise false alarms due pixel malfunctions, and I don't know what sort of sensor platform that entails. I have a suspicion that if you were mass producing these sort of sensors, they could be quite cheap and cover the whole sky range, and would obviously serve many other functions than just detecting stealth vessels. Many of them wouldn't need to be super-duper cold, either, although you'd want some that are cold, high sensitivity sensor platforms. Most could probably be radar and visible light sensors.

      The stealth defence would be to shape your vessel like an asteroid, probably the lowest density one you can get away with, and to strictly minimise the times you accelerate. That would really, really restrict where your stealth vessel could operate, though, because in a setting where these sensor platforms exist every notably large entity in the solar system would have been logged before the story starts, with new logged objects becoming rarer and rarer over time. Any sudden spike in asteroid-like objects occluding stars roughly in the same size range as vessels and heading towards roughly the same volume would be an anomaly, to say the least, and active sensors would start looking for them immediately. In fact if I was in charge of that military, I'd have standing orders to immediately fire a short kinetic barrage at every such object, on a trajectory such that the rounds will end up hitting a lifeless rock or the sun if the object moves unexpectedly, if I was a military officer for a solar system power. I'd probably broadcast that that's what I'm doing, too, to discourage anyone trying in the first place.

    6. It seems that occlusion suffers from the same dispersion problems than lasers, so even against a point source, it won't work after a few hundred thousand km. Which may still be useful to try and detect them in a planetary Hill sphere.

    7. @lucubratory: Actually, if you don't know the target's position, and only collect data from a single sensor, two, three or even more occlusion events won't tell you the velocity of the object you're tracking. You'd need a parallax measurement from multiple sensors... which is tricky for objects within the solar system.

      The occlusion events of the stealth ship would also have to be distinguished from the myriad other spaceships and natural objects in the same plane.

      @eth: At very short distances, the target spaceship will be covering up multiple stars at a time, which gives you more data.

    8. True, so you'd need at least two sensors to make those sorts of confirmations, and yep occluding multiple stars would be well into where even a small sensor net should have detected them days ago.

      I suppose it does depend on the setting. If we presume that there is an authority which is reasonably threatened by stealth vessels, we can probably also presume that it will put many hundreds of thousands, probably millions, of sensors around the solar system. If they all cost as much as a top-tier smartphone does at that time (which should encompass the required computing, image processing, and communication capabilities), plus a small amount for a solar power or nuclear battery system and some form of radiation protection like redundant capacity or shielding, plus a percentage of them having a large lens and sensor and cryogenics system for high sensitivity, an unfurlable wire-array that doubles as both communications antenna and radio receiver, etc, it should be possible to roll them off production lines in millions for a very small fraction of the military budget at that time. Doing that alone would probably cost proportionally less than what modern superpower militaries spend on surveillance and sensor technology today.

      I think the most limiting factor might be the lenses, mirrors, etc. A perfect large mirror currently costs a lot more than even many hundreds of smartphones. The cost would go down with a real assembly line and might go down with zero-g manufacturing, but I can't predict whether it would still be the most expensive part of the whole shebang.

      Anyway, interesting thought experiment. I'm still unconvinced that stealth in space can be hard sci-fi because I think it's a big stretch from the plausible mid-future to create settings where the technology you describe is possible but much cheaper sensor nets are not. I suppose a setting with very little or no computing could work, which could be interesting, but I have trouble even conceiving a spacegoing civilisation without computing.

  5. This is an interesting debate.

    Stealth in space would really require quite a bit of hand waving. Consider that astronomical sensors like the WMAP were used to map the cosmic radiation background and reveal the tiny fluctuations left behind by the Big Bang, and you can see that even a totally passive "reefer" at 20k will stand out quite brightly agains the 3K background. The issue is compounded with heat pumps, because while the "cold "end can be brought down to 3k, the "hot" end will be correspondingly warmer. The enemy ship will see the bright spot of the heat pump, and even if the resolution isn't great enough to reveal the rest of the missile or bus, there is now something to put a target number on. And the Laserstar can do a complete sky survey in 4 hours using the on board high resolution scanner. Multiple ships in the constellation will be running sky surveys starting at different points, so sneaking up on a constellation will be very difficult. It does provide a tactical distraction, however. The constellation will have to devote some resources and reserve some firepower against stealth missiles and busses while still keeping an eye on everything else.

