# Can I overcome the bandwidth limitation on my photonic railways?

The Humern Empire is a vast, well established empire that spans the galaxy. Despite having no FTL travel they have kept their empire together with a series of subluminal transport methods.

Heavy cargo loads are moved from system to system using a finely tuned web of Photonic Laser Propulsion base stations on various airless moons and planetoids, cargo ships being constantly accelerated across the void by lasers bounced off one base station before being decelerated at the other end by a receiving one. These 'photonic railways' are marvels of high precision engineering, and across the galaxy they are carefully balanced so no one base station is ever pushed too far out of position.

They do have a significant limitation though. Despite the Empire's best efforts base stations can't be established unless they're on a large planetoid with a regular orbit (which are in limited supply). This means each star can only establish a railway to the nearest few stars.

Not only that, but although each base station can have truly ridiculous amounts of power pumped through it (enough to make constant 1g acceleration in the interstellar void possible) any one railway can only really have a few ships being accelerated along the railway at any one time before the ships themselves start having to perform significant course corrections, which is less than ideal.

The Empire wants to maintain a constant stream of smaller transports rather than just relying on a few big shipments, both to keep the colonies culturally attached to the Empire and also to increase redundancy.

How can this be achieved? How can the Empire ensure it's cargo shipping capability scales with the amount of energy it has available instead of the number of railways established?

• How do you stop the railways energy output from destroying the vehicles in transit given many years of being pummelled by energy powerful enough to accelerate at 1G between stars? Given that a journey across the empire will literally take millions of years with major time dilation, there's a lot of potential plot devices that have been left unexploited – Tom J Nowell Oct 26 '17 at 22:37
• @TomJNowell, that's not within the scope of the question. The OP isn't asking us to "reality check" his transport method, but to help him understand ways it can be bolstered. – JBH Oct 26 '17 at 23:09
• Why is it that you can only have a few ships per railway at a time? (I can think of a few possible reasons but resolving that issue would require knowing which one is in effect here.) Also be aware that if you're writing in our universe it's a bit strange to have large planetoids with regular orbits in short supply, since given the number of exoplanets that have been observed that seems not to be the case. (There are now thought to be planets around almost every star, many of them large and rocky.) – Nathaniel Oct 27 '17 at 5:33
• @Nathaniel: if you consider star systems as points (not unreasonable given the distances involved) then the ‘line’ between them is 1d, supporting precisely one ship moving back and forth. Now, the number of planetoids is (as you say) large, but they move, so let’s assume we have ten base stations that are in the right orbit to fire lasers at any one ship at any given time. That’s ten ships on one ‘line’. Even if the beams are thick only so many ships can ride one beam simultaneously, and even if there are hundreds of suitable planetoids the network is still limited, and the Empire is Huge. – Joe Bloggs Oct 27 '17 at 8:03
• @Nathaniel: though you have just given me a thought for a parallax based answer, though it would be complex as anything. – Joe Bloggs Oct 27 '17 at 8:05

# Use a chain of ships.

I'm imagining a pyramid setup, in a sense. Take $n$ ships. Send out the first one, of relatively low mass - Ship $1$. Then send out Ship $2$. Ship $2$ is now directly propelled by the laser on the base, and it in turn is propelled Ship $1$. Repeat by launching Ship $3$ to power Ship $2$, and so on until you launch Ship $n$.

The thing here is that you end up using more power than you would if you simply launched each ship one-by-one, powered by the laser. If you gave each ship the same power, then only the very first ship would travel onwards, propelled like normal. However, you've solved the problem. You're only limited by energy, not by the number of railways.

Things to watch out for:

• You're not going to be able to go to many stars at once. It's possible that the station on each ship could rotate, so you could change its orientation to propel the ship ahead on a different path, but it would take work.
• Launching each new ship would be a delicate process. You have to line things up just right. If a ship slips outside the line of vehicles, you could be in trouble. Also, as not store bought dirt pointed out, the timing could be just as difficult as the spatial alignment. If the power was changed at the wrong time, the resulting disturbance would propagate through the system.
• As I said before, you're using a lot of power. But in an empire this large, I don't think that's going to be a huge problem. How you store the energy it is something that needs to be addressed; more energy means more mass on each ship, which is obviously not a good thing. It's plausible that solar energy could provide some help at the beginning, but it's not as useful in interstellar space.