    Smaller and lighter spacecraft do provide an advantage, much the same way an F-16 has higher performance in some aspects (acceleration, instantaneous turn) than larger and heavier aircraft like the F-15 Eagle or the SU-30. While there are obvious differences between spacecraft and fighter jets, the ability to suddenly accelerate at a higher rate than your opponent (to the ability to have the same performance using smaller engines) could provide that edge, especially if the contest is geared more towards missiles and kinetic energy impactors.

    1. Thanks for starting it!

      Cosmic radiation comes from fixed sources over very long periods of time. You can use a lower-than-required sensitivity sensor to detect very faint sources of energy by letting it accumulate photons over time. A spaceship will zip across your field of view, so you need to have your instantaneous sensitivity match or exceed what is required to detect the target's emissions.

      A heat pump removes heat and makes the 'hot sink' hotter. However, if the hot sink is sitting in liquid hydrogen, then it won't exceed 20K until you run out of propellant.

      Resolution allows you to get a precise measurement of the location of the target. Incredible resolution is never really required when warships measure a hundred meters long.

      The 4 hour sky scan figure is indicative of technology from... 1997. The origins is from figures given by a radioastronomer by the name of John Schilling. In practice, your sky scan depend on how many pixels you can process per second. If each pixel equals 1m, then you need incredible calculation time. If the pixels are hundreds of thousands of km wide, then you can scan the sly in a matter of hours with 1998 tech.

      I don't believe sensors consume enough power to matter, especially not to warships that produce a megawatt on backup power.

      I think the smaller/lighter spacecraft distinction does not make sense. The same acceleration can be produced by two spacecraft of wildly different size and mass. I think propulsion fraction is a more useful figure, but it still depends on how good the tech is.

      Acceleration still matters in a laser-dominated battlespace, as the warship with the higher acceleration can maintain a specific distance from its target and maintain its range advantage, or close the distance if it has shorter ranged weapons.

    2. Here,spectrum becomes important. Microwave frequencies produce much higher photon fluxes than IR, which greatly increases detectability. MF emissions do not suffer from the waste heat problem, as blackbody waste heat emissions (IIRC) generally do not produce microwaves (thus, the platform does not need to be cooled below the CBR source temperature). Finally, the CBR source contains a phenomenal amount of energy, even if that energy is widely dispersed.
      No one is doubting that a 20°K microwave source will stand out above a 3°K microwave source. The problem is that blackbody emissions emit at frequencies that are difficult to detect at low energy levels.

      As MatterBeam alludes, even 2000°K sources are undetectable if you have an insulating barrier between the source and the detector. Granted, there is always the tendency for sources to achieve thermal equilibrium... but this often takes time, especially with a heat sink absorbing the thermal load (so, no, the enemy does NOT see any bright spot).

      The 4hr sky scan is based on certain assumptions that really don't hold for distance observation. First, it wasn't taking into account physical limits of resolution over interplanetary distances (a problem solved through interferometry, but only if all detectors are directed toward the same source at the same time, and if those detectors can actually detect the source). It also does not take into account the exposure times required for detectors to register emissions from ow level sources (for instance, WISE/NEOWISE required 10 sec exposure times per frame). Third, it does not take into account whether or not a source can actually BE detected.

  6. Actually, larger ships have less acceleration: engine power depends on their surface, while ship mass depends on their volume.
    We can see that in Children of a Dead Earth, where small missiles routinely have several g of acceleration, while it is extremely rare with large capital ships, even using the same technologies.