Here's a diagram of what I'm picturing, for a simple set of three ships:

• You would need to put a Xg source on each ship for a chain of X ships, and the transitions as you added another ship to the chain or took one off the end would be headaches. – user25818 Oct 26 '17 at 19:33
• @notstoreboughtdirt That's true (and I think I mentioned it in the list), but with an empire like this, it should be doable. Difficult, but doable. – HDE 226868 Oct 26 '17 at 19:35
• You mention direction, but I mean timing. You need to be right on time or someone gets an extra g either forward or backward. – user25818 Oct 26 '17 at 19:44
• @notstoreboughtdirt That's quite a good point. Answer edited. – HDE 226868 Oct 26 '17 at 19:47
• @Shane: I’m using a variant of this answer where each ship is capable of ‘passing through’ various lasers with insanely precise mirror arrays. It even preserves the requirement to not waste photons rather than just turning each ship into a basic laser sail. – Joe Bloggs Oct 29 '17 at 10:28

Prioritize your shipping and stagger delivery times

Not everything needs to make port ASAP (though it does help commerce). Prioritize and consolidate cargo ships carrying cargo with similar expiration times. Use your laser such that ships acceleration times are proportional to their to prioritization.

The math for this becomes incredibly more complicated but possible. For instance your passenger ship would take the direct most route receiving more acceleration bursts such that its flight time is 1 period. A ship carrying recreational equipment could be launched on a different trajectory and accelerated with less attention such that its total flight time could be 50 periods or more. This way you eventually result in a fairly constant stream of ships.

This image sort of describes what im trying to get at, with different accelerations you can achieve different flight paths and different flight times while still hitting the same target.

Note: The added advantage of this method is that the rotation of your planets can now be used more efficiently. When the laser is no longer able to accelerate the priority ship (due to its angle with the horizon) it can now accelerate other ships as their position in the sky would be vastly different than the priority ship.

Also note: I would make cargo ships autonomous as there really isn't any need for crew especially since the rails are essentially controlling the majority of navigation.

• Interesting point about different trajectories potentially ending up at the same target. It would require a very detailed gravimetric (Is that even a word?) map of the galaxy, but in theory it's possible.. – Joe Bloggs Oct 26 '17 at 19:14
• You would need to know the exact orbit of planets so you could know their position at any given time. You would need to know fairly accurately what their mass is (to no their gravitational force). Beyond that you would need to know mass of the ship. After that its all doable, just going to need ALOT of computer time when creating and fixing flight plans. But its fairly simple physics and geometry, just a huge amount of calculations. – anon Oct 26 '17 at 19:22
• +1 because it made me think of gravitational lensing, which I would not have thought of if I didn't meditate on "different trajectories same target", the issue being that most laser propulsion systems require the laser and the receiver to be perpendicular. See my answer part #5. – Quaternion Oct 27 '17 at 2:26
• I'm sorry, "flight time of 1 week"? Interstellar craft with human-civilization scale caring about mere orbital velocity? This answer betrays a complete misunderstanding of the scale of physics involved. It is nonsense on the scale of interstellar travel. – Yakk Oct 27 '17 at 14:37
• @Yakk as I mentioned to Tom the flight time used was merely to illustrate the different flight times caused by the prioritization of different ships. Obviously, without knowing the interstellar distances and acceleration potentials there is no way to accurately state flight times. But since people seem to leap on to this irrelevant detail, I have changed my unit of time (weeks. years) to period. – anon Oct 27 '17 at 14:45

Make more transmission stations

If you can afford to push a ship at 1g for years, you can probably push asteroids or moons to eventually form into a body in the orbits you want.

It may take centuries of planning to maneuver the heavy objects out of gravity wells, collide them non-elastically without creating much debris, weld them together perhaps with ice and correct their orbits, but your slower-than-light galaxy-spanning people can't be strangers to patience.

Launch stations

You need it to leave solar orbit, but you don't need it to go any faster than that. Over time it will be somewhere far from the star so can push from a different angle.

If you can apply a counter push light-years away you can assist in transfers going past. This becomes a planning nightmare, since you are pretty much using death rays aimed years ahead of vessels with years lag in coordination between forces, and you only get a fraction (cosine of the angle) of the power used as velocity in the right direction, but maybe it could be a net benefit.

Make unpowered trips

From either planet's point of view keeping the ship at 1g longer than a few weeks doesn't make the trip take noticeably shorter (though from the cerw's point of view it would). If you only push for a short time you free to push the next ship.

And this method allows you to launch to any visible destination, since only the very start and end need to be powered.

Interstellar laser propulsion should never be based on a planet, the free space environment is far better in terms of collecting solar energy and building large structures for radiating away heat and focusing the laser beam. Dr Robert L Forward described such a system many years ago, and it already seems to have a far superior performance than what you are asking for. Since in the Forward system, the ships "coast" after being accelerated, and then receive laser power during the deceleration phase of the flight (once the ring mirror is detached), the laser is available to power other ships. With careful timing, the laser can be fully employed first accelerating trains of ships, then decelerating them years later.

Robert L Forward's laser lightsail system

Examples of the lightsails, to scale...

This isn't even the end of this idea. A reader of the NextBigFuture blog came up with a rather amazing idea to explain the strange dimming of "Tabby's Star"; a light "railroad" generated by a mirror orbiting the star:

Just imagine that, despite their size, these mirrors can be shaped to optical quality. What is the diffraction limit on what a telescope with a primary mirror more than half the diameter of our sun (I estimate D~ 8e8m to cause of 22% dip) can resolve in visible light (~5e-7m)? Somewhere on the order of 1e-16 radians. Since a parsec is 3e16 meters that means they could resolve objects on the order of 10’s of kilometers on Earth when viewed from 454 pc. So, great for astronomy.

When applied as the driving beam for a starship, you end up with this:

The light pressure in the ‘beam’ from the initial mirror would supply twice that pressure to this lightsail upon being reflected, sufficient to provide the sail a constant acceleration of several times g=10m/s^2. By hanging a starship of similar mass on the sail, the acceleration can be made equal to that of the passenger’s home planet, and all the inhabitants of the planet could indeed probably be accommodated in a ship of such mass simultaneously. At 1 g constant acceleration they could journey ‘anywhere’ experiencing only a couple of decades of onboard time due to time dilation (or less than one decade at 2 g’s). However, unless the home star is about to go up in flames and an ark is needed for evacuation, this ‘beam-filling’ sail is overkill.

Since you are not planning to send billions of beings at once on an interstellar ark the size of Mars, we can scale down to this:

We can assume the big beam is big just so it won’t spread and thinout due to diffraction. Smaller starships and with sails proportioned in scale with their small mass can travel within the bigger beam with the same level of acceleration. Energy not hitting the sail is not ‘wasted’ since stars always pour out energy, the megastructure aliens have just redirected it, their only expense is in building the original structures and maintaining focus and aim. The ships themselves are simple affairs. No need for huge stores of fuel, for Bussard ramjet complications, for antimatter, for generation ships. You just jump in a beam and the rest is free. (But you must trust the operators to remember you if you go off on a 1000 lyr expedition or something). It would have been the obvious way to go from the first, at least in principle, if we could just think ‘big enough’.

Now we don't just have a singular mirror near the star, a great deal of supporting structure is also required:

The primary ‘deflect and direct’ mirrors need to be positioned‘near’ the star’s surface and kept stationary relative to the sky (except those scanning it for astronomical purposes). Therefore they must be ‘floated’ on the star’s own light pressure in order to counter the star’s gravity without orbital motion. At a distance of 3 to 4 stellar radii (I am assuming the star is 1.25* solar radius and 1.5* solar mass) the gravitational acceleration would be in the range of a few g’s and they could be ballasted to float with structural forces similar to those involved in the active starship and sail situation.

The mirrors are probably stabilized on an open work spherical network of rings around the star, having enough mirrors to support and expand the whole structure equally under some degree of tension. Schaefer may have been documenting the addition of more mirrors to support and balance the whole as it was initially built up, or the accumulation of mirrors on the side of the star near us as commerce was increasing with other star systems in that general direction (or perhaps the addition of inactive mirrors just to combat global warming as their star ages).