    1. Ship performance does not follow square-cube law... and CODE's massive nozzles are ridiculous.

      A 1000 ton spaceship can accelerate at 1G with a 24.5GW nuclear thermal engine. Using water as propellant, it produces 9.81MN of thrust at 5000m/s exhaust velocity. The engine masses 62 tons, which is 16.7% of the dry mass. This is based on power to weight ratios from the 1968 NERVA project (2.3kg/MW).

      It throws out 1960kg per second. This is equivalent to 3.81 shuttle SSMEs or 31.2% less than an Apollo F1. No ridculous nozzles that mass more than the engine itself.

      A 10kT spaceship with 245GW engine throws 19600kg/s and accelerates at 1G. Just before burnout, it reaches 2.7G. It need the equivalent of an F1 nozzle, but 2.7 times wider.

  7. A 2000m squared radiator does massively dwarf out any interceptor drawing (as I am now discovering in my attempts to capture its aesthetics in MsPaint). I may have to have two drawings to show the interceptor in full detail. May I just double check that the 2000m sqd figure for the Martian Interceptor radiators is correct?

    Thanks for these figures, they are nice and clear, getting the dimensions drawn out has actually been quite simple. :)

    1. 2000 meters squared is the total area. It can be divided into 4 panels of 50m long and 10m wide.

      The whole warship is 160.6m long.

    2. *smacks head on desk* Well that WAS a stupid mistake to make! Total brain failure on my part there, sorry!

    3. Don't fret. You're already going above and beyond by trying to draw these ships.

    4. Oh don't worry, I was chuckling to myself at that mistake, thanks to a potential successful job application, I'm in a good mood.

      Thanks to the ability of ms publisher to measure things in 'cm', it is proving fairly easy to size things at the moment. The cones might be difficult due to the angles but I'm confident it can be sorted.

      Slightly off-topic, but if you want an AI related posting (I seem to remember it being mentioned) you might want to read up Yuval Harari's 'Homo Deus'. Its less innovatory than many people are saying, but it collects a lot of very good articles and examples of automation together. I can't think of a better text out there currently.

    5. since geoffrey mentioned "homo deus" as reading material for a post about artificial intelligence, i felt like suggesting two further books: nick bostrom's "superintelligence: paths, dangers, strategies" and eliezer yudkowsky's "from ai to zombies" (this latter one is more general about rationality and cognition, but certainly worth a read).

    6. Thank you both. I've downloaded audiobooks for Geoffrey SH's recommendations, and I will do the same for Superintelligence and From AI to Zombies.

  8. While not a vast game changer, the rotating fluidized bed NTR designs Atomic Rockets has placed on the engine list provide a much greater thrust/weight ratio and a far more compact package than solid core NTR's, and seem far more plausible than various sorts of gas core reactors. The idea of being able to "dump" the hot fissile material out the back provides an interesting option to dealing with reactor accidents or even simply cutting the throttle (the fissile elements will remain white hot, but without reaction mass flowing past there is no meat to get rid of the heat).

    And if this was developed in secret, the other side will still have fire control tables based on the T/W and acceleration if solid core NTR's.......

    1. That sounds great. The figures I used for the Solid-Core nuclear rockets are about 5x better than the actual specific power data obtained from 60's and 70's testing of such rockets.

      A rotating fluidized bed NTR could help achieve that performance.

      I am doubtful of the utility of dumping the fissile material, but the high surface area could instead be used to quickly flood the material in neutron poison for instantaneous shutdown of any reaction. It can then be washed away.

      As for fire control tables... well, you can vary your thrust randomly even with a standard NTR, so I don't think it will be much of an issue.

  9. Something I've not seen mentioned in this series (Apologies if it is here and I missed it) is the role of Carriers. Do you see larger carriers capable of transporting large amounts of personnel being a part of warfare? I've generally operated under the assumption some kind of vessel to carry your space marines for landing on Mars would be a necessity, and possibly be used at the centre of a task force for command and control/ co-ordination of the other vessels like a modern-day carrier.

    Do you think vessels that large and mass-heavy would have a place in interplanetary warfare?

    1. First of all, hi! Please use any name, even without registering, to distinguish you from other anonymous posters.

      Carriers, personnel carriers, depend strongly on the setting.

      In the setting used for 'space warship design', friendly troops are either already in place, or they're not. It takes a fleet over a month to reach the closest planets, and for comparison, the war in Irak was over in a week or so.

      Also, I do not see them as being military ship being escorted within a fleet in the first place. They can be individually targetted and destroyed by accurate lasers. If they are armored against lasers, they'd probably cost as 75-90% of a full military ship anyways!

      This is an important consideration when winning the war in space can make the presence of ground troops pretty irrelevant! If I can launch nuclear weapons and kinetic strikes from orbit, I can force demands while my troops arrive slowly from another planet.

    2. I think nuke weapon strikes wouldn't necessarily be a game-over factor, with a sizeable degree of bluffing involved. For example, would a commander necessarily want to risk messing up the Earth if it's the only planet capable of supporting life? What if the planet is balkanised and nuking a city could piss off the neighbours?

      Likewise, there's also stuff like bunkers, submarines etc that could be basically immune to orbital strikes. I don't think (in the case of a planet like Earth) you'd necessarily be able to just negate ground combat entirely.

      Plus, if Earth is controlling Mars or something, would you not want some ground troops to act as a police force to keep the locals in line and fight off rebel scum?

      I don't think that "Obey me or I'll render your planet a lifeless rock" would go down with the international community, particularly if you want to look good.

      I suppose it's, like you said, up to the individual tech level and setting.

    3. Nukes are kind of only useful on Earth, Venus and maybe Titan. Everywhere else in the solar system has very thin or no atmosphere at all. This means lasers can strike the ground with impunity.

      If they can hit targets at thousands of kilometers, they will have no problem doing precision destruction from orbit.

      Kinetics are equally deadly. They become accurate weapons that hit at orbital velocities.

      Using the two, you can surgically remove bridges, pop habitats and melt factories until the civilian population surrenders. If their leadership is hiding in a submarine while the country is set on fire bit by bit, a new leader will be elected. I doubt we will be able to install millions of inhabitants inside military-grade bunkers.

      As for the troops on the ground, it is more of a political question. In today's world, there are military outposts with UN or US soldiers in them, ready to intervene with backup from the closest navy or airfield. However, it is too sensitive to station them on Chinese land.

  10. Hi Matterbeam - great blog, it is really useful for a project of mine!

    Looking at these examples I thought I would investigate how a coilgun would fare in this scenario. Apologies if I have made some poor assumptions - I am still trying to digest the numbers!

    Assume a 1 on 1 engagement using your Terran ship and a Martian ship at a closing speed of 6km/s. I assume the Martian ship begins firing 4.5kg steel slugs at a rate of 0.5Hz and a velocity of 20km/s for a closing velocity of 26km/s, which delivers an impact energy of 1500MJ. While this will not penetrate the armour, I don’t think it needs to as the laser mirrors, radiators, electronics etc are going to suffer from shock effects and the ship will be disabled. In reality, the Martian ship fires a variety of different sized projectiles at a variety of velocities to achieve similar time on target.

    The Terran ship uses a W band targeting radar at 110GHz but the diameter of the projectiles is so much lower than the wavelength that while the radar can see them, it is unable to target them(!). My calcs are showing that at 500m the resolution of the radar is 0.5m but the laser spot size is less than 1mm. The projectiles are painted black so are invisible to LIDAR and some of them are ignited as flares during final approach which dazzles sensitive IR equipment that are trying to spot the cool (300K) incoming objects.

    The Martian ship begins firing at 10000km and Terran ship detects the launch of the projectiles immediately on radar. It begins to evade at its full acceleration of 0.1g and assuming it does not change direction will eventually end up 1400km from its initial projected position. This means that a simple engine on the projectiles only needs a deltav of 820m/s and they are guided to the target by the firing ship.