So by scaling your thinking a bit, you can increase the throughput of the "railway" serially (launching and accelerating/decelerating ships when prior ships are in coast phase), or in parallel (supermassive mirrors creating such a huge beam that you can launch a multitude of ships at once).

In fact, if a rival Empire starts thinking like this, you might discover their economic potential literally rocketing past yours......

• I like the idea of lensing the star to provide a beam! The photonic railway (note, that’s the designers name for it, not mine) is essentially a more efficient version of the laser sail with the reflected photons being bounced back onto the initial station and passed through a gain medium. The reason I wanted to put them on planetoids is to reduce the number of self sustaining, station keeping, structurally sound >1000km diameter space stations that need building, an also to provide some extra mass to ‘buffer’ the system. – Joe Bloggs Oct 27 '17 at 12:46

Shipping containers and transshipment

This doesn't seem to be the answer you want, but it really should be given anyway, since it is the way an analogous issue was solved here on Earth. So not only do we have good reason to think it would work, you can easily study how it works in practice and adapt to your setting.

Basically, just use the few rails with larger and larger ships containing standardized containers. Your planetoid sized base stations will then transfer the container to a ship going the next step.

This will do nothing to solve issues with latency to smaller ports, but it will solve the bandwidth scaling issue. Which is what you asked about. It will also reduce the shipping costs and increase economic integration, which seems to be something you'd want?

A few ideas:

1) If 1G acceleration can be maintained, then the speed of light will be reached in 1 year: reference. As such, constant 1g acceleration is not required over the whole distance. Greater accelerations of course result in bursts of energy (I'm calling 1 year a "burst"), rather than a non-stop stream. Keeping a small stream up "lighting up the lane" so the ship knows it is on target might not be bad.

2) Cryogenics and Intersteller-Packing-Foam (which fills the ship and makes it and everything frozen within it much more rigid) could allow for higher accelerations. Higher accelerations, combined with #1 (there is a limited thrust time before reaching C) results in increased transmission times. It also allows for rescue should people/cargo need to wait the ridiculous time frames which would be required for a rescue.

3) Lanes: More shipping lanes, more bandwidth. Each lane is a launch platform, and for various reasons each should be on its own airless planetoid within a solar system. While a solar system may seem big, regarding interstellar distances it is not, as all the lanes will converge on a new system.

4) Decouple Interstellar and interplanetary transportation systems: As you have already pointed out, your interstellar rail system uses planetoids without an atmosphere, and for safety reasons probably minimal life. By having your interstellar ships being accelerated outside of any significant gravity a lot of issues can be avoided.

5) Use gravitational lensing : both to keep track of your ships, and to fire them along different paths which result in the same destination. I was skeptical after reading another answer which had to do with using different flight paths but same origins, my issue is that laser based propulsion is like turning light into a fire-hose and using that to push something. Actually it is worse because, most of these system use a feedback system where the light is recycled or the power requirements go way up(meaning you can't use a sail, you have to have a perpendicular surface). You want the laser to push directly along the flight path. If it is hit at an angle it will start to spin. It may be possible to compensate somewhat but it is still not ideal. So how can we send multiple ships out from the same point on different paths, and have them converge at the same location? Gravitational lensing. If we find the path light is bent by the systems central star(s), you can send ships along the same flight path and the beam which is directly behind them will always perfectly aligned, because the light beam is traveling straight, it is space that is bent. Also gravitational lensing can focus a great amount of EM radiation from distant objects, making it useful for picking up the weak communications from distant ships, or observing distant planets in other systems.

6) Course corrections & Keeping your speed up: While theoretically not required, as you can send any rail-car to any location, after spending 100s of years of outside observers time maybe the receiver wants the cargo to go to another location, or the planet may have had a malfunction regarding its laser required for deceleration. It might be best to direct all traffic to the star for the system. As the train begins to make final approach (the distance to decelerate from near light speed would be 1 year if 1G is assumed) it could use the gravity well of the star to change its trajectory. Also a star is an excellent power source, if you are looking for something which must generate tremendous energy output. Also #2 (cryogenics&ridged packing-foam) I expect are quite essential if using a gravitational well for course correction. Even if people are to stay awake for most of the trip, #2 might be required during "lane-changes" or when transition to a different "leg". Where some aggressive deceleration may be expected, along with being accelerated along a new direction. If you can maintain much of your speed when transitioning to a new target, you also keep throughput up.