    With a 28 minute engagement time, the Martian ship can fire 830 times (for a total of 3.75tonnes) and this cloud of projectiles can be thickened with lightweight decoys to keep the target guessing. Based on a 4m projectile spacing, the cloud diameter is 130m and this means multiple guaranteed hits as long as the guidance is not completely jammed. The casaba PD missiles can destroy the kinetic swarm but at the cost of crippling the target ship anyway and it can be finished off with… more projectiles.

    What do you think? I am only just learning all this stuff so feel free to pick holes in my idea!

    1. Hello and welcome! Feel free to ask me more specific questions for your project either here or by email.

      Now before we start, we must consider the fact that it is quite unrealistic to have these spaceships duke it out 1 vs 1. I mentioned in A Constellation of Warships posts that fleets of at least 3 ships are the smallest formation, as all three axis of manoeuvre can be covered.

      Nonetheless, let's include a 1vs1, then imagine a full 3vs3.

      The Martians here are defenders. They don't travel across interplanetary space. The Terran battleship has a high Isp stage that delivers it to Mars. So, it is more likely that it is the Terran battleship that arrives at a high closing velocity. Looking at the deltaV of the Terran's interplanetary stage, we can assume that in case of a war, it would come looking for a fight at about 30 to 50km/s. It slowly down a little before intercepting Martian defenders.

      Either way, the two ships close at a velocity of much over 20km/s.

      The Martian Arsenal ship have a coilgun that lobs a missile on the Terran's path. The Arsenal ship fires 10x clusters of 1km/s deltaV shells at 5km/s. This means closing velocity for martian missiles is 25km/s+ and they can manoeuvre an additional 1km/s. Terran ships have a stock of missiles that can all be fired at once, to save the weight of a coilgun (replaced with a giant laser).

      The Martian shells have a maximum effective range determined by the battleship's acceleration first, detection range second. The battleship can put out 0.1G of acceleration. It takes 1000s to out-deltaV a 1km/s missile. In those 1000s, the missiles would cross more than 25000km.

      At this distance, radar returns are too weak to be useful. LIDAR relies on the target being bright and reflective, but the Martian shells are cold fuzzy carbon nanotubes called Vantablack. So, we'll use this as maximal firing distance, even if it is quite likely that the Terran battleship will only start acceleration once it sees the Martian coilgun being fired. However, if you read the design notes above, the Martian can fire a full salvo without being detected...

      Radar in space will be very short wavelength. There is not atmosphere to absorb these wavelengths, so even millimeter radar can be used for long range scanning. You need them anyway to detect micrometeorites. So, the diameter of the shells will always be too big to be hidden. CCD IR sensors can re-adjust their sensitivity in milliseconds today, so dazzling them in this way is unlikely to work. If you have several sensors, you can tune each of them to different brightness levels. The objects are at cryogenic temperatures anyways (20K) so they're very hard to spot with long range passive IR.

    2. What you have left is three successive scenarios.

      The first is the stealth shot. The Terran battleship does not see the Martian ship fire. It only detects the missiles once they are very close, possibly under 100km (4 seconds) away. The second is the tracking shot. The Terran warship sees the Martian ship fire and in which direction. The missiles are tracked by shining a 750MW laser on them until a hotspot is created, which allows them to be tracked by passive IR. Anti-missile defenses include reactive armor plates from 1km, casaba howitzer PD from 100km, anti-missile laser point defenses from 1000km and main laser up to 25000km.

      Firing all these lasers and taking down the missiles one by one heats up the equipment. A hot laser is a less accurate laser. A less accurate laser delivers less energy to the target, so you have to fire it for even longer and heat it up even more.

      This leads to the third scenario. The terran warship switches off its laser generator to cool it down and relies on kinetic and atomic point defenses during this period. They are incredibly effective against missiles but also very limited in ammo.

      So what is the likely result?

      The first scenario actually determines how all the rest goes. With no time to deploy a casaba howitzer far away enough from the hull or for the lasers to get a target lock, the Terran battleship can only rely on its armor. Each 30kg warhead that goes through delivers a devastating 9GJ impact. That's two tons of TNT focused on a point 10cm across. If this hits the propellant tanks, the engine, the lasers... the ship can only handle such an impact with its head-on armor. It has about 591mm armor depth combined with any equipment in the way. 400mm penetrators won't go all the way through. It might scramble the wire radiators, crack the laser lens or cause a radiation leak in the nuclear engines. Repairing this damage gives the second missile wave less chance of being intercepted.

      But, if you think all these interactions between one ship attacking another are complicated, imagine the standard 3vs3 scenario. Two Terran battleships can handle laser defenses while the third uses its laser offensively, to cut into the cold shell of the Martian Arsenal ships and prevent them from returning to stealth and drifting away. Or, the Martians Arsenal ships could be defended by a Martian Interceptor that put out 0.4 to 1.2G acceleration to catch up with the Terran battleships and zap them up the rear with laser blasts. Or, the Terran battleships can do a full deceleration burn and start the battle at zero relative velocity. Neither side's missiles go through laser defenses, making it a pure laser vs laser fight... but the Interceptor has long-range refocusing drones that can bounce a laser between them to out-range the Terrans, unless they are all swept away by Silver Bullet casaba howitzer strike.

      I hope that all these interactions and tactics inspire you to seek a more complex combat scenario and find the fun in it.

    3. Thanks for the long reply. I wanted to see what the effects of a third weapon type in the engagement would be, partially to see if I can reasonably use 3 weapon types (laser/missile/kinetic) in my project and also to get a better feel for how space combat works. All I did was swap the Martian laser for a 500MW coilgun to make sure that the tech was equivalent level.

      The reason why I chose a closing speed of 6km/s (roughly orbital velocities) was to choose a non-ideal scenario for kinetics, a 1 v 1 battle was just to make my spreadsheet a little simpler!

      With regard to deltav – you are correct: I messed up the calcs there – the projectiles would require 1.6km/s dv at the engagement range of 10000km.

      Cold vantablack shells: I was assuming that while this is worth the expense on a stealth missile, our coilgun rounds are cheap and as simple as possible so I did not include those features. I assume that firing the round would heat it up anyway? I am assuming a 2.7mm wavelength radar – surely any shorter than that and it would be absorbed by the black paint? I understand a flare will emit a range of wavelengths so a big enough flare should dazzle the sensors even if the peak wavelength is not right – the emissions given off a cool projectile are very small.

      The principle of this kinetic weapon is that you can detect the projectiles quite easily, but locating them with enough accuracy to shoot them down is impossible until they get less than 100m away. Modern CIWS like phalanx is expected to kill missiles at only a few hundred meters but missiles are much bigger and are moving 10x slower. The only chance a defending ship has is to evade the projectiles but since they don’t have turrets they can’t evade and shoot at the same time.

      On a fast approach, the Terran ship is either disabled or has deviated a long way from its approach vector and is moving 1.6km/s in the wrong direction. Their best bet would therefore be to approach cautiously, but then they have to burn a lot of dv to slow down from the transfer…

      This discussion is very interesting! My project will be announced soon but is a ‘tough sf’ 4x computer game. I would be interested in your thoughts when that happens!

    4. I think reading the Electric cannons posts ( will give you a handle on the limitations and tactics of rail/coilguns.

      I'd like to point out that even the fastest coilguns ( need to fire miniature missiles to reach their targets. So all space warfare weapons are either direct energy or guided projectiles.

      A fairer comparison between a coilgun and a laser is to provide the same reactor power instead of weapon power. This is because you'd miss out on the fantastic efficiency of coilguns vs lasers, sometimes reaching 90%+ for slow coilguns.

      In terms of closing speeds, I see three general situations: chasing orbits, interception and interplanetary flyby.