Just for fun: "Batteries Included" while some seem to hint that generating energies similar to what could destroy the surface of a planet every hour as not being a good idea... well here is a power proposal that makes even that look ridiculously safe! It is probably worth investigating how a large transportation hub could work. In my view it would be most ideal if the transportation hubs were situated around black holes, most ideally if the black hole was not too large (smaller the better to exploit tidal forces) and to avoid having to feed it, if it is in binary orbit around a star that would be really helpful. Such a setup would produce a vast amount of power. It is theoretically possible to produce sub-atomic black holes. Particle accelerators such as the LHC are still a couple magnitudes off what is required. Also such black holes evaporate! But still it is possible given enough energy to produce black holes, which are built in an accelerator. Now since they are produced at a high speed they are subject to time dilatation, while at rest they may only last seconds (or less) but at speeds approaching C, they can last much longer. Point of this is that a powerful accelerator which has access to massive amounts of energy (something pouring off gamma-rays seems like a good start) can store tremendous amounts of energy in black holes and ship them to planetoids in the system. Since black holes have magnetic fields see here, they can be bent and accelerated/decelerated with magnetic fields. So there is a means of capturing them and drawing off the energy (hawking radiation). As a small black hole decelerates it will produce more energy, and if accelerated less... so there is a means of controlling the output. Once the black hole starts to approach the limit where it will cease to be a black hole it should be fired back into the parent black hole... for people that thought your lasers were dangerous, having a 10000 kg black hole come apart when powering one your airless planetoid lasers would probably result in there being no planetoid!

• You post a lot of points.... Peoples attention is generally limited, when making answers you want to be concise and highlight main points allowing the reader to judge if they want to read your justification. Anyways your 1st method doesn't account for friction, which at subliminal speeds all the space dust would surely offer some. You should link gravitational lensing, but also if I remember correctly, that requires incredibly massive objects like black holes and maybe stars. Planets don't offer much in the way of lensing. – anon Oct 27 '17 at 13:38
• @anon regarding gravitational lensing: Using our own sun, the foci is 550 AU (astronomical units [the distance from the earth to the sun]). Pluto is 50 AU for reference. However the nearest star is 268,770 AU (about 0.002% of the total distance), so despite what at first seems like a backward way of doing things really amounts to almost no overhead. In terms of construction it would clearly be helpful if the star was significantly larger than ours or a black hole... as the foci would be a lot closer. – Quaternion Oct 27 '17 at 21:00
• @anon I'll take your points regarding attention into consideration. I'll think about the dust as well... I don't feel is is an issue regarding friction however even small grains of sand may cause serious issue. After all most shooting stars are nearly stationary grains sand that the earth runs into at 460 m/s. Running into a similar grain, or god forbid something even slightly larger at 0.5 C (149896229 m/s)can't possibly be good. – Quaternion Oct 27 '17 at 22:08

In keeping with the railway analogy....

Double Track: Build a second, parallel guide-way near the first. The Empire may already have two guide-ways for two-way travel, in which case this would mean four guide-ways (two in each direction). This will double the one-way frequency of pods.

Speed Up The Trains: Instead of accelerating at 1g, accelerate at 1.5g or 2g or more. On a 6 light-year trip to Barnard's Star, 2g will decrease travel time by about 30%, and increase pod frequency similarly.

Extend the rails in parallel.

Ship 1, top right, is accelerated by a blue beam from the laser. It reflects back in parallel, redshifted from its original color (because the mirror is receding).

Ship 2, middle left, is going the opposite direction. It is accelerated by the reduced frequency but still pretty powerful laser reflected from Ship 1. It reflects an even deeper redshifted beam back in the original direction to accelerate ship 3.

Depending on the efficiency of the ship mirrors, this could be carried on for several more levels if necessary. Furthermore, when a ship is decelerating it actually blueshifts the beam for the next recipient.

(Apologies for my terrible color choices.I'm color blind.)