      If you are on the same orbit, relative velocity is essentially nil. This is a huge advantage to fast coilguns, as they can deliver a projectile much faster than a missile that has to accelerate up to speed while being fired upon. Lasers should be devastating too.

      If you are in an interception scenario, then one fleet is sitting mostly still while the other slows down from interplanetary speeds. Slow projectiles are more useful, as you can use the velocity of the attacker against them.

      In an interplanetary flyby, closing velocities are huge. Slow coilguns are used before the intercept, missiles after the intercept. Lasers might have a hard time tracking targets which zip across the 'horizon'.

      Cold shells: if you only need to hide a projectile for a short period of time, then vantablack+cold hydrogen dumping is quite a simple option. The only expensive part is the thin cover of carbon nanotubes, but they aren't strictly necessary unless the target already knows where to shine their LIDAR. The round can be fired in a sabot. The sabot gets heated up while the payload remains cold.

      With regards to radar and vantablack, there are conflicting reports. Vantablack can be tuned to absorb a wide range of wavelengths, but it is however difficult to make a thin material interact with long wavelengths such as radar. Like an antenna.

      The problem with flares again is that even if you manage to cover an entire spectrum of frequencies, the LIDAR can be used like an AESA radar: the outgoing emissions can be shifted, coded and encrypted every millisecond. So you pop the flares, you get a laser beam covering the area. Instead of a single wavelength, you get a range of wavelengths (from a free electron laser) or coded flashes of light. The return signal must match the changes in wavelengths or the correct codes if it bounced off a target. Any other static return can be disregarded...

      Point defense targeting: I don't think you should stick to arbitrary ranges such as '100m' without giving a good reason why. Modern CIWS in a thick atmosphere trying to target a missile through sea haze is very different from futuristic laser point defense that operates in a 2.7K environment and can shoot whatever it sees with LIDAR immediately. Laser PD means you can have multiple small mirror turrets for defense even when your main mirror is too large to be turreted.

      As for missiles being bigger, modern ASMs can be several meters in length while even centimeter-sized shrapnel in space can be deadly at high closing velocities.

      With regards to deviation, 1.6km/s might be too small to exhaust a combat ship's deltaV reserves. Even the small and powerful Interceptor has 5km/s internal deltaV reserve. If a combat ship is pushed around for long enough, it might become a sitting duck... but if you leave alone long enough or if it flies past an intercept untouched, it can always be refuelled by a tender or docks with its interplanetary stage for a second pass!

      Good luck with your 4x game! Most of these discussions focus on the tactical aspect, so having them influence wars on a strategic scale is going to be a tough task.

    5. I am replying after a very long time, but I have just begun to reveal my game and if you are interested, please visit my blog at for more information.

      Development is slow but steady so expect a blog post every month describing the game in more detail, punctuated with status updates of internal releases. I would welcome feedback and advice, although I know you are busy with your own blog!

    6. Thanks for sharing this. I'll have a look.

  11. That rendered image of the ship is really nice, and I can see it's by someone calling themselves "grokodaemon". I'd like to track down more of his work to have a look at, there aren't all that many people doing art of highly realistic sci-fi spaceships so being able to find all the work of any artists who does this sort of stuff is really nice. But I followed the link to grokodaemon's reddit page, and it doesn't seem to link to anything like a blog of his. The imgur link doesn't give me links to his imgur user account either. So I'm a little stuck in being able to find more of his stuff, and there must be more of it because a lot of sill clearly went into the pictures on this page so grokodaemon must have learn his craft over many pieces. Can anyone point me to a blog of his, a devinatart page, anything which concisely has the rest of his image artwork on it. Thanks.

    1. I also could not find any of grokodaemon's other work. Try contacting him directly on reddit... but as he is not a frequent user, so don't hold your breath either.

      If you want more realistic scifi art, projectrho has a very good collection here:

      As for hard SF in general... we don't have many artists. Our best contributors are usually illustrators for NASA research projects, or the handful of 'military SF' artists who find it interesting to add a set of radiators to their warships. That's simply the state of things!